Segmentation techniques for high quality colour images.

by Wai Leung.Thesis (M.Phil.)--Chinese University of Hong Kong, 1994.Includes bibliographical references (leaves 148-150).Chapter 1. --- Introduction --- p.1Chapter 1.1. --- Image Selection/Segmentation --- p.3Chapter 1.2. --- Background Image Generation --- p.5Chapter 1.3. --- Thesis Organisation --- p.6Chapter 2. --- Fundamentals of Digital Image Segmentation --- p.7Chapter 2.1. --- Edge-based Segmentation methods --- p.7Chapter 2.1.1. --- Edge detection --- p.7Chapter 2.1.1.1. --- Gradient operators --- p.9Chapter 2.1.1.2. --- Compass operators --- p.10Chapter 2.1.1.3. --- Laplace operators and zero crossings --- p.10Chapter 2.1.1.4. --- Stochastic gradients --- p.11Chapter 2.1.1.5. --- Optimal edge detectors --- p.12Chapter 2.1.2. --- Boundary extraction --- p.13Chapter 2.1.2.1. --- Contour following --- p.13Chapter 2.1.2.2. --- Heuristic graph searching --- p.14Chapter 2.1.2.3. --- Dynamic programming --- p.14Chapter 2.2. --- Region-based Segmentation Methods --- p.15Chapter 2.2.1. --- Thresholding --- p.15Chapter 2.2.2. --- Region growing --- p.16Chapter 2.2.3. --- Region splitting and merging --- p.17Chapter 2.2.4. --- Texture segmentation --- p.19Chapter 2.2.4.1. --- Spectral approaches --- p.19Chapter 2.2.4.2. --- Statistical methods --- p.21Chapter 2.3. --- Works on Colour Image Segmentation --- p.23Chapter 3. --- Current Selection Tools for Image Retouching --- p.24Chapter 3.1. --- Selection by Region Growing --- p.24Chapter 3.2. --- Selection by Edge Finding --- p.26Chapter 3.3. --- Some Conclusions --- p.27Chapter 4. --- A New Segmentation Tool for Image Retouching --- p.28Chapter 4.1. --- Requirement and Development Strategy of the Selection Tool --- p.28Chapter 4.2. --- Basic Assumptions --- p.29Chapter 4.3. --- Algorithm of the Image Selector --- p.30Chapter 4.3.1. --- Boundary representation --- p.30Chapter 4.3.2. --- Colour edge detection --- p.31Chapter 4.3.2.1. --- Colour gradient --- p.31Chapter 4.3.2.2. --- Edge detector --- p.32Chapter 4.4. --- Boundary Searching --- p.34Chapter 4.4.1. --- The searching algorithm - A* --- p.35Chapter 4.4.2. --- The cost function and heuristic for A* algorithm --- p.37Chapter 4.4.2.1. --- Pixel cost formulation --- p.38Chapter A. --- The reference dependent function y --- p.39Chapter 1. --- The three similarity functions --- p.42Chapter 2. --- The similarity thresholds --- p.44Chapter 3. --- The n-reference dependent function --- p.45Chapter B. --- The reference independent function f --- p.46Chapter C. --- The actual pixel cost function W --- p.46Chapter 4.4.2.2. --- The heuristic function h --- p.47Chapter 4.5. --- Implementation --- p.48Chapter 4.5.1. --- Work-flow of the image selection tool --- p.48Chapter 4.5.2. --- Implementation of the user-input stage --- p.51Chapter 4.5.3. --- Implementation of the boundary searching phase --- p.54Chapter 4.5.3.1. --- OPEN and CLOSE lists --- p.56Chapter 4.5.3.2. --- Tracing back searching path --- p.57Chapter 4.5.3.3. --- Edge map --- p.58Chapter 4.5.3.4. --- Cost calculation --- p.59Chapter 4.5.4. --- Implementation of the boundary connection phase --- p.61Chapter 4.6. --- Experiments and Results --- p.63Chapter 4.6.1. --- Features exploration --- p.63Chapter 4.6.1.1. --- Cost function of the boundary tracer --- p.63Chapter 4.6.1.2. --- Reference sensitivity --- p.73Chapter 4.6.1.3. --- Boundary connection --- p.75Chapter 4.6.2. --- Comparison with current image selection tools --- p.76Chapter 4.7. --- Discussion --- p.87Chapter 4.8. --- Further Improvement --- p.91Chapter 4.8.1. --- A* algorithm --- p.91Chapter 4.8.2. --- Colour space --- p.92Chapter 4.8.3. --- Improvement in processing speed and quality --- p.93Chapter 5. --- Background Image Generation by Image Interpolation --- p.95Chapter 5.1. --- Current Filling Tools for Background Image Generation --- p.96Chapter 5.1.1. --- The Stamp tool --- p.96Chapter 5.1.2. --- The Gradient Fill tool --- p.96Chapter 5.2. --- Possible Approaches --- p.98Chapter 5.2.1. --- Surface Approximation --- p.98Chapter 5.2.2. --- Region Growing and Filling --- p.99Chapter 5.3. --- A New Image Interpolation Tool --- p.101Chapter 5.3.1. --- Problem analysis and requirement specifications --- p.101Chapter 5.3.2. --- The interpolation strategy --- p.103Chapter 5.3.3. --- Process overview of the image interpolation --- p.107Chapter 5.3.4. --- Data representation --- p.108Chapter 5.3.5. --- Implementation --- p.110Chapter 5.3.5.1. --- User-input - the first stage --- p.110Chapter 5.3.5.2. --- Interpolation preparation - the second stage --- p.112Chapter A. --- Boundary list rearrangement --- p.113Chapter B. --- Boundary parameterisation --- p.114Chapter C. --- Area marking --- p.114Chapter D. --- Colour pattern preparation --- p.115Chapter 5.3.5.3. --- Interpolation - the third stage --- p.118Chapter A. --- Filling curve generation --- p.119Chapter 1. --- Linear filling curve generation --- p.119Chapter 2. --- Non-linear filling curve generation --- p.120Chapter B. --- Colour pattern interpolation --- p.125Chapter 5.3.5.4. --- Post-filling - the fourth stage --- p.127Chapter 5.3.5.5. --- A working example --- p.128Chapter 5.3.6. --- Results --- p.130Chapter 5.3.7. --- Discussion --- p.140Chapter 5.3.8. --- Limitations and future improvement --- p.141Chapter 6. --- Conclusions --- p.144Appendix --- p.147References --- p.14