Abstract
Use of a single technique for the extraction of diverse features in a texture image usually shows limited capabilities for texture description. Texture features extracted using different techniques can be merged in an attempt to enhance their texture description capability. This paper explores the fusion of optimized moment and Gabor energy texture features. The Fisher linear discriminant analysis is used to show that the discrimination effectiveness of the features increases as a result of the proposed fusion. Preliminary simulation results have been validated experimentally through the classification and segmentation of real texture images.