Based Color Image Retrieval System

With the rapid development of Internet technology, image is widely used. It is an urgent task that how to search the image effectively and quickly. On the basis of widely referring to the material about Content Based Image Retrieval (CBIR) of home and abroad, matching algorithms of image similarity and the condition of image retrieval techniques based on the content was analyzed and discussed in this thesis.Generally speaking, current image database consists of color pictures, so in CBIR system, color is one of most useful characteristics.This thesis have discussed three key questions on how to make use of the characteristic of color, which include expressing color space, obtaining the characteristic of color and measuring the likeness based on color.Firstly, the thesis offers the relevant image preprocess methods to extract the image feature accurately and easily. Secondly, based on the shortage of full-image histogram, this thesis presents block histogram algorithm. Thirdly, we study the image retrieval algorithm of color pair, and make the retrieval image having color similarity in space. Finally, we apply HSV color-space corresponding to human visual characteristics to image retrieval, make use of human feeling on color to quantify the color weight, and analyze the two quantization methods.An experimental image retrieval system, which is based on the image color characteristic, is implemented with Visual C++ 6.0. And some retrieval results are analyzed. The works lay better foundation for integrating every kind of retrieval algorithms using color histograms and other image retrieval ways.