Published Paper Details:

The combination of these content-based features is needed for better retrieval of image according to the application. The main test of CBIR framework is to decide the specific/surmised matching image of database to the query image. Over the most recent two decades, CBIR frameworks have been enhanced a lot. In any case, there still stay a few issues which have not been addressed agreeably. Most importantly issue is of semantic gap, which exist between low level feature portrayal of images and the genuine visual view of the image. The similarity between two images (spoken to by their feature esteems) will characterize by a similarity measure. Choice of similarity metrics will have coordinate effect on the execution of content-based image retrieval. A CBIR using multistage retrieval strategy is proposed here to deal with these issues. Low level features of an image can be local and global. In this paper the A computer system having Pentium IV, 2.8 GHz processor and 1 GB RAM is used for conducting experiments. This system has been actualized in MATLAB.