Selection for Robust Color Image

This work addresses the issue of color feature selection for content-based retrieval from large, heterogeneous color image databases where no assumptions can be made about the images or the type of queries. The color features used to describe an image have been developed based on the need for speed in matching and ease of computation on complex images while maintaining invariance to diierences in scale, orientation , and location of the queried object in the database images and also the presence of signiicant, interfering backgrounds. The colors present and their spatial relationships are used as features to describe a color image. These features are used in an eecient, multi-phase retrieval system to produce retrieval results fast enough for use with an online user. Test results with multi-colored query objects from man-made and natural domains highlight the capabilities of the system. Any opinions, ndings and conclusions or recommendations expressed in this material are the authors and do not necessarily reeect those of the sponsors.