Key Technology Research For Region-based Image Retrieval

The continuing rapid growth in size of digital image databases leads to the increasing need for retrieval systems to satisfy more complex and sophisticated queries. Content-based image retrieval (CBIR) systems attempt to solve this problem by retrieval system based on text mode, become one of most active researches in computer vision, image and video processing, Data Mining and so on. Currently the researchers have already designed many content based image retrieval systems. While these researches establish the basis of CBIR, the retrieval performance is still far from users'expectations. The main reason is acknowledged to be the gap between low-level features and high-level concepts. Users accessing a CBIR system often look for images containing particular"objects", possibly arranged in a specific spatial organization. Therefore we need to develop the region-based image retrieval system (RBIR). Region-based image retrieval systems aim to improving the performance of content-based search by segmentation each image into a set of"homogeneous"regions. Thus, similarity between images is assessed by computing similarity between pairs of regions and then combining the results at the image level. Region-based image retrieval systems attempt to overcome the drawback of global features by representing images at object-level, which is intended to be close to the perception of human visual system. This paper mainly studies the key technology in the region-based image retrieval about image segmentation, region matching and relevance feedback.Image segmentation is the key step in region-base image retrieval. To transform the original image into more abstract and compact form through image segmentation, will make analyzing image and understanding image in higher level become possible. Region-base image retrieval hope use image segmentation technology to pick-up the object in the image to describe image feature in object level, and make retrieval more fit the semantic of image. The effect of segmentation has a direct influence in the effect of retrieval and feature extraction. This paper describes two segmentation algorithms. The first algorithm based on color and spatial information is automatic...