Machine Interpretation of Line Drawing Images : Technical Drawings, Maps and Diagrams

by Ablameyko, Sergey; Pridmore, Tony

• ISBN13: 9783540762072
• ISBN10: 3540762078
• Format: Hardcover
• Copyright: 2000-04-01
• Publisher: Springer Verlag

Summary

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.

Table of Contents

The Line Drawing Interpretation Problem	1	(20)
Motivation	1	(5)
Manual Input vs Document Scanning	6	(2)
Raster-Based vs Vector Representations	8	(1)
The Interpretation Problem	9	(1)
Engineering Drawings and Maps	10	(3)
Line Drawing Interpretation, Image Understanding and Pattern Recognition	13	(2)
Current Line Drawing Interpretation Systems	15	(4)
Commercial Systems	15	(1)
Laboratory Systems	16	(3)
The Line Drawing Interpretation Literature	19	(2)
Components of a Line Drawing Interpretation System	21	(24)
Design Criteria for Drawing Interpretation Systems	21	(3)
Five Stages of Line Drawing Interpretation	24	(3)
Intermediate and Target Representations	27	(12)
Raster Representations	27	(3)
Vector Representations	30	(3)
Universal Drawing Entities	33	(3)
Two-Dimensional Objects	36	(1)
Three-Dimensional Shape and Semantics	37	(2)
System Architectures and the Role of A Priori Knowledge	39	(6)
Sequential Architectures	40	(2)
Blackboard Architectures: The Expert System Approach	42	(3)
Document Image Acquisition	45	(12)
Scanning Devices	45	(5)
Image Coding	50	(3)
Image File Formats	53	(4)
Binarisation	57	(18)
A Taxonomy of Thresholding Techniques	57	(4)
Document Image Statistics	61	(3)
Binarising Line Drawings	64	(11)
Point-Dependent Threshold Selection	64	(7)
Region-Dependent Threshold Selection	71	(2)
Image Partitioning	73	(2)
Binary Image Processing and the Raster to Vector Transformation	75	(26)
Raster to Vector Conversion	75	(2)
Some Definitions	77	(4)
The Distance Transform	81	(9)
Mathematical Morphology	90	(1)
Reducing Noise in Binary Images	91	(5)
Reducing Noise in Binary Images of Line Drawings	96	(5)
Analysis of Connected Components	101	(18)
Nomenclature	101	(1)
Contouring	102	(5)
Goals	102	(1)
Classification of Contouring Techniques	102	(2)
A Contouring Algorithm	104	(3)
Skeletonisation	107	(10)
Motivation	107	(2)
Approaches To Thinning	109	(2)
A Thinning Algorithm	111	(2)
The Medial Axis Transform	113	(2)
Removing Noise from Thinned Data	115	(2)
Grey Level Skeletonisation	117	(2)
Vectorisation	119	(24)
Approaches To Vectorisation	119	(3)
Global Vectorisation Methods	122	(5)
Iterative Methods	122	(2)
Feature-Based Techniques	124	(1)
Hybrid Approaches	125	(2)
Local Vectorisation Methods	127	(3)
Line Following Methods	127	(1)
Raster Scan Techniques	128	(2)
The Hough Transform	130	(3)
Direct Vectorisation	133	(4)
A Vector Database	137	(3)
Removing Noise from the Vector Model	140	(1)
Alternative Raster to Vector Technologies	141	(2)
Interpreting Images of Maps	143	(20)
Introduction	143	(1)
System Overview	143	(2)
Map Interpretation Principles	145	(1)
A Classification of Map Entities	146	(6)
Line Objects	146	(1)
Area Objects	147	(4)
Symbols	151	(1)
Interactive Map Interpretation	152	(7)
Automated Image Interpretation under Operator Control	152	(5)
Operator-Supplied Context	157	(2)
Output Formats	159	(1)
Quality Issues	160	(3)
Basic Concepts	160	(1)
Maintaining Quality during Interpretation	161	(2)
Recognising Cartographic Objects	163	(22)
Recognising Isolines	163	(6)
Overview	163	(1)
Theoretical Background	164	(1)
An Isoline Recognition Procedure	165	(4)
Recognising Roads	169	(7)
Overview	169	(1)
Interpreting the Road Layer	170	(2)
Extracting Roads and Correlated Objects from the Black Layer	172	(4)
Recognising Texture and Area Objects	176	(4)
Overview	176	(1)
Texture Border Definitions	176	(4)
Recognising Symbols	180	(5)
Recovering Engineering Drawing Entities from Vector Data	185	(24)
Design Principles and System Architecture	185	(3)
Vectorisation and Entity Recognition Processes	188	(4)
Extracting Arcs and Straight Lines	192	(3)
Recognising Crosshatched Areas	195	(4)
Recognising Dimensions	199	(4)
Detecting Blocks	203	(6)
Knowledge-Directed Interpretation of Engineering Drawings	209	(34)
An Image Understanding Approach	209	(3)
Drawing Entities as Schemata	212	(2)
Image Analysis Facilities	214	(1)
The ANON Architecture	215	(2)
Control Issues	217	(2)
Entity Extraction: Chained Lines	219	(5)
Top-Down and Bottom-Up Control	224	(2)
Performance	226	(7)
Scene Formation	233	(7)
Searching Image and Memory	233	(3)
Coincidence Links	236	(3)
Partial Interpretations and Schema Fusion	239	(1)
ANON in Context	240	(1)
Discussion	241	(2)
Current Issues and Future Developments	243	(20)
Higher-Level and 3D Representations	243	(10)
Resolving Inconsistency	243	(1)
Editing and Parameterisation	244	(6)
3D Reconstruction	250	(3)
Exploiting Domain Knowledge	253	(3)
The Role of the Operator	256	(1)
Performance Measures	257	(1)
Topics for Future Development	258	(5)
References	263	(18)
Index	281