A Region-based Image Similarity Computation Approach
Guoqiang Han · Journal of Guangxi Normal University · 2008
Computation of image similarities is one of the key technologies for good retrieval results in Content-Based Image Retrieval(CBIR) system.In this paper,an approach of Region-based Image Similarity Computation is proposed.Firstly,the similarity of two regions with the combination features(color,texture and position),histogram and region shape features,is computed respectively.Then,the similarity of two regions is computed by the weighted product method.At last,the average of the max similarity of all regions is regarded as the similarity of two images.The proposed approach and implementation system are tested in the Corel natural image database,and the results show that the retrieval performance is greatly improved.