The Research And Implementation Of Color-Based Image Retrieval System

A large number of images come forth because of the prevalence of multimedia technology and the implementation of Internet technology. Since traditional text keyword based retrieval approach can't meet the demand of image data retrieval, so content-based image retrieval (CBIR) technology becomes the current research focus.The introductions of the application background of CBIR, as well as existing typical CBIR system are put in the first place. Then a discussion on key technologies used in CBIR field is given out.In current applications, the content-based image retrieval systems usually adopt the algorithms which are originally designed for the specific circumstances. According to that, the paper expatiates on key technologies used in CBIR researches, such as feature extracting, similarity measuring, and relevance feedback, etc. Afterwards, the application of perceptual grouping in object recognition, which is a principle in psychology, is discussed in this paper and the results of experimentation are given.Since color feature is relatively robust to background complication and independent of image size and orientation, it is widely used in image feature representations. In this paper, the image retrieval system adopts color feature as its retrieval method. Based on the research of image color feature extracting and similarity measure algorithm, an improved method for color features representation and similarity measuring is proposed. The results demonstrate that the improved approach can obtain better retrieval results than others.In addition, an image retrieval system based on color feature is proposed. The system supports various querying properties, such as querying by example, querying by draw, and querying by keywords, etc. Furthermore, users can intervene in the search process via relevancy feedback module. Main parts of the proposed system are described, including the system architecture, the system modules, and user interfaces. Also, some relevant key techniques in implementation are discussed in depth.Finally, the paper makes a summary of the current situation of the research in CBIR, and provides the CBIR's prospect.