Machine Interpretation of Line Drawing Images

Line drawing interpretation is a challenging area with enormous practical potential. At present, many companies throughout the world invest large amounts of money and human resource in the input of paper drawings into computers. The technology needed to produce an image of a drawing is widely available, but the transformation of these images into more useful forms is an active field of research and development. Machine Interpretation of Line Drawing Images

• describes the theory and practice underlying the computer interpretation of line drawing images and

• shows how line drawing interpretation systems can be developed.
  The authors show how many of the problems can be tackled and provide a thorough overview of the processes underpinning the interpretation of images of line drawings.

  Inhalt

  1. The Line Drawing Interpretation Problem.- 1.1 Motivation.- 1.2 Manual Input vs Document Scanning.- 1.3 Raster-Based vs Vector Representations.- 1.4 The Interpretation Problem.- 1.5 Engineering Drawings and Maps.- 1.6 Line Drawing Interpretation, Image Understanding and Pattern Recognition.- 1.7 Current Line Drawing Interpretation Systems.- 1.8 The Line Drawing Interpretation Literature.- 2. Components of a Line Drawing Interpretation System.- 2.1 Design Criteria for Drawing Interpretation Systems.- 2.2 Five Stages of Line Drawing Interpretation.- 2.3 Intermediate and Target Representations.- 2.4 System Architectures and the Role of A Priori Knowledge.- 3. Document Image Acquisition.- 3.1 Scanning Devices.- 3.2 Image Coding.- 3.3 Image File Formats.- 4. Binarisation.- 4.1 A Taxonomy of Thresholding Techniques.- 4.2 Document Image Statistics.- 4.3 Binarising Line Drawings.- 5. Binary Image Processing and the Raster to Vector Transformation.- 5.1 Raster to Vector Conversion.- 5.2 Some Definitions.- 5.3 The Distance Transform.- 5.4 Mathematical Morphology.- 5.5 Reducing Noise in Binary Images.- 5.6 Reducing Noise in Binary Images of Line Drawings.- 6. Analysis of Connected Components.- 6.1 Nomenclature.- 6.2 Contouring.- 6.3 Skeletonisation.- 6.4 Grey Level Skeletonisation.- 7. Vectorisation.- 7.1 Approaches To Vectorisation.- 7.2 Global Vectorisation Methods.- 7.3 Local Vectorisation Methods.- 7.4 The Hough Transform.- 7.5 Direct Vectorisation.- 7.6 A Vector Database.- 7.7 Removing Noise from the Vector Model.- 7.8 Alternative Raster to Vector Technologies.- 8. Interpreting Images of Maps.- 8.1 Introduction.- 8.2 System Overview.- 8.3 Map Interpretation Principles.- 8.4 A Classification of Map Entities.- 8.5 Interactive Map Interpretation.- 8.6 Output Formats.- 8.7 Quality Issues.- 9. Recognising Cartographic Objects.- 9.1 Recognising Isolines.- 9.2 Recognising Roads.- 9.3 Recognising Texture and Area Objects.- 9.4. Recognising Symbols.- 10. Recovering Engineering Drawing Entities from Vector Data.- 10.1 Design Principles and System Architecture.- 10.2 Vectorisation and Entity Recognition Processes.- 10.3 Extracting Arcs and Straight Lines.- 10.4 Recognising Crosshatched Areas.- 10.5 Recognising Dimensions.- 10.6 Detecting Blocks.- 11. Knowledge-Directed Interpretation of Engineering Drawings.- 11.1 An Image Understanding Approach.- 11.2 Drawing Entities as Schemata.- 11.3 Image Analysis Facilities.- 11.4 The ANON Architecture.- 11.5 Control Issues.- 11.6 Entity Extraction: Chained Lines.- 11.7 Top-Down and Bottom-Up Control.- 11.8 Performance.- 11.9 Scene Formation.- 11.10 ANON in Context.- 11.11 Discussion.- 12. Current Issues and Future Developments.- 12.1 Higher-Level and 3D Representations.- 12.2 Exploiting Domain Knowledge.- 12.3 The Role of the Operator.- 12.4 Performance Measures.- 12.5 Topics for Future Development.- References.