Abstract:

The ultimate reason of low retrieval rate of contourlet transform texture image retrieval systems lies in the coefficients oscillating and shift sensitive characters. In order to overcome the defects of contourlet transforms, a mapping based contourlet transform is proposed, and a texture image retrieval system based on the new transform is proposed in which the feature vectors are formed by cascading the energy and standard deviation of each subband in the mapping based contourlet domain, and a Canberra distance is used as similarity metric. Experimental results show that the image retrieval system is superior to that of the original contourlet transform with almost same length of feature vectors, retrieval time and memory.