摘要
A novel latent semantic indexing （LSI） approach for content-based image retrieval is presented in this paper. Firstly, an extension of non-negative matrix factorization （NMF） to supervised initialization is discussed. Then, supervised NMF is used in LSI to find the relationships between low-level features and high-level semantics. The retrieved results are compared with other approaches and a good performance is obtained.
A novel latent semantic indexing （LSI） approach for content-based image retrieval is presented in this paper. Firstly, an extension of non-negative matrix factorization （NMF） to supervised initialization is discussed. Then, supervised NMF is used in LSI to find the relationships between low-level features and high-level semantics. The retrieved results are compared with other approaches and a good performance is obtained.
基金
This work was supported by the Key Technologies R&D Program of Shanghai under Grant No. 03DZ19320.
