An Image Retrieval Algorithm Based on Local Color Features
Bai Qing-hai · Neimenggu Shi-da xuebao. Zhexue shehui kexue hanwen ban · 2008
An image retrieval algorithm based on local color features is proposed.Firstly,we split the image into sub-blocks and extract dominant colors in each sub-block.Secondly,according to the importance of the image center,the different feature weights are adopted in the dominant colors in each sub-block to enhance the feature vector of local image and improve the image retrieval accuracy.The experiments show that,compared with traditional global histogram method,the proposed algorithm has better retrieval results.