The Research Of Content-Based Image Retrieval Technique And System Implementation

Content-Based Image Retrieval (CBIR)is a kind of technique for retrieving images on the basis of automatically extracting visual features such as color, texture, and shape etc. It has been an active research area in recent years.In part one, the paper introduces the background, the application and some representative systems of CBIR.In part two, the paper describes the key techniques of CBIR in detail.In part three, after researching color feature-derived and feature matching, the paper bring forwards a new improved feature matching algorithm on the basis of histogram intersection.In part four, a scheme to extract spatial layout features is proposed, and a features matching algorithm is designed according to the spatial layout.In part five, the paper provide a novel Relevance Feedback with memory to adjust region weight in Image Retrieval, and the experimental results show that this algorithm achieves better performance .In part six, the paper depict the structure of my CBIR system , and introduce modules and functions of this system individually.In the last part, after epitomizing the CBIR technique and pointing out my work, the paper analyze the complexity of the CBIR and provide CBIR's prospect.