Content-based Image Retrieval

While the resources of multimedia such as images becomes more and more aboundant, demand for effectively managing the huge amount of image data becomes more and more urgent, because traditional retrieval methods based on keyword and description text are not competent for it. So content based image retrieval techniques have emerged as the times require.Content-based Image Retrieval(CBIR) is a technique for retrieving images on the basis of image features such as color,shape and color,etc.And the key issues in CBIR include extracting features form raw images,matching query and stored images in a way that reflects human similarity judgement.In this paper,the key technology of CBIR is studied.And the main aspects dealt with this are structured as follows:1. Some methods of extracting image content are analyzed detailedly in this paper. An actual CBIR system is designed aiming at concrete image database.2. Preprocess of image before CBIR. Mainly introducing the removal of noise, color image segmentation. And establishing the main object region of the image.3. Basing on the Human Vision Property, a retrieval algorithm is presented on L*a*b*, which combines color and aberration distance statistic between image pixels.4. With the confirm of the object region in image, a color image retrieval algorithm based on the main object region of the image is presented.5. After analysing the methods of shape retrieval, the property of image information entropy and shape moment is experimented. Finally, this paper presents Various similar formula for various retrieval methods. The retrieval result can be valued by quantification and precision.