Research And Implementation Of Content Based Image Retrieval System

Along with the development of computer and communication technology, digital image processing is widely used in many fields. Traditional information management and retrieval methods are no more appropriated for large image databases. In order to manage and retrieve large amount of images, the CBIR (Content-Based Image Retrieval) has emerged to be one of the hot research areas in image domain.CBIR is on the basis of images' content itself, which is extracted by computers, such as color, texture and shape. The system retrieves the feature vectors consisting of the above information from the large image database to satisfy the users' requirement.This dissertation deals with the content-based image retrieval system architecture and some key techniques are introduced. The key research issues of this dissertation are shape feature extraction and relevance feedback.The shape feature is a basis feature of the image, and the shape extraction is generally based on the image segmentation. In order to solve the problem in the edge detection based image segmentation, an effective closed contour extraction method based on edge linking and large region searching is proposed. The new method is applied in the image retrieval system to generate the image shape feature. The result of the experiment shows the effectiveness of the algorithm.Because of the weak correlation low level features and high level conception of images and different subjective perception of the system users, relevance feedback has been introduced to CBIR. And the system learns the requirements of different users from the feedback procedures. According to the research of hierarchy of image retrieval model and the conception of feedback, a new rough set based re-weighting feedback algorithm is designed.At last, a prototype image retrieval system DUTIRS is constructed, which realizes the content-based image retrieval and feedback image retrieval.