Study on a rough set approach to semantic image retrieval
Qingmin Cui, Wangao Li · 2010 3rd IEEE International Conference on Broadband Network and Multimedia Technology (IC-BNMT) · 2010
In this paper, semantic gap is a challenging issue in image retrieval. Firstly, in the process of constructing vector space model, the theory of Latent Semantic Indexing is introduced to mine the semantic information of images, and then, rough set theory is applied to retrieve and match the semantic feature of image database in the approximate space of tolerance rough set. Lastly, semantic image classification algorithm is implemented. Experimental results show that the performance of the classification is greatly improved.