Content-based Image Retrieval

With the development of the technology of multimedia and internet, visual information is used more widely . As a result , the management and retrieval of image information are more important . As a key technique , content-based image retrieval(CBIR) has become one of the most active research areas in the past few years .Content-based image retrieval(CBIR) is a technique for retrieving images on the basis of image features such as color, texture and shape,etc.Key issues in CBIR include extracting features from raw images, matching query and stored images in a way that reflects human similarity judgment . The thesis researches how to derive features automatically and how to match image perceptual similarity as well as possible .The main work of the thesis includes:The thesis researches the present, the future and the application of CBIR ,analyses the CBIR system ,and researches the key techniques of CBIR which consist of all kinds of indexing techniques, similarity measure and performance evaluation .We propose a low-dimensional region-based shape index to retrieve images. The initial step in our approach is to segment images into regions on dominant colors. Image regions thus obtained after segmentation are used as input to the shape module. The index is invariant to translation, rotation and scaling. Experiment is done to demonstrate that the method is more efficient and effective than the method before modified.We also propose a new shape similarity measure which is suitable for our method . Experiment is done to demonstrate that the method is more efficient and effective than the method before modified for retrieving images .We design an actual image retrieval system for the above retrieval method .The system is simple and convenient, and supports query by example and query by scanning.