A performance perspective: Content based image retrieval system

Abstract

The content-based image retrieval (CBIR) plays a significant aspect in digital image processing. The CBIR is novel and a high-speed process in information recovery. In order to sequentially access the huge set of database the oldest incisive search engines like google, yahoo, bingo was based on textual explanation of images. Hence, the performance of this system was not adequate. Therefore, a new process is required, which could be user friendly in order to access the image data. This paper used different set of various attributes like texture, shape and color derived from the image extracted from query and the training images for comparing and retrieving the image. Thus, it repeatedly produces descriptions directly from the media data by using digital image processing methods. In this paper we analysis some of the procedural aspects of CBIR method and even specify its advantages and disadvantages.