A WAVELET TRANSFORMATION-BASED MULTICHANN EL NEURAL-NETWORK METHOD FOR TEXTURE SEGMENTATION

A neural network texture segmentation method, in which multichannel fi ltering is embodied, was proposed. Multichannel filtering technology i s a very effective method for texture segmentation. Instead of using a general filter bank, the texture feature extraction and classificatio n tasks were performed in this paper by the same unified neural networ k. Decision-based neural network was adopted to improve the accuracy o f classification. Wavelet transformation of texture was used to decrea se the correlation of texture data and increase the efficiency of netw orks learning. Experiments show that the proposed method achieves lowe r error rates than other methods and a satisfactory result is obtained .