The Taub Faculty of Computer Science Events and Talks

Texture features have always been a key attribute in image recognition and classification. In this work we propose a pre-processing stage for enhancing the performance of widely used color texture recognition methods. One approach we investigated, Decorrelation Stretching, was employed historically for enhancing the interpretability of multi-channel satellite images. This is achieved by stretching the dynamic range of color data over its principal components. Another approach decomposes the image into cartoon-like and texture components and then re-combines them while amplifying the texture component. Conducting experiments on the VisTex texture image database we show that extracting auto- and cross-correlation features from images that went through the proposed pre-processing stages increases the classification accuracy significantly. Similar results were achieved when using wavelet correlation signature as texture features. Classification results will be presented and discussed. Our conclusion is that the proposed approach could be instrumental in texture recognition tasks.