An Algorithm of Texture Classification Based on Feature Extraction and BP Neural Network

Abstract

In this paper, we apply the texture conception of natural language to the texture classiflcation, and classify the natural texture into ten classes. Basing on the above, we found a small image library of natural texture. In the thesis, we discuss the common means of texture feature extraction, and bring forward a speciflc algorithm for Gabor fllter. In order to validity of the feature extraction, we adopt the BP network as the classifler to carry out our experiments, which bring us satisfying results.