Content-Based Image Retrieval System Learn And Realization

With the rapid development of the techniques of multimedia and Internet, theapplication of image is wider and winder. The CBIR has become one of the focuses ofresearch recently. Joint with techniques of image processing, pattern recognition,image comprehension and database application, it is a new potential technique in thefield of researches and applications.Based on the research of color characteristic, a new algorithm named domaincolor similarity histogram is presented; and the focus algorithm of staticallydifferential histogram on domain color similarity in this paper is presented. And aCBIR frame is designed. At the same time, a prime image retrieval system based onthis algorithm is realized using Visual C++ 6.0.The main contribution is as follows.1) To learn and analyses a few common file format, such as ".bmp" , ".jpeg"and ".gif" , etc. It has laid solid foundations for the research of CBIR.2) Based on the research of the preprocess of CBIR, especially the progresses ofimage denoised and grayed, the denoised and grayed methods of color image arepresented.3) To select the method of the HSV color space, which is suitable to the visualcharacteristic of human? by the way of discussion on color implication methods and anew domain color table for each image is established by color cluster in its colorspace,4) To put an improved method of Domain Color Histogram (DCH), DomainColor Similarity Histogram (DCSH) in order to improve the precise of imageretrieval.5) To present the algorithm of Domain Color Similarity Statistically DifferentialHistogram (DCSDH) in order to describe the domain color's special charlatanism.6) To combine DCSH with DCDSH to express the image characteristics of thecolor image.7) Finally, to analyze the retrieval result and given out an conclusion.