Comparing images using color coherence vectors
MULTIMEDIA '96 Proceedings of the fourth ACM international conference on Multimedia
Keyblock: an approach for content-based image retrieval
MULTIMEDIA '00 Proceedings of the eighth ACM international conference on Multimedia
Hi-index | 0.00 |
In this paper, a new image retrieving method is proposed, which integrates color features, and integrates color histogram, relevance feedback, the partition color and the genetic algorithms. First, by analyzing images' color histogram, 'example images' which have the same global color feature are founded out. Then, by using relevance feedback for these images, images which user thought are similar to the example image are found out and then classify these images by spatial position. After that, genetic algorithm is applied to adjust automatically retrieval parameters, reduce the user's selecting operations in a feedback process, and improve the intelligent degree of the system. By comparative experiments with other two kinds of methods from search efficiency and feed back time, the global accumulated histograms and the partitions and relativity feed backing, this method is proved efficient.