Abstract
More and more digital images can be obtained by users from the world-wide-web. From the large number of images, it is very important for users to retrieve desired images via the efficient and effective mechanisms. In this paper, we proposed two efficient approaches to facilitate image retrieval by using a simple way to represent the content of images. Each image is partitioned into m×n equal-size sub-images (or blocks). A color that has enough number of pixels in a block is extracted to represent the content of this block. In the first approach, the content of an image is represented by these extracted colors of the blocks directly. The spatial information between images is considered in image retrieval. In the second approach, the colors of the blocks in an image are used to extract objects (or regions). A block-level process is proposed to perform the region extraction. The spatial information between regions is considered unimportant in the similarity measurement. Our experiments show that the block-based information used in these two approaches can speed up the image retrieval. Moreover, the two approaches are effective on different requirements of image similarity. Users can choose a proper approach to process their queries based on their similarity requirements.