Handbook of Computer Vision Algorithms in Image Algebra

Handbook of Computer Vision Algorithms in Image Algebra provides engineers, scientists, and students with an introduction to image algebra and presents detailed descriptions of over 80 fundamental computer vision techniques. These techniques represent a core of knowledge that all computer vision practitioners should have. The book also introduces the portable iac++ library, which supports image algebra programming in the C++ language. Image algebra provides a concise, high-level mathematical language with which one can specify computer vision and image processing transformations and algorithms. It is clear, lucid, and easy to learn and use. Because of their high level of abstraction, image algebra specifications can be directly implemented on a variety of computer architectures. Programs using the iac++ library can be executed efficiently on both sequential and parallel computer architectures without modification. To aid practitioners in formulating algorithms in succinct and precise mathematical language, computer vision methods are presented with clear mathematical specifications. To assist algorithm implementors in developing robust programs, the practical impact of algorithm variants and implementation choices is discussed.

Table of Contents:
Image Algebra Introduction Point Sets Value Sets Images Templates Recursive Templates Neighborhoods The p-Product Image Enhancement Techniques Introduction Averaging of Multiple Images Local Averaging Variable Local Averaging Iterative Conditional Local Averaging Max-Min Sharpening Transform Smoothing Binary Images by Association Median Filter Unsharp Masking Local Area Contrast Enhancing Histogram Equalization Histogram Modification Lowpass Filtering Highpass Filtering Edge Detection and Boundary Finding Techniques Introduction Binary Image Boundaries Edge Enhancement by Discrete Differencing Roberts Edge Detector Prewitt Edge Detector Sobel Edge Detector Wallis Logarithmic Edge Detection Frei-Chen Edge and Line Detection Kirsch Edge Detector Directional Edge Detection Product of the Difference of Averages Crack Edge Detection Local Edge Detection in Three Dimensional Images Hierarchical Edge Detection Edge Detection Using K-Forms Hueckel Edge Operator Divide-and-Conquer Boundary Detection Edge Following as Dynamic Programming Thresholding Techniques Introduction Global Thresholding Semithresholding Multilevel Thresholding Variable Thresholding Threshold Selection Using Mean and Standard Deviation Threshold Selection by Maximizing Between Class Variance Threshold Selection Using a Simple Image Statistic Thinning and Skeletonizing Introduction Pavlidis Thinning Algorithm Medial Axis Transform (MAT) Distance Transforms Zhang-Suen Skeletonizing Zhang-Suen Transform - Modified to Preserve Homotopy Thinning Edge Magnitude Images Connected Component Algorithms Introduction Component Labeling for Binary Images Labeling Components with Sequential Labels Counting Connected Components by Shrinking Pruning of Connected Components Hole Filling Morphological Transforms and Techniques Introduction Basic Morphological Operations: Boolean Dilations and Erosions Opening and Closing Salt and Pepper Noise Removal The Hit-and-Miss Transform Gray Value Dilations, Erosions, Openings, and Closings The Rolling Ball Algorithm Linear Image Transforms Introduction Fourier Transform Centering the Fourier Transform Fast Fourier Transform Discrete Cosine Transform Walsh Transform The Haar Wavelet Transform Daubechies Wavelet Transforms Pattern Matching and Shape Detection Introduction Pattern Matching Using Correlation Pattern Matching in the Frequency Domain Rotation Invariant Pattern Matching Rotation and Scale Invariant Pattern Matching Line Detection Using the Hough Transform Detecting Ellipses Using the Hough Transform Generalized Hough Algorithm for Shape Detection Image Features and Descriptors Introduction Area and Perimeter Euler Number Chain Code Extraction and Correlation Region Adjacency Inclusion Relation Quadtree Extraction Position, Orientation, and Symmetry Region Description Using Moments Histogram Cumulative Histogram Texture Descriptors: Gray Level Spatial Dependence Statistics Neural Networks and Cellular Automata Introduction Hopfield Neural Network Bidirectional Associative Memory (BAM) Hamming Net Single Layer Perceptron (SLP) Multilayer Perceptron (MLP) Cellular Automata and Life Solving Mazes Using Cellular Automata Appendix: The Image Algebra C++ Library Index