The field of machine vision has expanded extensively since the First Edition of Machine Vision was published by Academic Press in 1990. As a result, this Second Edition contains significant amounts of new material on artificial neural networks, mathematical morphology, motion, invariance, texture analysis, x-ray inspection, and foreign object detection. Intermediate level vision is examined in depth (especially Hough transforms), and automated visual inspectionis discussed. The author takes care to consider theoretical aspects as well as practical applications, including perspective invariants and robust statistics. Written in a user-friendly style and full of up-to-date methods, Machine Vision, Second Edition will be an essential volume for students and professionals in the field.
Reihe
Auflage
Sprache
Verlagsort
Verlagsgruppe
Elsevier Science & Technology
Zielgruppe
Für höhere Schule und Studium
Für Beruf und Forschung
Maße
Höhe: 229 mm
Breite: 152 mm
Gewicht
ISBN-13
978-0-12-206092-2 (9780122060922)
Copyright in bibliographic data is held by Nielsen Book Services Limited or its licensors: all rights reserved.
Schweitzer Klassifikation
Roy Davies is Emeritus Professor of Machine Vision at Royal Holloway, University of London. He has worked on many aspects of vision, from feature detection to robust, real-time implementations of practical vision tasks. His interests include automated visual inspection, surveillance, vehicle guidance, crime detection and neural networks. He has published more than 200 papers, and three books. Machine Vision: Theory, Algorithms, Practicalities (1990) has been widely used internationally for more than 25 years, and is now out in this much enhanced fifth edition. Roy holds a DSc at the University of London, and has been awarded Distinguished Fellow of the British Machine Vision Association, and Fellow of the International Association of Pattern Recognition.
Autor*in
Emeritus Professor of Machine Vision, Royal Holloway, University of London, UK
Vision, the Challenge. Part I: Low-Level Processing: Images and Imaging Operations. Basic Image Filtering Operations. Thresholding Techniques. Locating Objects via Their Edges. Binary Shape Analysis. Boundary Pattern Analysis. Part II: Intermediate-Level Processing: Line Detection. Circle Detection. The Hough Transform and Its Nature. Ellipse Detection. Hole Detection. Polygon and Corner Detection. Part III: Application Level Processing: Abstract Pattern Matching Techniques. The Three-Dimensional World. Tackling the Perspective n-Point Problem. Motion. Invariants and their Applications. Automated Visual Inspection. Statistical Pattern Recognition. Biologically Inspired Recognition Schemes. Texture. Image Acquisition. The Need for Speed: Real-Time Electronic Hardware Systems. Part IV: Perspectives on Vision: Machine Vision, Art or Science? Appendices. References. Subject Index. Author Index. Vision, the Challenge: Introduction-Man and his Senses. The Nature of Vision. Automated Visual Inspection. What This Book is About. The Following Chapters. Part I: Low-Level Processing: Images and Imaging Operations: Image Processing Operations. Convolutions and Point Spread Functions. Sequential Versus Parallel Operations. Basic Image Filtering Operations: Noise Suppression by Gaussian Smoothing. Median Filtering. Mode Filtering. Bias Generated by Noise Suppression Filters. Reducing Computational Load. The Role of Filters in Industrial Applications of Vision. Sharp-Unsharp Masking. Thresholding Techniques: Region-Growing Methods. Thresholding. Adaptive Thresholding. Locating Objects via Their Edges: Basic Theory of Edge Detection. The Template Matching Approach. Theory of 3 x 3 Template Operators. Summary-Design Constraints and Conclusions. The Design of Differential Gradient Operators. The Concept of a Circular Operator. Detailed Implementation of Circular Operators. Structured Bands of Pixels in Neighbourhoods of Various Sizes. The Systematic Design of Differential Edge Operators. Problems with the Above Approach-Some Alternative Schemes. Binary Shape Analysis: Connectedness in Binary Images. ObjectLabelling and Counting. Metric Properties in Digital Images. Size Filtering. The Convex Hull and Its Computation. Distance Functions and Their Uses. Skeletons and Thinning. Some Simple Measures for Shape Recognition. Shape Description by Moments. BoundaryTracking Procedures. Boundary Pattern Analysis: Boundary Tracking Procedures. Template Matching-a Reminder. Centroidal Profiles. Problems with the Centroidal Profile Approach. The (s, () Plot. Tackling the Problems of Occlusion. Chain Code. The (r, s) Plot. Accuracy of Boundary Length Measures. Concluding Remarks. Bibliographical and Historical Notes. Part II: Intermediate-Level Processing: Line Detection: Application of the Hough Transform to Line Detection. The Foot-of-Normal Method. Longitudinal Line Localization. Final Line Fitting. Circle Detection: Hough-Based Schemes for Circular Object Detection. The Problem of Unknown Circle Radius. The Problem of Accurate Centre Location. Overcoming the Speed Problem. The Hough Transform and Its Nature: The Generalized Hough Transform. Setting Up the Generalized Hough Transform-Some Relevant Questions. Spatial Matched Filtering in Images. From Spatial Matched Filters to Generalized Hough Transforms. Gradient Weighting Versus Uniform Weighting. Summary. Applying the Generalized Hough Transform to Line Detection. An Instructive Example. Tradeoffs to Reduce Computational Load. The Effects of Occlusions for Objects with Straight Edges. Fast Implementations of the HoughTransform. The Approach of Gerig and Klein. Ellipse Detection: The Diameter Bisection Method. The Chord Tangent Method. Finding the Remaining Ellipse Parameters. Reducing Computational Load for the Generalized Hough Transform Method. Comparing the Various Methods. Hole Detection: The Template Matching Approach. The Lateral Histogram Technique. The Removal of Ambiguities in the Lateral Histrogram Technique. Application of the Lateral Histogram Technique for Object Location. A Strategy Based onApplying the Histograms in Turn. Appraisal of the Hole Detection Problem. Polygon and Corner Detection: The Generalized Hough Transform. Application to the Detection of Regular Polygons. The Case of an Arbitrary Triangle. The Case of an Arbitrary Rectangle. Lower bounds on the Numbers of Parameter Planes. An Extension of the Triangle Result. Discussion. Determining Orientation. Why Corner Detection? Template Matching. Second-Order Derivative Schemes. A Median-Based Corner Detector. The Hough Transform Approach to Corner Detection. The Lateral Histogram Approach to Corner Detection. Corner Orientation. Part III: Application Level Processing: Abstract Pattern Matching Techniques: A Graph-Theoretic Approach to Object Location. Possibilities for Saving Computation. Using the Generalized Hough Transform for Feature Collation. Generalizing the Maximal Clique and Other Approaches. Relational Descriptors. Search. The Three-Dimensional World: Three-Dimensional Vision-the Variety of Methods.Projection Schemes for Three-Dimensional Vision. Shape from Shading. Photometric Stereo. The Assumption of Surface Smoothness. Shape from Texture. Use of Structured Lighting. Three-Dimensional Object Recognition Schemes. The Method of Ballard and Sabbah.The Method of Silberberg et al. Horaud's Junction Orientation Technique. The 3DPO System of Bolles and Horaud. The IVISM System. Lowe's Approach. Tackling the Perspective n-Point Problem: The Phenomenon of Perspective Inversion. Ambiguity of Pose Under Weak Perspective Projection. Obtaining Unique Solutions to the Pose Problem. Motion: Optical Flow. Interpretation of Optical Flow Fields. Using Focus of Expansion to Avoid Collision. Time-to-Adjacency Analysis. Basic Difficulties with the Optical Flow Model. Stereo From Motion. Applications to the Monitoring of Traffic Flow. Invariants and their Applications: Cross Ratios: The 'Ratio of Ratios' Concept. Invariants for Non-Collinear Points. Invariants for Points on Conics. Automated Visual Inspection: The Process of Inspection. Review the Types of Object to be Inspected. Summary-the Main Categories of Inspection. Shape Deviations Relative to a Standard Template. Inspection of Circular Products. Inspection of Printed Circuits. Steel Strip and Wood Inspection. Inspection of Products with High Levels of Variability. X-Ray Inspection. Bringing Inspection to the Factory. Statistical Pattern Recognition: The Nearest Neighbour Algorithm. Bayes' Decision Theory. Relation to the Nearest Neighbour and Bayes' Approaches. The Optimum Number of Features. Cost Functions and Error-Reject Tradeoff. Cluster Analysis. Principal Components Analysis. The Relevance of Probability in Image Analysis. Biologically Inspired Recognition Schemes: Artificial Neural Networks. The Back-Propagation Algorithm. MLP Architectures. Over-Fitting to the Training Data. Optimising the Network Architecture. Hebbian Learning. Case Study: Noise Suppression Using ANNs. Genetic Algorithms. Texture: Some Basic Approaches to Texture Analysis. Grey-Level Co-Occurrence Matrices. Laws' Texture Energy Approach. Ade's Eigenfilter Approach. Appraisal of the Laws and Ade Approaches. Fractal-Based Measures of Texture. Shape from Texture. Markov Random Field Modelsof Texture. Structural Approaches to Texture Analysis. Image Acquisition: Illumination Schemes. Cameras and Digitization. The Sampling Theorem. The Need for Speed: Real-Time Electronic Hardware Systems: Parallel Processing. SIMD Systems. The Gain in Speed Attainable with N Processors. Flynn's Classification. Optimal Implementation of an Image Analysis Algorithm. Board-Level Processing Systems. LSI. Part IV: Perspectives on Vision: Machine Vision, Art or Science? Parameters of Importance in Machine Vision. Tradeoffs. Future Directions. Hardware, Algorithms, and Processes. A Retrospective View. Just a Glimpse of Vision? Appendices. References. Subject Index. Author Index.
