Image retrieval based on weighted blocks and color feature

Study the global histogram algorithm and the rigid block algorithm in image retrieval, in order to improve the lack of spatial information in the global histogram,and to improve the lack of rotation invariance and image integrity in the rigid block, this paper proposed block weighting the color feature algorithm. The algorithm first conduct color image a rigid segmentation, then sort the second moment, again add different weights in different regions. Through the image retrieval results show, the algorithm can obtain results superior to the traditional search algorithm.