Multi-modal CBIR Algorithm Based on Latent Semantic Indexing

The paper presents a new multiple feature fusion (MFF) based on latent semantic indexing (LSI) method to achieve an improved image retrieval performance. The proposed method extracts different physical features, which come from not the whole image but its main objects, and constructs a multi-modal semantic space, each dimension of which represents a different feature component of the image. Furthermore, semantic relevance feedback information from the users is also integrated to improve the feedback performance of the system. The experimental results demonstrate the good robustness of LSI-MFF and have shown that this method is especially suitable for mass image database such as web environment.