Computer automated system and method for converting source documents bearing symbols and alphanumeric text relating to three dimensional objects

Abstract

The computer automated system and method of converting a digitized raster image of a scanned source document, bearing alphanumeric text relating to a plurality of physical dimensions and to a plurality of edges of a three dimensional object and of a moiety a symbol represents and of an insertion point of the moiety into the three dimensional object, at least one orthographic drawing view having a plurality of lines oriented in a direction to each other and corresponding to the edges of the three dimensional object, and the symbol, into mathematically accurate three dimensional vectors corresponding to the physical dimensions and the edges of the object and moiety and into a mathematically accurate computer drawing file. The digitized raster image is organized into an orthographic viewpoint file corresponding to the view. The file is imported into a corresponding orthographic viewport in a CAD drawing file having three dimensional vector generating capability in an existing CAD system having a COGO subroutine and using an OCR and an OSR operating within the CAD system. The alphanumeric text relating to the symbol is recognized by the OCR and an attributed symbol vector file is created using CAD block attribution techniques. The alphanumeric text relating to the plurality of physical dimensions and to the plurality of edges of the three dimensional object and the insertion point is recognized by the OCR; the symbol is recognized by the OSR. The recognized alphanumeric text and recognized symbol and the attributed symbol vector file are converted by the COGO subroutine into mathematically accurate vectors which can be used for producing accurate drawings and for Computer Assisted Manufacturing. Mechanical, engineering and architectural drawing (plans) are converted by the present invention.

Claims

What is claimed is:

1. A computer method of converting alphanumeric text relating to a plurality of physical dimensions and to a plurality of edges of a three dimensional object, from a hard copy source document having recorded thereon the alphanumeric text and a plurality of drawing views of the three dimensional object, into mathematically accurate vectors corresponding to the physical dimensions and the edges of the three dimensional object, the method comprising the steps of:

(a) acquiring a computer useable raster image of a hard copy source document having

(i) a plurality of drawing views thereon of a three dimensional object having a plurality of physical dimensions and a plurality of edges, said drawing views having a plurality of lines corresponding to said edges of said three dimensional object, said lines oriented in a direction relative to each other; and

(ii) alphanumeric text relating to said plurality of physical dimensions and to said plurality of edges of said three dimensional object and recorded on said drawing view in association with said lines on said drawing views; and

organizing said raster image according to an orthographic viewpoint raster file for each said drawing view by selecting one said drawing view and said alphanumeric text associated with said lines on said drawing view;

(b) setting up a drawing file in a CAD applications program in a computer; said CAD applications program having a plurality of orthographic viewports, a coordinate geometry subroutine for creating three dimensional vectors, and an optical character recognition subroutine; and selecting one of said orthographic viewports to correspond in orthogonality to one of said orthographic viewpoint raster files;

(c) importing one said orthographic viewpoint raster file into one said orthographic viewport which corresponds in orthogonality to one said drawing view;

(d) repeating step (c) for each drawing view on said document;

(e) recognizing said alphanumeric text in said optical character recognition subroutine in each said orthographic viewport separately, and creating a recognized alphanumeric text;

(f) converting said recognized alphanumeric text in said coordinate geometry subroutine into said mathematically accurate three dimensional vectors corresponding to said alphanumeric text recorded on said document.

2. The computer method of claim 1, wherein said document for use in step (a) is a mechanical drawing.

3. The computer method of claim 1, wherein said document for use in step (a) is an engineering drawing.

4. The computer method of claim 1, wherein said document for use in step (a) is an architectural plan having a plurality of drawing sheets.

5. The computer method of claim 1, wherein step (a) includes acquiring said raster image of said drawing view of said document and said alphanumeric text by scanning said document in an automated digitizing unit and receiving said raster image from said unit into said CAD applications program in said computer,

wherein step (b) of organizing each said viewport includes setting up said drawing file in each said viewport corresponding to each said drawing view on said document;

wherein step (f) includes

(i) selecting an origin and a start point on a viewpoint raster image of one said drawing view in one of said viewports for the generation of a first vector, corresponding to one said edge;

(ii) selecting said recognized alphanumeric text corresponding to said one edge and transporting said recognized text into said coordinate geometry subroutine;

(iii) creating said first vector corresponding to said one edge, said first vector having an endpoint coordinate;

(iv) selecting an adjacent edge on said raster image of said drawing view in said viewport, selecting said recognized alphanumeric text corresponding to said selected adjacent edge, transporting said recognized text into said coordinate geometry subroutine and creating a next vector corresponding to said adjacent edge, using an endpoint of a last vector generated as a beginning point of a next vector;

(v) repeating step (iv) until all adjacent edges have been selected and corresponding vectors created;

(vi) creating next said vectors for each said orthographic viewpoint corresponding to said edges on each said drawing view on said document, until all said adjacent edges have been selected; and

(vii) generating an orthographic vector file of said mathematically accurate vectors corresponding to said alphanumeric text relating to the physical dimensions and said edges of said three dimensional object.

6. The method of claim 5 wherein step (f)(vi) includes repeating steps (f)(i) through (f)(v) for each said orthographic viewport corresponding to each said drawing views on said document.

7. A computer method of converting alphanumeric text relating to a plurality of physical dimensions and a plurality of edges of a three dimensional object, relating to a plurality of physical dimensions and a plurality of edges of a moiety represented by a symbol and its symbol property and used in association with the three dimensional object, and relating to an insertion point of the moiety which the symbol represents within the three dimensional object, from a hard copy source document having recorded thereon a plurality of drawing views of the three dimensional object, the alphanumeric text, and the symbol, into mathematically accurate vectors corresponding to the physical dimensions and the edges of the three dimensional object and the moiety the symbol represents; the method comprising the steps of:

(a) acquiring a computer useable raster image of a hard copy source document having

(i) a plurality of drawing views thereon of a three dimensional object having a plurality of physical dimensions and a plurality of edges, said drawing view having a plurality of lines corresponding to said edges of said three dimensional object, said lines oriented in a direction to each other;

(ii) at least one symbol disposed on at least one said drawing view, said symbol having a symbol property, said symbol associated with a moiety in the three dimensional object; said moiety having a plurality of physical dimensions and a plurality of edges expressed as said symbol property, said symbol having an insertion point on said drawing view corresponding to a moiety insertion point of the moiety in the three dimensional object; and

(iii) a first alphanumeric text relating to said plurality of physical dimensions and to said plurality of edges of said three dimensional object and recorded on said drawing view in association with said lines on said drawing views; a second alphanumeric text relating to said symbol property for each said symbol; and a third alphanumeric text relating to said insertion point; and

organizing said raster image according to an orthographic viewpoint raster file for each said drawing view by selecting one said drawing view and said first alphanumeric text associated with said lines on said drawing view, said symbol and said third alphanumeric text; and organizing a symbol property viewpoint raster file for said second alphanumeric text;

(b) setting up a drawing file in a CAD applications program in a computer, said drawing file having a symbol library; said CAD applications program having a coordinate geometry subroutine for creating three dimensional vectors, an optical symbol recognition subroutine, an optical character recognition subroutine and a plurality of viewports; and selecting said viewports into at least one orthographic viewport and customizing at least one floating viewport;

(c) importing said symbol property viewpoint raster file into one said floating viewport;

(d) repeating step (c) for each said second alphanumeric text for each said symbol on said document;

(e) importing one said orthographic viewpoint raster file of one said drawing views into one said orthographic viewport which corresponds in orthogonality to both said drawing view and said orthographic viewpoint raster file;

(f) repeating step (e) for each drawing view on said document;

(g) recognizing in said optical character recognition subroutine said second alphanumeric text in each said floating viewport then recognizing said first and third alphanumeric texts in each said orthographic viewport; and creating a first, a second and a third recognized alphanumeric text;

(h) creating an attributed symbol by converting said second recognized alphanumeric text relating to said symbol property of said symbol into a block of mathematically accurate three dimensional symbol vectors representing said physical dimensions and said edges of the moiety said symbol represents and creating a vector symbol file of said symbol vectors;

(i) recognizing each said symbol in said optical symbol recognition subroutine in each said orthographic viewport separately;

(j) converting in each said orthographic viewpoint said first recognized alphanumeric text in said coordinate geometry subroutine into a plurality of mathematically accurate vectors corresponding to said first alphanumeric text recorded on said drawing and to the physical dimensions and edges of the three dimensional object; and placing said vectors into an orthographic vector file;

(k) selecting said attributed symbol in said orthographic viewport and converting said third recognized alphanumeric text inserting said symbol vector file into said orthographic vector file at said insertion point; and creating a vector file and a drawing file corresponding to the physical dimensions and to the edges of the three dimensional object with the moiety inserted therein.

8. The computer method of claim 7, further comprising the step of:

(l) displaying a drawing view in one said viewport generated by said orthographic vector file of step (j).

9. The computer method of claim 8, further comprising the step of:

(m) displaying a drawing view in one said orthographic viewport of said drawing file of step (k).

10. The method of claim 7, wherein said drawing file of said CAD applications program of step (b) includes a symbol property library and wherein step (h) includes calling up said symbol property library and adding said attributed symbol to said symbol library.

11. The computer method of claim 7, wherein said document for use in step (a) is a mechanical drawing.

12. The computer method of claim 7, wherein said document for use in step (a) is an architectural plan having a plurality of drawing sheets.

13. The computer method of claim 7, wherein said document for use in step (a) is an engineering drawing.

14. The computer method of claim 7,

wherein step (a) includes acquiring said raster image of said drawing view of said document and said alphanumeric text by scanning said document in an automated digitizing unit and receiving said raster image from said unit into said CAD applications program in said computer,

wherein step (b) includes

(i) setting up a drawing file in each said orthographic viewport corresponding to each said drawing view on said document, and

(ii) setting up a drawing file in said floating viewport corresponding to said second alphanumeric text;

wherein step (j) includes

(i) selecting an origin and a start point on said raster viewpoint image of said drawing view in one of said orthographic viewports for the generation of a first vector, corresponding to one said edge;

(ii) selecting said first recognized alphanumeric text and transporting said first recognized alphanumeric text into said coordinate geometry subroutine;

(iii) creating said first vector corresponding to said edge, said first vector having an endpoint coordinate;

(iv) selecting an adjacent edge on said raster viewpoint image of said view on said orthographic viewport, selecting said recognized alphanumeric text corresponding to said selected adjacent edge, transporting said recognized text into said CAD applications program, and creating a next vector corresponding to said adjacent edge, using an endpoint of a last vector generated as a beginning point of a next vector;

(v) repeating step (iv) until all adjacent edges have been selected and corresponding vectors created;

(vi) creating said next vectors for each said orthographic viewpoint corresponding to each said drawing view on said document until all said adjacent edges have been selected; and

(vii) generating said orthographic vector file of said mathematically accurate vectors corresponding to said alphanumeric text recorded on said drawing relating to said physical dimensions and said edges of said three dimensional object;

wherein in step (h) includes

(i) selecting one of said orthographic viewports containing said viewpoint raster image of said view containing the most of said symbols;

(ii) selecting one said symbol; selecting said second recognized alphanumeric text for said symbol and converting said second recognized alphanumeric text into said block of said symbol vector representing said three dimensional properties of said symbol;

(iii) transporting said block of symbol vectors into said vector symbol file corresponding to said one symbol; and

(iv) repeating step (h)(i) through step (h)(iii) for each said symbol.

15. The method of claim 14, wherein step (j)(vi) includes repeating steps (j)(i) through step (j)(v) for each said orthographic viewport corresponding to each said drawing view on said document.

16. The method of claim 7, further comprising the step of (n) using said CAD applications program to generating a plurality of orthographic drawings of said drawing file of step (k).

17. An automated conversion system for converting alphanumeric text, relating to a plurality of physical dimensions and a plurality of edges of a three dimensional object and of properties of a symbol used in association with said three dimensional object and of an insertion point of the symbol within the three dimensional object, recorded on a hard copy source document having at least one drawing view of said three dimensional object, said alphanumeric text and at least one said symbol thereon, into mathematically accurate three dimensional vectors corresponding to said physical dimensions and said edges of said three dimensional object and said properties of said symbol and said insertion point, said system comprising in combination

an automatic digitizing unit for document scanning and

a computer having a three dimensional CAD applications program, said computer including receiving software and recognition software, said program, including conversion software, transport software and vectorization software;

said automatic digitizing unit (i) scanning a hard copy document having at least one drawing view thereon, having at least one symbol thereon and having alphanumeric text relating to a plurality of physical dimensions and a plurality of edges of a three dimensional object and of properties of said symbol used in association with said three dimensional object, and of an insertion point of the symbol within the object (ii) creating a digitized raster file corresponding to said alphanumeric text, said drawing view and said symbol, and (iii) outputting said digitized raster image;

said receiving software operatively associated with said automated digitizing unit, for receiving said digitized raster image into said CAD applications unit and creating a digitized raster viewpoint file, transporting said viewpoint file into a CAD drawing file, having a viewport;

said recognition software for recognizing said digitized raster viewpoint file text in said viewport, and creating an ASCII text file and comprising an optical character recognition subroutine operating in said CAD applications program for recognizing said alphanumeric text, and comprising an optical symbol recognition subroutine operating in said CAD application program for recognizing said symbol;

said conversion software, operatively associated with said recognition software, for converting said ASCII text file into a converted file useable in said coordinate geometry subroutine;

said transport software, responsive to said converted file, for transporting said converted file into said coordinate geometry subroutine; and

said vectorization software, operatively associated with said transport means, for converting said converted file into mathematically accurate three dimensional vectors having X, Y, Z coordinates and representing said plurality of physical dimensions and said plurality of edges of said three dimensional object and of said properties of said symbol used in association with said three dimensional object, and comprising a coordinate geometry subroutine.

18. The system of claim 17, wherein said document is a mechanical drawing.

19. The system of claim 17, wherein said document is an engineering drawing.

20. The system of claim 17, wherein said document is an architectural plan having a plurality of drawing sheets.

21. The system of claim 17, wherein said vectorization software further including software for arranging sequentially vectors according to a sequence of adjacent edges of said three dimensional object wherein said arranged vectors form a graphical representation in plurality of orthographic views of said three dimensional object.

22. The system of claim 17, wherein said vectorization software further includes software for arranging said vectors for a graphical representation of said property associated with said symbol and wherein said property of said symbol is introduced into said graphical representation of said object.

23. A computer quality control method for a mechanical drawing, the method comprising the steps of:

(a) scanning a mechanical drawing hard copy source document using an automated digitizing unit, said document having at least one drawing thereon of a three dimensional object and alphanumeric text relating to a physical dimension and an edge of said three dimensional object having a plurality of physical dimensions and a plurality of edges, to yield a digitized output;

(b) receiving said digitized output, said output including said digitized alphanumeric text and a digitized drawing from an automated digitizing unit into a CAD applications program, said CAD applications program having a coordinate geometry subroutine;

(c) displaying said digitized drawing; recognizing said alphanumeric text using a conversion software operating in said CAD applications program, said conversion software having a graphics optical character recognition subroutine, and recognizing said alphanumeric text using said textual optical character recognition subroutine;

(e) transporting said alphanumeric text into said coordinated geometry subroutine;

(f) converting said alphanumeric text in said coordinate geometry subroutine into mathematically accurate vectors representing said physical dimensions and said edges of said three dimensional object;

(g) generating a set of orthographic views of said three dimensional object using said mathematically accurate vectors and overlaying each said orthographic view on a corresponding view of said digitized drawing.

24. The method of claim 23, wherein said mechanical drawing scanned in step (a) is an engineering drawing.

25. The method of claim 23, wherein said mechanical drawing scanned in step (a) is an architectural plan having a plurality of drawing sheets.

26. A computer method of converting a mechanical drawing having alphanumeric text relating to a plurality of physical dimensions and a plurality of edges of a three dimensional object, relating to a plurality of physical dimensions and a plurality of edges of a moiety represented by a symbol and its symbol property and used in association with the three dimensional object, and relating to an insertion point of the moiety the symbol represents within the three dimensional object, from a hard copy source document having recorded thereon a plurality of drawing views of the three dimensional object, the alphanumeric text, and the symbol, into mathematically accurate vectors corresponding to the physical dimensions and the edges of the three dimensional object and the moiety the symbol represents; the method comprising the steps of:

(a) acquiring a computer useable raster image of the mechanical drawing having

(i) a plurality of drawing views thereon of a three dimensional object having a plurality of physical dimensions and a plurality of edges, said drawing view having a plurality of lines corresponding to said edges of said three dimensional object, said lines oriented in a direction relative to each other;

(ii) at least one symbol disposed on at least one said drawing view, said symbol having a symbol property, said symbol associated with a moiety in the three dimensional object; said moiety having a plurality of physical dimensions and a plurality of edges expressed as said symbol property, said symbol having an insertion point on said drawing view corresponding to a moiety insertion point of the moiety in the three dimensional object; and

(iii) a first alphanumeric text relating to said plurality of physical dimensions and to said plurality of edges of said three dimensional object and recorded on said drawing view in association with said lines on said drawing views; a second alphanumeric text relating to each said symbol property; and a third alphanumeric text relating to said insertion point; and organizing said raster image according to an orthographic viewpoint raster file for each said drawing view by selecting one said drawing view and said first alphanumeric text associated with said lines on said drawing view, said symbol and said third alphanumeric text; and organizing a symbol property viewpoint raster file for said second alphanumeric text;

(b) setting up a drawing file in a CAD applications program in a computer; said CAD applications program having a coordinate geometry subroutine for creating three dimensional vectors, an optical symbol recognition subroutine, an optical character recognition subroutine and a plurality of viewports; said drawing file having a preset symbol library, said symbol library having an attributed symbol therein corresponding to said symbol on said drawing view, said attributed symbol having a block of mathematically accurate three dimensional symbol vectors representing said physical dimensions and said edges of the moiety said symbol represents in a vector symbol file; and selecting said viewports into at least one orthographic viewport and customizing at least one floating viewport;

(c) importing said symbol property viewpoint raster file into one said floating viewport;

(d) repeating step (c) for each said second alphanumeric text relating to each said symbol on said document;

(e) importing one said orthographic viewpoint raster file of one said drawing views into one said orthographic viewport which corresponds in orthogonality to both said drawing view and said orthographic viewpoint raster file;

(f) repeating step (e) for each drawing view on said document;

(g) recognizing in said optical character recognition subroutine said second alphanumeric text in each said floating viewport, then recognizing said first and third alphanumeric text for each said orthographic viewport; and creating a first, a second and a third recognized alphanumeric text;

(h) calling up said symbol library and selecting said attributed symbol corresponding to said symbol and selecting said second recognized alphanumeric text for said symbol and incorporating said symbol property into said vector symbol file;

(i) recognizing each said symbol in said optical symbol recognition subroutine in each said orthographic viewport separately;

(j) converting in each said orthographic viewpoint said recognized first alphanumeric text in said coordinate geometry subroutine into a plurality of mathematically accurate vectors corresponding to said alphanumeric text recorded on said drawing and to the physical dimensions and edges of the three dimensional object; and placing said vectors into an orthographic vector file; and

(k) selecting said attributed symbol in said orthographic viewport and converting said third recognized alphanumeric text; inserting said symbol vector file into said orthographic vector file at said insertion point; and creating a vector file and a drawing file corresponding to the physical dimensions and to the edges of the three dimensional object with the moiety inserted therein.

Description

STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT

Not Applicable

BACKGROUND OF THE INVENTION

This invention relates generally to a computer automated system and method for converting a hard copy source document bearing alphanumeric text and symbols relating to the physical dimensions and edges of a three dimensional object and of symbol properties of the symbol used in association with the three dimensional object into a drawing file consisting of three dimensional coordinates and vectors and corresponding to the physical dimensions and edges of the object and to the symbol properties of the symbol, and in particular, to a computer automated system and method for converting engineering drawings and architectural plans. The invention is particularly well-suited for utilizing and converting a raster image of a scanned source document bearing a drawing view, symbols and alphanumeric text relating to heights, widths, depths (lengths), and angles of edges of the three dimensional object and of the symbols into mathematically accurate vector computer drawing files based on the symbols and alphanumeric text scanned from the source document. The invention is also well suited for utilizing a raster image produced by scanning an engineering drawing document having a having at least one view thereon showing the edges of a three dimensional object and having alphanumeric text relating to the lengths and directions of lines and curves corresponding to the edges of the object, where the invention constructs a mathematically accurate drawing of the view from the raster file. The present invention is also well-suited for scanning a hard copy architectural plan bearing symbols and alphanumeric text relating to the size, shape, location and elevation of various structural components, such as, but not limited to, walls, windows, and doors and converting the symbols and alphanumeric text into a mathematically accurate computer drawing of the architectural plan.

With the development of various interactive Computer Aided Drafting (also often referred to as Computer Aided Design) (hereinafter, CAD) software, architects, engineers, their draftspersons, and/or technicians (collectively, hereinafter "users") are able to produce architectural plans and engineering drawings using computer drawing files more easily, quickly and accurately than using traditional hand drafting techniques. Besides the ease, speed and accuracy of producing these plans and drawings, the resultant computer drawing files are easier to edit and alter to create new drawings and plans. They are easier to store; they are easier to share with other technicians; and the drawing files can be exported to other computer applications. In the case of architectural plans/drawings, for example, the finished drawings can be rendered in different views, materials lists can be generated, loads can be calculated and structural members can be tested for integrity. In the case of engineering drawings, for example, structural calculations can be made testing the integrity of design elements and the drawing can be exported to a Computer Aided Manufacturing (CAM) environment. In the CAM environment, exact instructions from the computer program related to the dimensions of an object can be fed to an external manufacturing machinery which in turn produces the object ported in the drawing file. The overall advantages of being able to rapidly test and alter designs is well known in the art.

Revising the CAD drawings is easier and more accurate than revising hand drafted drawings. In CAD, the lines on the drawings are represented as vectors. CAD vectors can be manipulated electronically. With the click of a mouse button, lines can be copied, erased, bisected, offset or rotated. The same operations in a hand drawing environment would involve the use of several tools such as erasers, pencils, protractors, scaling rulers and straight edges. But, what makes CAD qualitatively different from hand drawing, is that the vectors representing the lines in the drawing are actually mathematical expressions and not mere representations of the dimensions of the lines. Therefore, any CAD operation performed on an accurately constructed vector would produce mathematically meaningful results. Complex geometric functions can be performed by manipulating vector entities.

Unfortunately, CAD use is relatively new, having become a significant drafting modality in the past 15 years. Even today, only about 50% of the architectural plans and engineering drawings are being produced in this manner. It is estimated that about 80% of drawings and plans exist in the form of hard copy paper documents.

One of the major tasks facing industry, business and government is organization and storage of architectural and engineering information for further utilization. Currently, much of the information is stored in the form of hand drawn hard copy source documents, e.g., mechanical drawings, such as, engineering drawings and architectural plans/drawings, having various formats and scales. These documents, which are commonly drawn on paper, typically contain a scaled hand drawing of various views and components of the three dimensional (3D) object. As used hereinthroughout, the term "3D" means three dimensional; and the term "2D" means two dimensional. The 3D object may be, but is not limited to, an article of manufacture as drawn in an engineering drawing, or to a structure as drawn in an architectural drawing. As is known in the art, each view (e.g., graphical representation/drawing) reveals information about the shape of the 3D object, showing a plurality of edges of the 3D object and any penetrations within the 3D object. Since each drawing view is 2D, the edges and the penetrations are typically depicted by lines and/or curves. Along with the lines and curves recorded on the view, is alphanumeric text relating to the actual physical dimension and edge of the 3D object, any penetration within the 3D object, and symbol property information regarding any symbol on the drawing. The alphanumeric text which is recorded on the document is frequently written, typed or printed on the document. The alphanumeric text provides information relating to the heights, widths, depths (lengths), radius of curvature (if the line is curved), as well as, directions and locations of these lines and curves. Sometimes, the direction of the line is implicit from the drawing view, as when a pair of adjacent lines are orthogonal to or collinear with each other (e.g. at 90.degree., 180.degree. or 270.degree. to each other).

Symbols may also be present on the view. The symbols include, but are not limited to, symbols for placement of moieties within the 3D object e.g., windows, doors, toilets, electrical outlets, or other features, or symbols for the sizes and shapes of apertures which may fully or partially penetrate the 3D object, or symbols which are indicia of certain shapes of cuts or surface features of the 3D object.

A vexatious problem, largely unattended in the art, is the lack of an accurate easy, quick and cost effective conversion of these hand drawn hard copy source documents depicting 3D objects into computer based drawings which are mathematically accurate and based on the recorded alphanumeric text present on the hard copy source document. Conversion is the process of taking a analog hard copy source document and changing it into a digital format suitable for use in a digital computer environment.

Some prior art has attempted to respond to some of the problems of converting this paper based information into a usable, reliable computer file. Unfortunately current methods are too slow, too inaccurate, too costly (because of labor intensiveness) or useful only for two dimensional representations. The prior art methods can be grouped into two major categories: manual entry and automated methods.

Manual entry of drawing information into a CAD program is slow and labor intensive and requires the user to read the alphanumeric text and symbols from sometimes a large unwieldy paper drawing and enter it into the computer. The user must constantly go back and forth between paper document and computer. This slow process is subject to possible error and delay. The advantages of the manual entry method are accuracy, when done correctly, and avoiding the costs of the hardware and software required by automated methods. Because it is labor intensive, it is not well suited for performing large volumes of conversions.

A number of automated methods are known in the art of conversion of cartographic documents (describing two dimensional (2D) land areas surveyed) for inputting hard copy document survey data into a computer file, manipulating the data using known coordinate geometry (COGO) software and CAD software packages to produce vectorized computer drawings. The most popular methods fall into two categories: the "manual digitizing method," and the "scanning method used with a vectorization of the graphics." [See, P. J. Stevenson, Scanning for Automated Data Conversion of Cartographic Documents, 1994, titled "Report No. 426 Department of Geodetic Science and Surveying, The Ohio State University, Columbus, Ohio/Report No. CFM-R-94-101, The Ohio State University Center for Mapping Columbus, Ohio", pp. 1-95, the disclosure of which is incorporated herein by reference; G. Omura, Mastering AUTOCAD B for DOS, SYBEX, Alameda, Calif. 1995, pp. 452-480]. The scanning method used with a vectorization of the graphics typically uses heads-up digitizing, line following and automated raster vectorization for converting scanned raster files to vector files. Major problems with using these methods include inaccurate drawing of the area surveyed on the hard copy document, scaling errors, typographical errors in keying typing in text, and scanner distortion of the information on the face of the hard copy document. Both the aforementioned manual digitizing method and automated methods for use with cartographic documents are described in detail along with the limitations of each method, in U.S. patent application Ser. No. 08/445,687 filed May 22, 1995, page 3, line 20 through page 5, line 27, the disclosure of which is incorporated herein by reference.

Also known in the art is an apparatus and method for manipulating scanned documents in CAD. [see, e.g. U.S. Pat. No. 5,353,393 to Bennett et al.] CAD generated images are overlaid over scanned raster images. The 2D CAD image is generated by tracing over the raster image or by using standard CAD commands. This is basically a 2D technique, since 3D CAD drawing packages were unknown at that time. The creation of the resultant 2D CAD drawing view has potential inaccuracy due to using a scaled raster drawing (which itself may be inaccurately drawn or may suffer from scanner distortion) or the data may be typed in incorrectly by the user in response to the CAD queries, or may be misread by the user of the CAD program.

Also optical character recognition has been most effective when recognizing text that is perfectly horizontal and of a standard font type. Recently, non-standard fonts and hand written text are recognized by the OCR. In a typical drawing or map to be converted, the alphanumeric text of concern is often written along the same angle as the line to which it refers. Techniques for recognizing hand drawn graphic symbols on scanned engineering drawings to produce vectorized graphic data, are known in the art. [See, Bhaskaran (U.S. Pat. No. 4,949,388).] Likewise, methods of processing information from scanned hard copy documents containing text or text and graphics and use of OCR recognized text is known. (Lech et al., U.S. Pat. No. 5,258,855). However this recognition process ignores the meaning behind the symbols and the text. It simply replaces the raster entity with either a vector representation or an ASCII based text string.

The same limitation exists with Optical Symbol Recognition (OSR). OSR works by comparing raster images of symbols with a pre-established library of symbols. The OSR recognizes the raster symbol and converts the raster symbol into a vector version of the same symbol and then places the vector version of the symbol in approximately the same location on the raster drawing image as the raster image of the symbol. The user is still left to relate a symbol property to the symbol. By "symbol property" or "symbol property information" are meant hereinthroughout a physical dimension(s) and shape of a moiety that the symbol represents. The user must locate the symbol property and manipulate the drawing to incorporate this information into the drawing.

Recently a computer automated system and method for converting source documents bearing alphanumeric text relating to the length and directions of the bounding lines of an area surveyed using OCR recognition of the alphanumeric text has been discovered.(WO 96/37861). The alphanumeric text relating to the length and direction of the bounding lines of a land area surveyed treats only a two dimensional (2D) representation, e.g. the bounding lines are treated as in the same XY plane.

Engineering drawings and architectural plans are significantly different from the bounding lines of a land area surveyed because the lines (and/or curves) designating the edges of a surface of the 3D object are not necessarily in the same plane, although on the drawing (or plan) which is a 2D representation of the 3D object, they may appear to be so. The aforementioned inaccuracies of automated methods are particularly critical in a three dimensional environment because various edges of the surface of the 3D object must fit together exactly. Therefore, conventional automated conversion strategies are, at best, only capable of producing two dimensional results.

Also engineering drawings and architectural plans further differ from land survey maps in that the former have symbols thereon which are also associated with physical three dimensional symbol properties related to the moiety the symbol represents and to the placement of the moiety in the 3D object. The symbols may also have associated alphanumeric text providing the physical dimensions of the symbol property.

Despite recognition and study of various aspects of hard copy source document conversion to digital format, the prior art has produced very little in the way of providing an accurate computer automated system and method of converting a raster file of a scanned hard copy source document bearing a drawing view, symbols and alphanumeric text relating to the physical dimensions and edges of the 3D object being rendered, as well as the symbol associated with the 3D object, into mathematically accurate 3D vectors of the 3D object and of moiety the symbol represents and into a mathematically accurate vector computer drawing file based on the drawing, the symbol and alphanumeric text which was scanned from the source document.

BRIEF SUMMARY OF THE INVENTION

The present invention responds specifically to the long-felt need heretofore unmet by the prior art and especially with a way to overcome the intrinsic inaccuracies of converting a raster file image of a scanned hard copy source document depicting 3D objects into computer based drawings which are mathematically accurate and based on the recorded alphanumeric text, the graphical drawing, the symbol (if present) present on the hard copy source document. The present invention provides an accurate easy, quick and cost effective conversion of these scanned hard copy documents. The present invention lends itself to use by those with certain disabilities since there is no need to manually type any recognized alphanumeric text, since this is easily accomplished by "picking" the recognized alphanumeric text. The unique attributed symbols of the present invention advantageously provide time and cost savings eliminating the need for the user to draw the moiety the symbol represents in each of the views in the converted hard copy source document.

The foregoing, and other advantages of the present invention, are realized in one aspect thereof in a computer method of converting alphanumeric text relating to a plurality of physical dimensions and to a plurality of edges of a three dimensional object, from a hard copy source document having recorded thereon the alphanumeric text and a plurality of drawing views of the three dimensional object, into mathematically accurate vectors corresponding to the physical dimensions and the edges of the three dimensional object. The method comprising the steps of: (a) acquiring a computer useable raster image of the document having a plurality of drawing views thereon of a three dimensional object having a plurality of physical dimensions and a plurality of edges; the drawing views having a plurality of lines corresponding to the edges of the three dimensional object; the lines oriented in a direction to each other; and alphanumeric text relating to the plurality of physical dimensions and to the plurality of edges of the three dimensional object and recorded on the drawing view in association with the lines on the drawing view; and organizing the raster image according to an orthographic viewpoint raster file for each of the drawing views by selecting one of the drawing views and the alphanumeric text associated with the lines on the drawing view; (b) setting up a drawing file in a CAD applications program in a computer; the CAD applications program having a plurality of orthographic viewports, a coordinate geometry subroutine for creating three dimensional vectors, and an optical character recognition subroutine, and selecting one of the orthographic viewports to correspond in orthogonality to one of the orthographic viewpoint raster files; (c)importing one orthographic viewpoint raster file into one orthographic viewport which corresponds in orthogonality one drawing view; (d) repeating step (c) for each drawing view on the document; (e) recognizing the alphanumeric text in the optical character recognition subroutine in each of the orthographic viewports separately; (f) converting the recognized alphanumeric text in the coordinate geometry subroutine into the mathematically accurate three dimensional vectors corresponding to the alphanumeric text recorded on the document. The document for use in step (a) is a mechanical drawing or an engineering drawing or an architectural plan having a plurality of drawing sheets.

In another aspect, the present invention provides a computer method of converting alphanumeric text relating to a plurality of physical dimensions and a plurality of edges of a three dimensional object, relating to a plurality of physical dimensions and a plurality of edges of a moiety represented by a symbol and a symbol property and used in association with the three dimensional object, and relating to an insertion point of the moiety the symbol represents within the three dimensional object, from a hard copy source document having recorded thereon a plurality of drawing views of the three dimensional object, the alphanumeric text, and the symbol, into mathematically accurate vectors corresponding to the physical dimensions and the edges of the three dimensional object and the moiety the symbol represents. The method comprising the steps of: (a) acquiring a computer useable raster image in a raster file of the document having (i) a plurality of drawing views thereon of a three dimensional object having a plurality of physical dimensions and a plurality of edges, the drawing view having a plurality of lines corresponding to the edges of the three dimensional object, the lines oriented in a direction to each other; (ii) at least one symbol disposed on at least one drawing view, the symbol having a symbol property, the symbol associated with a moiety in the three dimensional object; the moiety having a plurality of physical dimensions and a plurality of edges expressed as the symbol property, the symbol having an insertion point on the drawing view corresponding to a moiety insertion point of the moiety in the three dimensional object; and (iii) a first alphanumeric text relating to the plurality of physical dimensions and to the plurality of edges of the three dimensional object and recorded on the drawing view in association with the lines on the drawing views; a second alphanumeric text relating to the symbol property; and a third alphanumeric text relating to the insertion point; and organizing the raster image according to an orthographic viewpoint raster file for each drawing view by selecting one drawing view and the first alphanumeric text associated with the lines on the drawing view, the symbol and the third alphanumeric text; and organizing a symbol property viewpoint raster file for the alphanumeric text relating to the symbol properties of each symbol, and; (b) setting up a drawing file in a CAD applications program in the computer, the drawing file having a symbol library, the CAD applications program having a coordinate geometry subroutine for creating three dimensional vectors, an optical symbol recognition subroutine, an optical character recognition subroutine and a plurality of viewports, and selecting the viewports in at least one orthographic viewport and customizing at least one floating viewport; (c) importing the symbol property viewpoint raster file into one of the floating viewports; (d) repeating step (c) for each alphanumeric text relating to the symbol properties of each symbol on the document; (e) importing one of the orthographic viewpoint raster files of one drawing views into one of the orthographic viewports which corresponds in orthogonality to both the drawing view and the orthographic viewpoint raster file; (f) repeating step (e) for each drawing view on the document; (g) recognizing the alphanumeric text in the optical character recognition subroutine in each floating viewport and in each orthographic viewport separately and creating recognized alphanumeric text; (h) creating an attributed symbol by converting the recognized alphanumeric text relating to the symbol property of the symbol into a block of mathematically accurate three dimensional symbol vectors representing the physical dimensions and the edges of the moiety the symbol represents and creating a vector symbol file of the symbol vectors; (i) recognizing each symbol in the optical symbol recognition subroutine in each orthographic viewport separately; (j) converting in each orthographic viewpoint the recognized alphanumeric text relating to the plurality of physical dimensions and the plurality of edges of the three dimensional object in the coordinate geometry subroutine into a plurality of mathematically accurate vector corresponding to the alphanumeric text recorded on the drawing; and placing the vectors into an orthographic vector file; (k) selecting the attributed symbol in the viewport and converting the alphanumeric text relating to the insertion point and inserting the symbol vector file into the orthographic vector file and creating a vector file and a drawing file corresponding to the physical dimensions and to the edges of the three dimensional object with the moiety inserted therein. The document for use in step (a) is a mechanical drawing or an engineering drawing or an architectural plan having a plurality of drawing sheets.

In still another aspect for a mechanical drawing, one follows the aforementioned steps (a)-(k), however the symbol library in step (b) is a preset symbol library having an attributed symbol therein corresponding to the symbol on the drawing view. The attributed symbol has a block of mathematically accurate three dimensional symbol vectors representing the physical dimensions and the edges of the moiety the symbol represents. The symbol vectors are in a vector symbol file. Step (h) comprises calling up the symbol library and selecting the attributed symbol corresponding to the symbol on the drawing view and selecting the recognized alphanumeric text relating to the symbol property of the symbol recorded on the drawing view, and incorporating the symbol property into the vector symbol file.

In yet another aspect, the present invention discloses an automated conversion system for converting alphanumeric text, relating to a plurality of physical dimensions and a plurality of edges of a three dimensional object and of properties of a symbol used in association with the three dimensional object and of an insertion point of the symbol within the three dimensional object, recorded on a hard copy source document having at least one drawing view of the three dimensional object, the alphanumeric text and at least one symbol thereon, into mathematically accurate three dimensional vectors corresponding to the physical dimensions and the edges of the three dimensional object and the properties of the symbol and the insertion point. The system comprising in combination an automatic digitizing unit for document scanning and a computer having a three dimensional CAD applications program. The computer includes receiving software and recognition software. The program includes conversion software, transport software and vectorization software. The automatic digitizing unit scans a hard copy document having at least one drawing view thereon, having at least one symbol thereon and having alphanumeric text relating to a plurality of physical dimensions and a plurality of edges of a three dimensional object and of properties of the symbol used in association with the three dimensional object, and of an insertion point of the symbol within the object and then creates a digitized raster file corresponding to the alphanumeric text, the drawing view and the symbol, and then outputs the digitized raster image. The receiving software is operatively associated with the automated digitizing unit, for receiving the digitized raster image into the CAD applications unit and creates a digitized raster viewpoint file and transports the viewpoint file into a CAD drawing file, having a viewport. The recognition software for recognizing the digitized raster viewpoint file text in the viewport, and creating an ASCII text file comprises an optical character recognition subroutine operating in the CAD applications program for recognizing the alphanumeric text, and comprises an optical symbol recognition subroutine operating in the CAD application program for recognizing the symbol. The conversion software is operatively associated with the recognition software, for converting the ASCII text file into a converted file useable in the coordinate geometry subroutine. The transport software is responsive to the converted file and is for transporting the converted file into the coordinate geometry subroutine. The vectorization software is operatively associated with the transport means and is for converting the converted file into mathematically accurate three dimensional vectors having X, Y, Z coordinates and representing the plurality of physical dimensions and the plurality of edges of the three dimensional object and of the properties of the symbol used in association with the three dimensional object, and comprises a coordinate geometry subroutine. The document used in the system is a mechanical drawing or an engineering drawing or an architectural plan having a plurality of drawing sheets.

In yet still another aspect, the present invention provides a computer quality control method for a mechanical drawing. The method comprises the steps of: (a) scanning a mechanical drawing hard copy source document using an automated digitizing unit, the document having at least one drawing thereon of a three dimensional object and alphanumeric text relating to a physical dimension and an edge of the three dimensional object having a plurality of physical dimensions and a plurality of edges, (b) receiving digitized output, the output including the alphanumeric text and the digitized drawing from an automated digitizing unit into a CAD applications program, the CAD applications program having a coordinate geometry subroutine, the alphanumeric text and the drawing having been scanned from the mechanical drawing hard copy source document; (c) displaying the digitized drawing; recognizing the alphanumeric text using a conversion system operating in the CAD applications program, the conversion system having a graphics optical character recognition subroutine, and recognizing the alphanumeric text using the textual optical character recognition subroutine; (e) transporting the alphanumeric text into the coordinated geometry subroutine; (f) converting the alphanumeric text in the coordinate geometry subroutine into mathematically accurate vectors representing the physical dimensions and the edges of the three dimensional object; (g) generating a set of orthographic views of the three dimensional object using the mathematically accurate vectors and overlaying each orthographic view on a corresponding view of the digitized drawing. The mechanical drawing scanned in step (a) is an engineering drawing or an architectural plan having a plurality of drawing sheets.

Other advantages and a fuller appreciation of the specific attributes of this invention will be gained upon an examination of the following drawings, detailed description of preferred embodiments, and appended claims. It is expressly understood that the drawings are for the purpose of illustration and description only, and are not intended as a definition of the limits of the invention.

BRIEF DESCRIPTION OF THE DRAWING(S)

The preferred exemplary embodiment of the present invention will hereinafter be described in conjunction with the appended drawing wherein like designations refer to like elements throughout and in which:

FIGS. 1A and 1B illustrate the automated conversion system of the present invention;

FIG. 2 is a diagram illustrating the seven steps of the automated conversion method of the present invention;

FIG. 3 is an example of an engineering drawing used to demonstrate the method of the present invention for use in engineering drawings;

FIG. 4 is a diagram illustrating the automated conversion system and method of the present invention and further detailing Step 1 of FIG. 2 for the engineering drawing of FIG. 3;

FIG. 5 is a diagram illustrating the automated conversion system and method of the present invention and further detailing Step 2 of FIG. 2 for the engineering drawing of FIG. 3;

FIG. 6 is a diagram illustrating the automated conversion system and method of the present invention and further detailing Step 3 of FIG. 2 for the engineering drawing of FIG. 3;

FIG. 7 is a diagram illustrating the automated conversion system and method of the present invention and further detailing Step 4 of FIG. 2 for the engineering drawing of FIG. 3;

FIG. 8 is a diagram illustrating the automated conversion system and method of the present invention and further detailing Step 5 of FIG. 2 for the engineering drawing of FIG. 3;

FIG. 9 is a diagram illustrating the automated conversion system and method of the present invention and further detailing Step 6 of FIG. 2 for the engineering drawing of FIG. 3;

FIG. 10 is a diagram illustrating the automated conversion system and method of the present invention and further detailing Step 7 of FIG. 2 for the engineering drawing of FIG. 3;

FIG. 11 is a diagram illustrating the automated conversion system and method of the present invention and further detailing Step 1 of FIG. 2 for the architectural plan of FIG. 18;

FIG. 12 is a diagram illustrating the automated conversion system and method of the present invention and further detailing Step 2 of FIG. 2 for the architectural plan of FIG. 18;

FIG. 13 is a diagram illustrating the automated conversion system and method of the present invention and detailing Step 3 of FIG. 2 for the architectural plan of FIG. 18;

FIG. 14 is a diagram illustrating the automated conversion system and method of the present invention and further detailing Step 4 of FIG. 2 for the architectural plan of FIG. 18;

FIG. 15 is a diagram illustrating the automated conversion system and method of the present invention and further detailing Step 5 of FIG. 2 for the architectural plan of FIG. 18;

FIG. 16 is a diagram illustrating the automated conversion system and method of the present invention and further detailing Step 6 of FIG. 2 for the architectural plan of FIG. 18;

FIG. 17 is a diagram illustrating the automated conversion system and method of the present invention and further detailing Step 7 of FIG. 2 for the architectural plan of FIG. 18; and

FIG. 18 is an example of an architectural plan used to demonstrate the method of the present invention for use in architectural plans.

DETAILED DESCRIPTION OF THE INVENTION

The present invention relates broadly to a computer automated system and method for converting a hard copy source document bearing alphanumeric text relating to a plurality of physical dimensions and edges of a three dimensional object and at least one drawing view of the three dimensional object into a computer drawing file consisting of three dimensional coordinates and vectors corresponding to the edges of the three dimensional object. The computer automated system and method of the present invention is most particularly adapted for use in converting engineering drawings and architectural plans. The invention is particularly well-suited for utilizing and converting a raster image of a scanned hard copy source document bearing a drawing view of the three dimensional object, symbols and alphanumeric text relating to heights, widths, depths (lengths), and angles of edges of the three dimensional object and of the symbol into mathematically accurate vector computer drawing files, which files are based on the symbols and alphanumeric text scanned from the source document. The invention is also well-suited for utilizing a raster image produced by scanning an engineering drawing document having at least one view thereon showing the edges of a three dimensional object and having alphanumeric text relating to the lengths and directions of the lines and curves of the edges, where the invention constructs a mathematically accurate drawing of the drawing view from the raster file. The present invention is also well-suited for scanning a hard copy architectural plan bearing symbols and alphanumeric text relating to the size, shape, location and elevation of various objects, such as, but not limited to walls, windows and converting the symbols and alphanumeric text into a mathematically accurate computer drawing of the architectural plan.

Accordingly, the present invention will now be described in detail with respect to such endeavors; however, those skilled in the art will appreciate that such a description of the invention is meant to be exemplary only and should not be viewed as limitative on the full scope thereof.

As used throughout herein, "document" means the hard copy source document. The document may be of paper, vellum, blue prints, polyester film, plastic transparencies, microfilm, or microfiche. The present invention also can use images scanned by video recording devices and computer or digital cameras.

Conventionally, as is known in the mechanical drawing art, each drawing view on the document is an orthographic projection (or drawing) on a plane of the edges of the 3D object. Perpendicular lines or projectors are drawn from all points on the edge (linear or contour) of the 3D object to the plane of projection. Thus X, Y, Z information of a Cartesian coordinate system is revealed in combinations of the views, or possibly in a single view with additional dimensional information recorded thereon (e.g. a top view of a shim having a depth dimension). In orthographic projection drawings, the 3D object is viewed from 6 mutually perpendicular directions, thus 6 orthographic views may be drawn. These orthographic views are a top view, a bottom view, a front view, a back view, a right side view and a left side view. The principle orthographic projection views are the top (or bottom), front (or rear), right (or left side). The orthographic views of the three-dimensional object are conventionally placed on a drawing sheet in sequence of adjacent sides of the 3D object of interest. Views are most commonly arranged by a third-angle projection system in engineering drawings, in which the top view is drawn in a location vertically above the front view, and the side view is placed horizontally in direct line with the front view (see, FIG. 3). Thus a horizontal plane is used for a top (or bottom) view. A profile plane is used for a side view and a frontal plane is used for the front view(a rear plane is used for the rear view). The horizontal plane, frontal plane and profile plane are orthogonal to each other. Most engineering drawings also include an isometric view or a perspective view of the 3D object in the upper left-hand corner of the completed source document.

Conventionally, for architectural plans (drawings), including structural drawings, a separate sheet is used for each plan view (top view or bottom view) or building elevation view (rear view, front view, left side view or right side view) of a structure, and the individual parts of the structure are shown in assemblage. Most architectural and structural drawings are third-angle projections.

Each of the mechanical drawing views (e.g., engineering drawings and/or architectural plans or drawings and/or other technical drawings) show lines (straight lines or curved lines) corresponding to the various edges of a surface of the three dimensional object, as well as, alphanumeric text associated with these edges.

Symbols may also appear on the engineering drawing views, as well as on the architectural plan views. Conventionally, there are two types of symbols, simple symbols and content symbols. Each of these symbols is drawn on the document using a symbol graphic which has a particular shape associated with a moiety that the symbol represents, or is a single alphanumeric character, or a string of alphanumeric characters. A simple symbol represents a moiety which is not part of the structure, per se, of the 3D object. Thus the simple symbol is merely a conventional shape (graphic) located appropriately on the drawing to show the presence of the moiety represented by the symbol. Simple symbols include, but not limited to, symbols for electrical outlets, toilets, chairs, desks or other furniture. There is no 3D information on the drawing view associated with the moiety represented by the simple symbol.

The term "content symbol" means hereinthroughout that the symbol is related to a moiety which has a symbol property, e.g. 3D content information regarding the physical dimensions and edges of moiety. The moieties represented by content symbols include, but are not limited to, items incorporated into the structure of the 3D object, voids or penetrations within the 3D object, a surface feature of the 3D object, or other features which somehow affect the physical dimensions and edges of the 3D object. An example of an item incorporated into the surface of a 3D object includes, but is not limited to, doors, windows, etc. An example of penetrations (either partial or full) within the 3D object include, but are not limited to, holes, threads for machine screws or bolts, etc. Indicia of surface features of the 3D object include, but not limited to, contours or angulation of surfaces, such as but not limited to, radius of curvature of a surface, e.g., "R" or for a special slope of the edges.

Content symbols have two aspects: A symbol graphic and a symbol property. The symbol graphic is merely a graphical drawing representing a shape for the symbol. For example, is an architectural symbol meaning window, and is an architectural symbol meaning door. Or, for example, the following engineering symbols: , , .O slashed. or DIA, R respectively mean countersink, deep (depth), diameter, and radius. The , , , , .O slashed. or DIA, and R shapes are the symbol graphics of the respective symbols. The other aspect of the content symbol is the symbol property. The symbol property is 3D information pertaining to the physical dimensions and edges of the moiety the symbol represents. This information is found (in whole or in part) as alphanumeric text in drawing blocks, tables, charts, legends or schedules on the drawing or as alphanumeric text on the face of the drawing in proximity with the symbol or in a CAD symbol library. In conventional symbol recognition (OSR), the raster symbol (whether it is a simple symbol or a content symbol) shape is compared to a number of vector based shapes in a symbol library. The vector shape most closely resembling the raster shape is selected and then placed in approximately the same location as the raster version. This is basically a "cut" and "paste" procedure. Conventionally, whenever a content symbol is found, the user is expected to refer to the table or legend to discover the meaning of the symbol and to draw the moiety the symbol represents into the drawing views.

Many simple symbols and content symbols are standardized symbols which are known in the mechanical drawing art. The term standardized symbol(s) means hereinthroughout those simple symbols and content symbols which have standard symbol shapes and/or alphanumeric text and/or which bear three dimension information as to a moiety the symbol represents; standardized symbols, include, but are not limited to, those described in industry standards, such as, ANSI Y14.5 M-1982P29-34, (1983) pp 29-34; 29-133 (American National Standard, Dimensioning and Tolerancing, ANSI Y14.5 M-1982, New York, N.Y.); ANSI Y14.6-1978, (1978) pp. 7-13 (American National Standard, Screw Thread Representation, ANSI Y14.6-1978, New York, N.Y.); ISO 725-1978(E) pp. 151-153, ISO-263-1973(E) pp. 146-150, ISO 724-1978(E) pp. 72-73 (International Organization for Standardization, Switzerland, "ISO Standards Handbook 18, Fasteners and Screw Threads", (1984); as well as those in standard technical drawing books, such as, but not limited to, T. E. French and C. J. Vierck, "Engineering Drawing and Graphic Technology", 12th Ed. McGraw Hill Book Company, New York, N.Y., (1978), pp. 253-254, Appendix C, p. A48-A49, A55; French et al., "Mechanical Drawing CAD-Communications, 12th Ed., pp. 308-309, 723 Glencoe-McGraw-Hill Company, New York, N.Y., (1997); J. W. Grachino et al., "Engineering Technical Drafting", 4th Ed., American Technical Society Publication Chicago, Ill. (1978) pp. 702,704, 705; J. H. Earle, "Graphics for Engineers" Addison Wesley Publishing Co., (1996), Reading, Mass., pp. Appendices A-6, A-7, A-8, A-14; Giesecke et al., "Engineering Graphics", MacMillan Publishing Co., Inc., 3rd Ed. New York, N.Y. (1985) pp. 357-359, 370-373, 834-836, 843); and in existing CAD drawing programs, and manuals such as, but not limited to, T. M. Schumaker et al. "AutoCAD and its Applications", The Goodheart-Willcox Company, Inc., Tinley Park, Ill. (1993) p. G-1, 18-60, 18-61, 18-32, 18-33, 18-31, 18-37, all the aforementioned disclosures of which are incorporated herein by reference.

The present invention creates a third symbol, the "attributed symbol". As used herein, "attributed symbol", means a content symbol whose symbol properties have been converted by the method of the present invention and exist as a 3D vector file where the vectors correspond to the physical dimensions and edges of the moiety the symbol represents and which is assembled into the mathematically accurate vectors corresponding to the physical dimensions and edges of the 3D object and is inserted at the location of the moiety in the 3D object, as shown as the insertion point of the content symbol in the drawing view. This aspect of the present invention, eliminates the time consuming hand entry of the symbol property information for each content symbol and eliminates errors in miskeying and misreading this information. The content symbols, if present on the document, are advantageously converted to attributed symbols according to the present invention. Many standardized content symbols will have been converted into attributed symbols which are part of the default symbol library.

The format as used herein for recording the alphanumeric text are well known in the mechanical drawing drafting art. Where the recorded alphanumeric text describes a straight line, typically the line has two properties, a distance and a direction. The distance is expressed in units of measurements, such as, feet and inches, represented by the alphanumeric text (ft) or (in) or by the symbols (') and (") or in units of measurement, such as, meters, represented by the symbol (m). The direction is expressed as an angle or an azimuth. Alternatively, the direction is ascertained by viewing the drawing on the document. Usually the lines on the drawings are arranged in an orthogonal mode, which means most of the straight lines are in the cardinal directions: 0.degree., 90.degree., 180.degree. and 270.degree.. By definition, lines drawn in these cardinal directions are parallel or perpendicular to each other. Forming corresponding orthogonal vectors is particularly easy using the methods of this invention.

The invention includes hardware and software necessary to extract, retrieve and convert the drawing view, the symbols and the alphanumeric text, relating to the depth, width, height and shape of edges of the three dimensional object and of any symbol property of any content symbol recorded on the document into mathematically accurate vectors.

As used herein, the term "vector" or "vectors", means straight lines having length and direction, as well as, curved lines having length and direction. A vector "straight line" or vector line is described by two points: the start point and the endpoint. A vector "curved line" or vector curve (or arc) can be described by three points: the start point, the endpoint and the center point. As is known in the art, the direction of the curve is implicit in the relationship among these points.

As used herein, the term "mathematically accurate" vector(s) or vector file means a vector(s) or vector file comprising vectors which relate accurately to the mathematical data expressed in the alphanumeric text recorded on the source document which describes the physical dimensions and edges of the three dimensional object. When "mathematically accurate" vector(s) are used in association with the content symbol, the term means a vector(s) or a vector file which relate accurately to the mathematical data expressed in the alphanumeric text recorded on the source document as a symbol property or a vector file present in the symbol library, each of which describes the physical dimension and the edges of the three dimensional moiety the content symbol represents. The term "mathematically accurate" in reference to "computer "drawing(s)" or "computer drawing files" means hereinthroughout creating a computer drawing or computer drawing file using the aforementioned mathematically accurate vectors for the 3D object and/or for the moiety.

As used hereinthroughout, the term "file" is meant to encompass, i.e., that group of information which is in the form of a certain data type. For example, a "viewpoint raster text file" is that group of alphanumeric text expressed as raster text; "a mathematically based ASCII file" is that group of ASCII text which pertains to the mathematical properties of lines and arcs. In another example, "a viewpoint raster symbols file" is a group of raster shapes for the symbol graphics for simple symbols and content symbols. For convenience, these files are typically separated into layers within the CAD drawing itself which can be deleted, frozen, turned on or off, and otherwise manipulated, as is commonly known in the CAD art. The word "file" does not necessarily refer to a discrete bundle of information located at a specific address in the memory of the computer and may include information which is separated within the drawing itself.

CONVERSION SYSTEM AND METHOD

The automated conversion system 100 and method of the present invention is generally shown in FIGS. 1A-18. FIGS. 1A-B, shown without an isometric viewport or viewpoint, illustrate the hardware and software needed to practice the method of the present invention. The hardware for practicing the present invention includes the following: an automated digitizing unit 102 and a computer 104 in a computer work station 106. The automated digitizing unit 102 is suitably any type of a scanner 108, for example an Optical Reader, which extracts information 110 off a document 112, having a document face 116. The document 112 has one or more drawing views 160 thereon. The scanned information which is a digitized output 118 from the face 116 of the document 112 can be saved as simple raster file on a storage device 120 such as a floppy disk, a ZIP.TM. disk (made by lomega Corporation, Roy, Utah), a compact disk (CD), a standard DVD disk, or a hard disk and converted, according to the method of the present invention at a later time. A raster image 119 of the digitized output is shown on the screen. The digitized output 118 is preferably immediately organized into a plurality of viewpoints 122, with each viewpoint 122 corresponding to one of the drawing views 160. This results in the creation of a plurality of digitized viewpoint raster files 124. Each viewpoint raster file 124 is named and then stored on a storage device 120 such as a floppy disk, a ZIP.TM. disk, a compact disk (CD), standard DVD disk or hard disk and then retrieved for use to continue the method of the present invention.

The computer work station 106 suitably includes the computer 104, a high resolution color display monitor (either a VGA or SVGA) 128 with a screen 130 and a high resolution color display card, a keyboard 132, a two button mouse 134, a hard disk (not shown), and a plotter 136. The computer has at least 32 MB (megabytes) of RAM, at least one disk drive capable of reading a 3.5 inch 1.44 MB disk, or a 100 MB ZIP.TM. disk and a hard disk with at least 1 Gigabyte or more of free memory space. The computer 104 should have at least one serial port or a switch box (not shown) installed. An optional printer 138 may be used instead of, or in addition to, the plotter 136.

The software in accordance with the present invention is commercially available and includes a document scanning software program, an operating systems program having "windows," a conversion program with an OCR subroutine and with an OSR subroutine and with automated graphics recognition vectorizing software, a CAD program having a COGO subroutine, and, if necessary, a subroutine for converting a scanned image file into a format that can be used in the specific CAD environment.

In a preferred embodiment of the system and method of the present invention, the following hardware and software are used: the scanner 108 is an Optical Scanner, Contex FSC 8000DSP Full Scale Color Scanner by Contex Scanning Technology (hereinafter "CONTEX") and distributed by Ideal Scanners and Systems of Rockville, Md. The computer 104 is a Dell Dimension XPS R400 MHz Mintower Base personal computer with MMX Technology and 512K Cache by Dell of Round Rock, Tex. with a Pentium II chip, 256MB 100 MHz SDRAM memory, and 6.4 GB of hard disc space. The display monitor 128 is a Dell UltraScan 1600 HS 21" Trinitron Color Monitor with 19.8" viewable image size (Model #1626 HT for Dimension). The plotter 136 is a Hewlett Packard 500 series ink jet plotter. The keyboard 132 is a Silitek Quiet Key 104 key keyboard, factory installed by Dell. The mouse 134 is a two-button MICROSOFT Intellimouse, shipped with the Dell system. The hard disk is a 6.4 GB EIDE Ultra ATA hard disk, factory installed by Dell. The printer 138 is a HEWLETT PACKARD LASERJET 6P.

The preferred software for use in the present invention includes a scanning software CAD/IMAGE/SCAN.TM. software by CONTEX accompanying the aforementioned scanner 108, a CAD software system (AutoCAD.TM. Release 14.0, hereinafter "AUTOCAD", by Autodesk, Sausalito, Calif.) with a COGO subroutine. The preferred recognition and vector conversion software programs used in the AutoCAD.TM. 14.0 environment, include preferably the IMAGE TRACER PROFESSIONAL.TM. (IIS f/k/a I.G.S., a division of Hitachi Inc. of Boulder, Colo.). The IMAGE TRACER PROFESSIONAL.TM., software includes IMAGE EDIT software for editing images, and IMAGE AutoGT.TM. software and IMAGE TRACER.TM. software, both of which are software for fully automatic raster-to-vector conversion. The IMAGE TRACER PROFESSIONAL.TM. for AutoCAD.TM. software is a user interactive fully automatic raster-to-vector conversion tool that runs on AutoCAD.TM. Release 14.0 and requires a Pentium Processor, 32 MB RAM, 10 MB hard disk space, and WINDOWS.RTM. 95 software. It is the core software to which the preferred embodiment of the present invention is written. The present invention uses the text recognition capabilities within IMAGE TRACER PROFESSIONAL.TM. software for automatic conversion of raster text, including handwritten, upper/lower case, and rotated text, into true AutoCAD.TM. text strings. The aforementioned software provides the user interactive series of commands permitting the user to control automatic recognition of graphics, text, or both, as functions in AutoCAD.TM.. The IMAGE TRACER PROFESSIONAL.TM. software also contains utility functions permitting the user to set parameters for specific types of drawings, specific graphics within the drawing, and text recognition. The AutoCAD.TM. software includes a Reject Editor "FRED" software to alert the user to text recognition errors noted by the IMAGE TRACER PROFESSIONAL.TM. software. The IMAGE TRACER PROFESSIONAL.TM. software digitizes all parts of a scanned drawing graphics into vector files as instructed by the user and converts both text and graphics into vector files, e.g. vectorized text and vectorized graphics, when instructed to do so by the user. The "vectorized text" created by the IMAGE TRACER PROFESSIONAL.TM. software means that each text letter is shown in vector format. This is distinguished from the present invention where, in a mechanical, engineering, or architectural drawing, the alphanumeric text recorded on the drawing view relating to the physical dimensions and edges of a three dimensional object and to the symbol properties of a symbol used in association with the three dimensional object is converted to vectors corresponding to edges of a three-dimensional object and to vectors corresponding to the moiety the symbol represents.

The IMAGE TRACER PROFESSIONAL.TM.'s software has a "Graphic Defaults" program which assists the user's vectorizing of the drawing on the scanned document. For example, by the selecting certain options, the "Graphic Defaults" program sets parameters to recognize line intersections and places vectors at each intersection (node). The IMAGE TRACER PROFESSIONAL.TM. software's text recognition module identifies and converts typed or handwritten alphanumeric characters and creates CAD useable ASCII text strings. The converted graphics and converted text are given different colors and different CAD layers. Dimension text is stored on a separate CAD layer.

The preferred software for the OSR for use in the present invention is SYMBOL.TM. symbol recognition software which is commercially available from GRAPHIKON GmbH of Berlin, Germany, and useable with a raster to vector conversion program, such as VECTORY.RTM. by GRAPHIKON GmbH (supra). An additional DFX interface is used to directly connect SYMBOL.TM. software to the CAD software. The SYMBOL.TM. software permits symbols to be recognized on both vector and raster images. This software allows the user to search a vectorized drawing for definable symbols and exchange them with standardized symbols or customized CAD symbols.

As best shown in FIGS. 1A, 3, 11 and 18, the document 112 has one or more drawing views 160 of a 3D object thereon. The 3D object has a plurality of physical dimensions and a shape. The shape of the 3D object is represented as a plurality of edges, which are drawn as lines 164 or curves 166 on the drawing view 160 forming a drawing graphic 167. Each line 164 and each curve 166 has associated with it alphanumeric text 168 relating to the physical dimension of the edge, and hence of the shape of the 3D object. There may also be alphanumeric text 170 relating to the direction of the line and/or curve with respect to an adjacent line and/or curve. The alphanumeric text 168, 170 provides information relating to a height, a width, a depth (length), a radius of curvature (for curves ) as well as direction of the line (in certain instances) and a location of the line 164 and/or the curve 166. Alphanumeric text 168, 170 is also hereinthroughout referred to as a "first alphanumeric text" relating to the physical dimensions and edges of the three dimensional object which is recorded on the document 112. The drawing view 160 also contains information regarding adjacent pairs of edges and their directions on the drawing view 160. Thus the lines 164 may be at right angles to one another and oriented at orthogonal directions: 0.degree., 90.degree., 180.degree. and 270.degree.. The document 112 may have one or more symbols 174 recorded thereon. The symbol may be a simple symbol 175 or a content symbol 176 having a symbol property 178. The symbol property 178 is expressed as an alphanumeric text 180. The alphanumeric text 180 is also referred to hereinthroughout as "a second alphanumeric text" recorded on the document face 116 of the document 112 scanned and is contained in tabular form, such as, a chart, a schedule, a table 188, a legend, etc. The symbol property has 3D information as to the physical dimensions and edges of a moiety the symbol represents, e.g. a height, a width, a depth (length), a radius of curvature (if curved). Or, the symbol property 178 may be a block of vectors in a vector symbol file, containing or associated with this 3D information. The moiety the symbol represents may be, but is not limited to, a penetration (full or partial) 184 within the 3D object, or an extrusion of the 3D object, or an object (door, window, etc.) incorporated into the 3D object. Where a content symbol 176 is present on the document 112, an alphanumeric text 181 is present which relates to an insertion point 182 of the moiety of the symbol into the 3D object. Alphanumeric text 181 is hereinthroughout also referred to as a third alphanumeric text. Furthermore, a drawing scale 187, as well as, a measurement of units 189 are on the source document 112.

The present method of conversion of a document 112 such as, but not limited to, a mechanical drawing, such as an engineering drawing 190 or an architectural plan 192, is practiced by following steps 1-7 as illustrated in FIG. 2. The same steps 1-7 are illustrated in further detail in FIGS. 4-10, in Example 1, for the engineering drawing 190 and in FIGS. 11-17, in Example 2, for the architectural plan 192 and in FIG. 1.

The computer method of the present invention has 7 simple steps. These steps are (1) acquiring the image, (2) setting up the drawing, (3) importing the viewpoint raster files, (4) recognizing the text, (5) recognizing the symbols, (6) converting the text, and (7) converting the symbols. If no symbols are recorded on the document 112, the computer method of the present invention includes steps (1)-(4) and (6). If only simple symbols are present on the source document 112, the computer method of the present invention includes steps (1)-(6).

The term "text" in FIGS. 1A, 1B, 2, and 4-17 refers to the alphanumeric text 168, 170, 180 and 181.

STEP 1

Step 1 of acquiring the image is illustrated in steps 1.1 to 1.4 as best seen in FIGS. 1A, 2, 3, 4, 11, and 18. In FIGS. 4-10, the document 112 of FIG. 3 is used. In FIGS. 11-17, the document 112 of FIG. 18 is used. In step 1.1, in the conventional manner, the source document 112 is scanned using a scanner 108. The resolution of the scanner 108 is set to at least 300 dots per inch (300 DPI). The hard copy source document 112 has one or more drawing views 160 thereon, as well as alphanumeric text 168, 170 (relating to the physical dimensions and edges of the 3D object) and symbols 174 (if present), and alphanumeric text 180, 181 (if content symbols 176 are present) on the face 116 of the document 112. The document 112 may have other alphanumeric text 194 which does not relate to the depths, widths and heights of the 3D object or to the symbol property 178. This alphanumeric text 194 may include information on a project name, a revision number of the document 112, a name of the drafter, a drawing scale 187 of the drawing views 160 or, a unit 189 of measurement for which the drawing views 160 is dimensioned, etc. A title block 196 is typically in a corner of the engineering drawing 190.

In step 1.1, all the information is extracted from the face 116 of the document 112 and stored in the scanner memory as an unedited digitized raster image 119. The raster image 119 can be saved as a simple raster file 200 in a user designated for mat (e.g. GIF, TIF). Most preferably, the raster image 119 is directly imported into the computer 104 without being stored in the scanner memory. In step 1.1, the digitized raster image is displayed on the display monitor 128 as raster image 119. Raster image 119 displays all the scanned content on the face 116 of the document 112, showing the drawing views 160 in the source document 112 as well as text 168, 170, 180 and 181 and any symbols 174. For ease of illustration, the hidden lines on the drawing in step 1.1 have been omitted and are omitted in subsequent illustrations. However in actual practice, any hidden lines on the drawing appear on the raster image 119.

In step 1.2, raster image 119 is edited, if needed, using software provided with the scanner and/or with other editing computer software as is known in the art, to remove scanner artifact, e.g., noise. For example, the raster image 119 may be reoriented or realigned, artifact, such as "noise", removed, etc. If the raster image 119 is edited, an edited raster file 204 is created. At this point, the raster files 200, 204 may be stored in a storage device 120 as a simple raster file 200 in computer memory, on hard drive, or on a floppy disk, or compact disk (CD), or output as a simple raster drawing 203 using a standard printer 138 or a standard plotter 136. This file 200, 204 may be brought up to practice the present invention.

After the raster image 119 has been adequately edited, in order to practice the present invention, it is necessary for the user to organize the digitized raster file 200, 204 or the edited raster image 119 into viewpoint raster files which correspond to corresponding views on the document and to organize any tables, charts or schedules on the document into an appropriate viewpoint raster file(s).

In step 1.3, each of the digitized raster file(s), either the unedited raster file 200 or the edited raster file 204, or edited raster image 119 is then organized according to a viewpoint 122 corresponding to one of the views 160 on the document 112 or to a symbol property 178 in the document 112, and in step 1.4 suitably stored in a storage device 120 as a named viewpoint raster file in computer memory, on a hard drive, or on a floppy disk, on ar compact disk (CD), or is output as a viewpoint raster drawing 209 using the printer 138 or the plotter 136.

As discussed previously, mechanical drawings 190, 192 have conventional drawing views 160 of a 3D object. The viewpoints 122 include orthographic viewpoints 210. There are six orthographic viewpoints 210 which correspond to the six principle orthographic views 212 used to represent a 3D object. The six orthographic views 212 are a top view 214, a left side view 216, a right side view 218, a front view 220, a back view 222, and a bottom view. The corresponding orthographic viewpoints 210 are a top viewpoint 226, a left side viewpoint 228, a right side viewpoint 230, a front viewpoint 232, a back viewpoint 234 and a bottom viewpoint and are given the respective abbreviations T, L, R, F, B, and Bo. There are also viewpoints 122 corresponding to any isometric view 238 or perspective view on the source document 112. These corresponding view points are an isometric viewpoint 240 and a perspective viewpoint and are given the abbreviation I and P respectively. The alphanumeric text 168, 170, 181 associated with the specific view 160 is included in the drawing viewpoint 210 for the specific view 160.

Other viewpoints 122 include one or more floating viewpoints 242 which contain the symbol property 178 and the alphanumeric text 180. If the symbol property 178 is not present in the source document 112, then a floating viewpoint 242 is not set up. The floating viewpoint 242 is given the short abbreviation FV. The floating viewport 242 appears as a moveable window 244 on the screen 130 which are manipulated, enlarged, or diminished in a manner well known to anyone familiar with the WINDOWS.RTM. operating environment software. These windows 244 are set to activate automatically as a part of a command or are called up deliberately at any time by the user.

For the engineering drawing 190, typically one source document 112 is scanned which may contain more than one drawing view 160, as shown in FIG. 4. The digitized raster image 119 is divided into separate orthographic views 214, 216, 218, 220, 222, and bottom view corresponding to each separate view 160 on the engineering drawing 190, and stored in a separate viewpoint 122 corresponding to the view 160, as a corresponding separate viewpoint raster file 226', 228', 230', 232', 236', and bottom viewpoint raster file. If an isometric view 238 is present, this is placed in the isometric viewpoint 240 and stored as an isometric viewpoint file 240'. If a perspective view is present, this is placed in the perspective viewpoint and stored as a perspective viewpoint file. If symbol property 178 is present on the engineering drawing 190, this is placed in the floating viewpoint 242 and stored as described in step 1.4 as a floating viewpoint raster file 242'.

Alternatively, for architectural plans 192 since different views 160 are typically represented on a plurality of separate sheets 112' as shown in FIG. 18. The sheets are scanned sequentially, as shown in FIG. 11, and a separate raster image 119 is created, for each sheet 112' scanned. Each raster file image 119 is then edited to remove artifact and organized according to a viewpoint 226, 228, 230, 232, 236, 240 and bottom viewpoint and to any floating viewpoint 242, named as described in step 1.4 and stored as a set of digitized viewpoint raster file(s) 124.

In step 1.4, each of the digitized viewpoint raster files 124 is given a file name 252 consisting of three components: a general file name 254, a viewpoint designation 256 (corresponding to the view on the hard copy source document) and a suffix indicating the raster file format 258. The digitized viewpoint raster file 124 is then stored under the file name. The general file name 254 can be any name that relates to the 3D object or to the project. The viewpoint designation 256 using the aforementioned short abbreviations, T, L, R, F, B, Bo, I, P or FV is used. The abbreviation GIF is used to indicate the raster file format 258.

For the engineering drawing 190, the digitized viewpoint raster file is given a general file name 254 such as "WIDGET". Each view is named by adding viewpoint designations 256 to the general file name 254 followed by the raster file format 258, e.g. WIDGET-T.GIF, WIDGET-F.GIF, WIDGET-R.GIF, and WIDGET.-I.GIF. Thus, a top viewpoint raster file 226' named WIDGET-T.GIF corresponds to the top view 214. A front viewpoint raster file 232' named WIDGET-F.GIF corresponds to the front view 220. A right side viewpoint raster file 230' named WIDGET-R.GIF corresponds to the right side view 218, and an isometric viewpoint raster file 240' named WIDGET-I.GIF corresponds to the isometric view 238 of the engineering drawing 190 scanned. Other engineering drawings may have other views shown and suitable viewpoints are assigned to these views and file names given to the digitized viewpoint raster files. For any tabular information, charts or schedules on the document, the general file name 254 may be used along with the abbreviation "FV" for floating viewpoint 242 and the raster file format 258 "GIF.", to provide a floating viewpoint raster file 242', e.g. "WIDGET-FV.GIF". In FIG. 4, this raster file corresponds to the "TABULAR DIMENSION" table scanned off the engineering drawing 190, shown in FIG. 3. Alternatively, as shown in FIG. 11, the general file name 254 may be an indicator of what it represents, e.g. "TABLE" with a raster block named by adding a viewpoint designation 256 to the general file name 254 followed by the raster file format 258, e.g. "TABLE-1.GIF". In FIG. 11, this viewpoint raster file "TABLE-1.GIF" is a floating viewpoint raster file 242', corresponding to the "DOOR SCHEDULE" scanned from the document 112.

The viewpoint raster file 124 has a viewpoint raster text 125 file corresponding to the alphanumeric text, a viewpoint raster graphic file 126 (corresponding to a drawing graphic 167) and a viewpoint raster symbol file 127, corresponding to the symbols 174 on the drawing.

Steps 1.1-1.4 are accomplished using the scanner software. The named raster viewpoint files 124 are preferably stored in an AUTOCAD directory in the CAD operating program in raster file format.

STEP 2

Step 2 of setting up a CAD drawing file is illustrated in steps 2.1-2.3 of FIGS. 2, 5 and 12. Prior to conversion of the alphanumeric text 168, 170, 180, 181, 194 and symbols 174 in steps 6 and 7, the user creates a CAD drawing file 260 which includes a raster file format and a set of parameter settings 264 of a drawing 266 to be constructed. In step 2, the user enters the CAD environment in the computer 104. Creation of the CAD drawing file 260, including selection of basic parameter settings 264 and setting up of the drawing file 260 is radically different from prior art methods. In the prior art, conventional CAD drawing methods have been used to produce a set of vector drawings which look like drawings on the face 116 of the original document 112. Conventional methods use some form of line tracing or graphics recognition to produce vectors which overlay the raster image. The resulting prior art vectors are only visually accurate and, at that, only as accurate as the drafter of the source documents. Since a three dimensional drawing demands that the edges of the 3D object fit together exactly, prior conversion practices are not accurate enough to be suitable to generate the necessary vectors to do this.

Since this invention uses the alphanumeric text 168, 170 relating to the physical dimensions and edges of the 3D object to produce mathematically accurate vectors 268 corresponding to these physical dimensions and edges, it is possible to achieve mathematically accurate results in each view 160 and therefore it is possible to generate an accurate three dimensional vector file 269 corresponding to the physical dimensions and edges of the 3D object. Most advantageously, the edges of the 3D object fit together using the present invention. The selection of basic parameter settings 264 and the setup of the CAD drawing file 260 is accordingly radically different from that used in the prior art.

In step 2.1, the user assigns the CAD drawing file 260 a file name 270 that corresponds with the general file name 254 of the viewpoint raster file name 252 given in step 1.4. In step 1, e.g. for FIG. 4, the file name 254 is "WIDGET" and in step 1 for FIG. 11, the file name is "HOUSE".

The user then selects a prototype drawing file 274 which matches a type 276 of document(s) 112 scanned. In the present invention, prototype drawing files 274 fall into two major types 276: "ARCH" drawing types 278 for architectural drawings/plans 192 and "MECH" drawing types 280 for mechanical engineering drawings 190. These two major types 276 are further subdivided into prototypes drawing files 274 based on a sheet size 282 of the hard copy document 112. These sheet sizes 282 are "A" for 8.5 inch by 11 inch, "B" for 11 inch by 17 inch, "C" for 18 inch by 24 inch, "D" for 24 inch by 36 inch and "E" for 36 inch by 48 inch documents. Thus the prototype drawing files 274 are labeled "ARCH-A", "ARCH-B", "ARCH-C", "ARCH-D", "ARCH-E", "MECH-A", "MECH-B", "MECH-C", "MECH-D" and "MECH-E" according to the type 276 of source document 112 and the sheet size 282 of the source document 112.

The chosen prototype drawing file 274 is preset with one or more parameter settings 264 to cover most typical mechanical drawings. The parameter settings 264 include settings for precision, dimension type (e.g. meters, inches), line types, menus, layering and font styles. This invention requires use of Optical Character Recognition (OCR) software and Optical Symbol Recognition (OSR) software. The parameter settings 264 for optimizing both OCR recognition and OSR recognition is also included as part of the prototype drawing file 274. This will be discussed in more detail in steps 4 and 5.

Also, the user has the opportunity to customize any parameter settings 264. These customized prototype drawing files are saved as separate prototype drawing files or are limited for use in the current prototype drawing file.

In addition to the preset parameter settings 264, each prototype drawing file 274 also includes a set of viewports 290 corresponding to the previously described views 214, 216, 218, 220, 222, 224, 238 and the perspective view. These viewports 290 conform with conventional mechanical drawing views, but may be customized to include a floating viewport 320, as needed. The viewports 290 are set up in what is conventionally known in AUTOCAD as "paperspace". Depending upon the prototype drawing file 274 chosen, the user is presented with several preset page layouts 292, hereinafter, "pages", of appropriate size of the viewports 290. The user may plot these page layouts 292 at desired settings (e.g., MECH-A prototype pages plot out on a paper at 11 inches by 8.5 inches, or e.g., ARCH-E prototype pages plot out on a paper at 36 inches by 48 inches). Each of the preset "pages" 292 has an array of viewports 290. The viewports 290 can be thought of as holes cut out of paper through which a particular view 160 can be seen. The user selects the "page" 292 which matches the drawing views 160 as recorded on the source document(s) 112. Here in Example 1, shown as step 2.3 on FIG. 5, an 11 inch by 8.5 inch "page" 292 containing four views 160 (see, FIG. 3) is selected. All the other "pages" 292 are automatically not displayed. The viewports 290 may be modified, if needed, in several ways. More "pages" 292 can be created; viewports 290 can be expanded or contracted in size; viewports 290 can be located on separate "pages" 292.

Each viewport 290 has a name 294 based on one of the standard six orthographic views 212. Thus, six orthographic viewports 296 are named, a TOP viewport 300, a FRONT viewport 302, a RIGHT viewport 304, a BOTTOM viewport (not shown), a BACK viewport 308, and a LEFT viewport 310, to correspond to the corresponding orthographic views, 214, 220, 218, 224, 222 and 216, respectively. In addition, an ISOMETRIC viewport 312 and a PERSPECTIVE viewport, may be set up to correspond to the isometric view 238 and the perspective view, if present on the source document 112. The isometric viewport 312, the perspective viewport and the orthographic viewports 296 are also called fixed viewports 314. The viewports 290 selected in the CAD drawing file 260 are those as shown/recorded on the source document 112. For example, for the mechanical drawing of FIG. 3, the CAD drawing file 260, as shown in FIG. 5, is given the CAD drawing file name 270, "WIDGET", and the viewports 290 are named TOP, FRONT, RIGHT and ISOMETRIC. Along with the fixed viewports 314, one or more floating viewports 320 may also be used. The floating viewport 320 is called FLOATING VIEWPORT. The floating viewport 320 is primarily used to capture and present the symbol property 178. Similarly, for architectural drawings 192, appropriate fixed viewports 314 and floating viewports 320 containing the symbol property 178 are created. Fixed viewports 314 and floating viewports 320 are collectively referred to as viewports 290. The floating viewport 320 appears in a moveable window 244 on the screen 130. The moveable window 244 is manipulated, enlarged, or diminished in a manner well known to anyone familiar with WINDOWS.RTM. operating environment software. The floating viewport 320 "floats" over a currently used underlying fixed viewport 314. The window 244 is set to activate automatically as a part of a command, or is called up at any time by the user. For example, the command sequence "Attribute Symbols" (step 5.3) (which will be discussed subsequently) automatically displays the OSR program and the floating viewport 320 containing the symbol property 178, e.g. information/data regarding the symbol 174.

For the purpose of converting a mechanical drawing 190, 192, the user is primarily concerned with the basic three orthographic projections: the top view 214, the front view 220 and the side view 216 or 218, and hence, their corresponding viewports, the top viewport 300, the front viewport 302 and the side viewport 304 or 310. As is known in AUTOCAD, each of the fixed viewports 314 has its own User Coordinate System (UCS). The UCS conveniently allows the user to view each view 160 as though it is a two-dimensional plan view. In the UCS, for each of the viewports there is a horizontal direction, hereinafter, "X" or "X coordinate" and a vertical direction, hereinafter, "Y" or "Y coordinate". This is true even if the viewport, e.g. the view is in the XY, XZ, or YZ plane corresponding to the front view, top view, or side view. Hereinafter prompts "X" or "Y" are in the UCS and refer to the "horizontal direction" and the "vertical direction". Using the AUTOCAD software, a UCS vector created in each UCS for a particular orthographic viewport is automatically converted into a corresponding vector 268 having 3D coordinates, e.g. "X, Y, Z" in the three-dimensional World Coordinate System (WCS). Because of the AUTOCAD software, the drawing in the ISOMETRIC viewport 312 (or the perspective viewport) is constructed indirectly. As the vectors 268 of the orthographic viewports 296 are produced according to the present invention, the vectors 268 appear simultaneously as vector lines 322 or vector curves 324 in the isometric viewport 312 (or the perspective viewport). The vectors 268 are combined to create a vector file 269 representing the 3D-object in a 3D space. As a vector 268 is produced in the UCS, it is given a set of 2D coordinates 332 in the UCS and also automatically given a corresponding set of 3D coordinates 334 in the WCS. The WCS serves as an absolute framework within which all the vectors 268 are integrated and the 3D object defined in the vector file 269. The vector file 269 may be used to create a 3D computer model, e.g. perspective view, or isometric view which can be further manipulated. The vector file 269 also may be used in association with a standard Computer Aided Manufacturing (CAM) program to generate the 3D object

In addition, each viewport 290 has its own current viewport layer 336 which serves as the layer upon which the digitized viewpoint raster file 124 is placed. The viewport layers 336 are named according to the viewpoint raster files 124 (raster-t for top, raster-f for front, raster-FV for floating view, etc.). The layer 336 for one viewport 290 is frozen in all other viewports 290 so that each viewport 290 has only its corresponding viewpoint raster file 124 view as the current viewport layer 336.

Also the CAD applications program advantageously has another preset parameter setting 264, a preset symbol library 171. The preset symbol library 171 according to the present invention includes the aforementioned standardized symbols, (e.g., common abbreviations and common graphical symbols used in technical drawings, such as, but not limited to, those described in industry standards, such as, those in standard technical drawing books, and in existing CAD drawing programs), the disclosures of which are incorporated herein by reference. The symbol library 171 contains vector-based shape files of each of the symbols as well as preset "attributed" symbols corresponding to specified dimensions of the standardized symbols, (e.g., the physical dimensions and edges of a moiety the standardized symbol represents). Creation of attributed symbols will be explained in step 5 and in step 5 in Examples 1 and 2.

Also the standardized symbols which are content symbols 176 may be imported from databases having the vector-based shaped file for the symbol as well as the symbol properties 178 in CAD useable vector format corresponding to the standardized symbol.

In step 2.2, the aforementioned preset parameter settings 264 of step 2.1 are further customized if needed. In some cases it may be necessary to change some of the preset parameter settings 264, such as, text parameters for measurement unit type (e.g. inches, meters), precision settings, text size (measured in pixels) and dimensioning style. Preset parameter settings 264 are chosen according to the prototype drawing type 276. These parameter settings 264 are selected to achieve optimal recognition levels for the symbols 174 and alphanumeric characters 168, 170, 180, 181, 194 typically used in different types of drawings. For example, the OCR is optimized to recognize dimensions expressed in feet (') and inches (") and to recognize numbers "0, 1, 2, 3, 4, 5, 6, 7, 8, 9" as well as the alphabet letters, and text strings, such as those alphanumeric text strings used for the symbols 174 in the preset symbol library, the disclosure of the alphanumeric text strings of the aforementioned standardized symbols is herein incorporated by reference, and any punctuation marks and frequently used sequences of text, such as "cm" or "ft" or "feet". The alphanumeric text strings used for the standardized symbols include, but are not limited to standardized thread descriptors, such as, but not limited to standard metric and English thread notes e.g., "3/8-24 UNF-2A", symbols "", "", terms such as, "DEEP", "DEPTH", "COUNTERBORE", "CBORE", "CSINK", "COUNTERSINK", "DEGREES", "SQUARE", etc. The OSR is optimized to recognize the symbols 174 found in the symbol library and also for such structural elements as doors or windows.

If symbols 174 are used in the document 112, which are not present in the symbol library 171, the user can add symbols 174 to the library by constructing vector drawings to match the raster symbol. A toolbox containing several common shapes, simple vector elements and basic CAD tools enables the user to easily assemble vector based symbols. This is discussed in step 5.2.

In step 2.3, the viewports 290 are customized, if needed. The user sets each viewport 290 to a current layer 336 named after the view 160/viewpoint 124. The user can customize the viewports 290 if the preset page layout 292 is inadequate. An orthographic viewport 296 can be deleted; additional viewports can be created, such as, an isometric viewport 312 or a perspective viewport. The user may also customize the CAD drawing file 260 by setting up one or more floating viewports 320. In step 2, an organized array of viewports 290 is created in the CAD environment into which the viewpoint raster files 124 are systematically imported.

STEP 3

Step 3 of importing the raster files is illustrated in steps 3.1-3.4 of FIGS. 2, 6 and 13. In the present invention, in step 3, each digitized viewpoint raster file 124 is imported from storage into a corresponding viewport 290 of the CAD drawing file 260 using a "RASTERIN" command, available in the AutoCAD.TM. 14.0 software environment. The CAD environment also includes software capable of converting the viewpoint raster graphic files 126, the viewpoint raster symbol files 127, and viewpoint raster text files 125 into a format usable in the CAD environment, such as Hitachi's IMAGE TRACER PROFESSIONAL.TM. software.

Practically the entire computer automated conversion system 100 and method uses the mouse 134 to enter selected choices from the screen 130 into the computer 104. As is known in the art, a position of the mouse 134 is indicated by a flashing cursor 340 on the screen 130. When some information appearing on the screen 130 is to be selected or chosen, the user moves the cursor 340 to a position on the screen at the location of the information and then the user depresses, e.g. clicks, an appropriate button, on the mouse 134. This transports the selected choice of information into the computer 104 for use in practicing the method of the present invention is accomplished in the WINDOWS.RTM. software environment. The mouse 134 has a left mouse button 344 and a right mouse button 346. The left mouse button 344 is used to execute a function by picking the selected information, e.g., either the alphanumeric text 168, 170, 180, 181, the symbols 174, the viewports 290, a starting point 402, an origin 400, the viewpoint raster files 124, vectors 268, a vector end point 406, or a vector beginning point 404, etc. The right mouse button 346 generally passes on the default or a pick and invokes an option or a series of options, as will be explained later. As used throughout herein, "pick," "picking," "picks on," "picking on," "pick upon," or "picked" means to move the cursor 340 to the location of the information or choice on the screen and to depress the appropriate mouse button of the mouse 134 and thereby transport the selected information choice to the appropriate location in the method, e.g. OSR, OCR, COGO routine, etc.

In step 3.1, the user picks a chosen viewport 290, or names the desired viewport 290 using a AUTOCAD "VIEW" command. This activates the selected viewport 290. As is known in the art, the viewport 290 may be enlarged to fill the screen, using standard CAD features, such as "FIT TO SCREEN" commands.

In step 3.2, once the selected viewport 290 is active, the user selects the appropriate viewpoint raster file 124 which corresponds in orthogonality to the viewport 290 selected. The user selects the name of the raster viewpoint file 124 to be imported. For example, "WIDGET-F.GIF", the front viewpoint raster file 228', is selected by picking the import option in the "file" pulldown menu familiar to AUTOCAD users. Thus, the user selects a raster viewpoint file 124 corresponding in orthogonality to one of the selected orthographic viewports 296, or to an isometric viewport 312, or to a perspective viewport, or to a floating viewport 320 for symbol properties 178. For example, if a top viewport 300 is activated, then the top viewpoint raster file 226' for the top view 214 on the scanned source document 112 is selected.

In step 3.3 the AUTOCAD program prompts the user for the drawing scale 187. The user enters the drawing scale 187 shown on that part of the source document 112 corresponding to the viewpoint raster file 124 to be imported or shows up on the title block 196.

In step 3.4, the user then imports the appropriate viewpoint raster file 124 into the viewport 290 by picking an appropriate command (e.g. "RASTERIN" in AUTOCAD 14). The viewpoint raster file 124 is automatically imported at that selected drawing scale 187 into the activated viewport 290.

In the CAD environment, the physical dimensions are considered to be real world dimensions. The drawing scale 187 is generally a function of the plotting process and it is typically chosen to advantageously display the drawing views 160 on the document 112. Thus, an edge of the 3D object which is five inches long in the real world is also five inches in the CAD environment regardless of the drawing scale 187. A plan view (top 214, bottom 224, front 220, or back 222) drawn at a 1/4"=1' scale meshes perfectly in the CAD environment with a side view 216, 218 drawn at 1/8"=1' scale.

Drawing scale 187 is also extremely important in properly introducing a viewpoint raster file 124 since the scale expresses a ratio of between the size of a viewpoint raster file image 350 and the actual size of the 3D object represented to the CAD environment. Generally, keying in the drawing scale 187 in response to the prompt for scale 187 is sufficient to enter the viewpoint raster file 124 into the viewport 290 in the proper drawing scale 187. However, if the document 112 was, for example, a reduced copy of an original drawing view 160, the viewpoint raster file 124 would need to be adjusted to fit the scale 187. The CAD environment enables the user to check the accuracy of the importation process. The user merely inquires as to an apparent length between two points of a known physical dimension, recorded on the document 112. If the apparent length and the known dimension, (which is shown as alphanumeric text 168 on the document 112 in a position which relates it to the specific edge of the 3D object) are essentially equal, then the entry has been properly performed. If they are different, then the viewpoint raster file 124 can be adjusted by a factor of the known dimension divided by the apparent length. For example, if the known dimension, e.g. the length of an edge is given as 5 inches. However, suppose the viewpoint raster file image of that edge measures only 2.5 inches (the apparent dimension). By dividing the known dimension by the apparent dimension, one would determine a factor of 2. The image viewpoint raster file 350 is then appropriately enlarged to 2 times its original size.

Checking and adjusting the scale 187 of the viewpoint raster file 124 is important because the viewport raster file image 350 in the viewport 290 serves as a backdrop to the production of the vectors 268. The role the of the viewpoint raster file image 350 in the process of this invention radically differs from the role in prior art raster to vector conversion method. Typical prior art conversion software uses the raster image 119 as the basis for the vectors. Prior art 3D vector drawings are effectively graphic tracings of the raster images 119. In the method of this invention, 3D vectors 268 are produced according to the mathematical data content inherent in the alphanumeric text 168, 170, 180, 181, the symbol properties 178, and the symbols 174. The viewpoint raster file image 350 in the viewport 290 serves two functions. The viewpoint raster file image 350 provides the basis for OSR and OCR conversions. That is, the OCR recognizes raster text from the alphanumeric text 168, 170, 180, and 181 in the viewpoint raster file 124 and converts it into a corresponding ASCII text 352 (recognized alphanumeric text), and the OSR recognizes raster symbols and converts them into vector based symbol blocks, as will be explained later. ASCII text 352 corresponds to a first recognized alphanumeric text 168, 170, a second recognized alphanumeric text 180 and a third recognized alphanumeric text 181. Secondly, the viewpoint raster file image 350 in the viewport 290 provides a backdrop for which the vectors 268 overlay and in this capacity serves as immediate quality assurance on the accuracy of the conversion.

The viewpoint raster file image(s) 350 are each on separate layers. Thus even if these images 350 originate from different drawing scales 187, they are all converted to the same overall dimensioning framework. However, most mechanical drawings 190, 192 are composed of views 160 in the same scale. Thus, when importing the viewpoint raster file 124 sequence of steps 3.1-3.4, in step 3.3, the user responds to the drawing scale 187 prompt by picking a particular drawing scale 187 from a list of possible drawing scales 187. This selected drawing scale 187 becomes the default scale for all views 160 and viewpoint raster files 124.

Ultimately, all the lines 164 and curves 166 of the drawing views 160 of the document 112 are expressed as vectors 268 in the World Coordinate System (WCS). As is known in the art, this system is composed of three dimensional X, Y, Z coordinates. A three dimensional vector file 269 corresponding to the physical dimensions and edges of the 3D object is created by aligning the lines 164/curves 166 corresponding to edges of the 3D object, using the common origin X=0, Y=0, and Z=0 as an index. However, for simplicity's sake, the orthographic projection of the 3D object represented in each view 160 is drawn in its own User Coordinate System (UCS). This allows the user to approach each drawing view 160 from essentially a plan view. Thus, a front view 220 of the "WIDGET" is presented as though it was lying flat on the drafting table.

Once the viewpoint raster file 124 has been imported into the viewport 290 in step 3.4 it is, by default, placed on the current viewport layer 336 in that viewport 290. The current viewport layer 336 (e.g. RASTER-F is current layer of viewport front) in one viewport 290 is frozen in all other viewports 290 so that only the viewpoint raster file 124 pertinent to each specific viewport 290 is displayed accordingly. At this point the arrangement of the viewpoint raster file image 350 superficially resembles the digitized raster file 200,204 in a prior art conversion. The important difference is that in the prior art, all the images on the digitized raster file 200, 204 are essentially on one viewport 290. In the method of this invention, each viewpoint raster file image 350 is in a separate viewport 290 and represents a specific orthogonal projection (and orthographic projection) of the 3D object.

The user repeats steps 3.1 to 3.4 to import each digitized viewpoint raster file 124 into its corresponding viewport 290 in the CAD environment. For the floating viewpoint files 242, the scaling step is not performed. In each viewport 290, only the current raster layer 336 of that viewport 290 is displayed on the screen 130.

STEP 4

Step 4 of recognizing the text is illustrated in steps 4.1-4.4 of FIGS. 2, 7 and 14. In step 4, in the CAD environment, the alphanumeric text 168, 170 relating to a plurality of physical dimensions and a plurality of edges of the three dimensional object and the alphanumeric text 180 relating to the symbol properties 178 of a symbol 174 and the alphanumeric text 181 relating to the moiety's insertion point 182 into the 3D object is recognized and converted into ASCII text 352 via the optical character recognition (OCR) software. The viewpoint raster file 124 is processed by the OCR subroutine. The preferred standard OCR conversion program uses Hitachi's IMAGE TRACER PROFESSIONAL.TM. OCR software.

In step 4.1, the alphanumeric text 168, 170, 180 and 181 displayed on the viewpoint raster file image 350 in each viewport 290 is recognized separately. It is at this point that the user chooses different strategies depending upon the type of prototype drawing 276 and the information present on the mechanical drawing 190, 192. In a mechanical drawing, such as an engineering drawing 190 or an architectural drawing (or plan) 192 having symbols 174 thereon, there are four important sets of information which are captured by the recognition process (both OCR and OSR). These sets of information are: the alphanumeric text 168, 170 relating to the physical dimensions and edges of the 3D object itself, the alphanumeric text 180 relating to the symbol property 178, and the alphanumeric text 181 information relating to the insertion point of the moiety represented by symbol 174, and the symbol 174 used in the mechanical drawing 190, 192. The viewpoint raster file images 350 in the viewports 290 consist of three main types of information: a viewpoint raster graphic 358, relating to the drawing view 160, alphanumeric text 168, 170, 180, 181, and raster images 359, hereinafter, also "raster symbols 359", of the symbols 174. In the method of the present invention, the alphanumeric text 168, 170 associated with the physical dimensions and the edges of the 3D object and the alphanumeric text 180 of the symbol property 178 of any symbol 174 and the alphanumeric text 181 is the basis for the vector production corresponding to mathematically accurate vector 268 for the 3D object and mathematically accurate vectors 500 for the moiety the symbol represents. The role of the viewpoint raster graphic 358 is merely a rough check on the geometry of the vectors 268 being produced by the method of the present invention but the viewpoint raster graphic 358 is not directly converted into the vectors 268.

It is not necessary for the conversion program software to convert the raster graphic 358 into vectors 268 because this is accomplished more accurately by the present invention.

The viewpoint raster alphanumeric text 168, 170, 180, 181 is converted into ASCII text 352 (e.g. recognized alphanumeric text 168, 170, 180, 181) with an OCR operating in the CAD environment. The OCR is capable of recognizing alphanumeric text at any angle. As has been explained, OCR text parameters are preset in the prototype CAD drawing file 274 to maximize recognition of the dimension and style of the type of prototype drawing file 276 to be converted. After the alphanumeric text 168, 170, 180, 181 has been recognized, the user can then edit the resulting ASCII text 352 if necessary. The preset OCR parameters can automatically subject questionable text recognitions to user review.

In step 4.2, the alphanumeric text 180 (of the symbol property 178) in the floating viewports 320, is selected for recognition first because the symbol properties 178 are essential to the symbol recognition process of step 5. For these viewports 320, the OCR text recognition parameters are preset to maximize the recognition of alphanumeric text in a tabular format with tolerances for acceptable degrees of ambiguity and to alert user to possible errors. Recognition of the alphanumeric text 180 in the floating viewports 320 is optimized by setting the text recognition parameters to the horizontal mode. Typically recognition rates in this mode are faster and more accurate than a mode set to read text at any angle.

In step 4.3, the alphanumeric text 180 in the floating viewport 242 is recognized. The recognized tabular text (ASCII text 352) is automatically placed on an appropriate text layer (e.g. WINDOW-TXT, DOOR-TXT, CURVE-TXT ) which indicates the symbol property 178, or the ASCII text 352 on the floating viewport 242 is placed on layer TEXT-FV. If necessary, the user may edit the ASCII text 352 to correct OCR recognition errors. After the alphanumeric text recognition has taken place, any lines which comprise an element of the table 188, schedule, legend, chart or graph are recognized by the graphics recognition software operating in the CAD environment according to preset parameters optimizing the vectorization of typically evenly spaced vertical and horizontal lines.

In step 4.4, the alphanumeric text/raster text 168, 170, 181 in the other viewpoints 290 is recognized. (Any other alphanumeric text 194 is also recognized in steps 4.3 and 4.4 and converted into ASCII text 356.) Here the OCR parameter settings are preset to read text at any angle. The OCR parameters settings are preset to distinguish nonmathematical text strings 366 such as "BEDROOM," "KITCHEN," from physical dimension text mainly by size and somewhat by content. OCR parameters are set such that any text over a given height