Transform Texture Classification.
WOODS HOLE OCEANOGRAPHIC INSTITUTION MA
Pagination or Media Count:
This thesis addresses the three major components of a texture classification system texture image transform, feature extractionselection, and classification. For the first component, a unique investigation of texture analysis, drawing on an extensive survey of existing approaches, defines the interrelations among 11 types of texture analysis methods. A novel unification of the different methods defines a framework of transformation and representation in which three major classes of transform matrices capture texture information of increasing coherence length the spatial domain method co-occurrence, the micro-structural method run-length, and the frequency multichannel method Fourier spectrum. For the second system component, we apply the Karhunen-Loeve Transform KLT directly to the transform matrix to extract a vector of dominant features, optimally preserving texture information in the matrix. This approach is made possible by the introduction of a novel Multi-level Dominant Eigenvector Estimation MDEE algorithm, which reduces the computational complexity of the standard KLT by several orders of magnitude. Experimental results of applying the new algorithm to the three transform matrix classes show a strong increase in performance. Using the same MDEE algorithm, the three extracted feature vectors are then combined into a more complete description of texture images. The same approach is also used for a study of object recognition, where the combined vector also include granulometric, object-boundary, and moment-invariant features. The plankton object recognition experiments use a Learning Vector Quantization LVQ neural-net classifier to achieve superior performance on the highly non-uniform plankton database. By introducing a new parallel LVQ learning scheme, the speed of network training is dramatically increased.