Application of support vector machines in a texture segmentation system based on wavelet features
Date
2004
Authors
Ng, B.
Editors
Rubinov, A.
Sniedovich, M.
Sniedovich, M.
Advisors
Journal Title
Journal ISSN
Volume Title
Type:
Conference paper
Citation
Proceedings of the 6th International Conference on Optimization: Techniques and Applications, 9-11 December 2004, Ballarat, Australia.
Statement of Responsibility
Conference Name
International Conference on Optimization: Techniques and Applications (6th : 2004 : Ballarat, Australia)
Abstract
This paper investigates the application of Support Vector Machines (SVMs) to segment images based on textural information. The textures are subjected to a wavelets-based feature extraction process with the extracted features used by the SVM for classification. Both binary and multi-class cases are considered, with the latter using a one-against-one approach. Experimental results show an improvement over SVM classification using direct grayscale values as input vectors, while being more robust than alternative classifiers with the same texture features.