Non-negative matrix factorization with L0 sparseness constraints and its applications to face recognition

作者：Han, Dong; Gai, Shan^*

来源：Journal of Intelligent and Fuzzy Systems, 2018, 35(1): 729-737.

DOI：10.3233/JIFS-171116

• Sparseness
• NMF
• L-0 norm
• sparse coding
• feature extraction

In order to solve the problem of unstable sparseness of non-negative matrix factorization (NMF), the improved NMF algorithms with L-0 sparseness constraints are proposed. With the constraining the L-0 norm of the coefficient matrix, we applied inverse matching principle into non-negative least square (ISNNLS) which enhances the reconstruction ability of the decomposition matrix. In addition, the L-0 sparseness constraints are added to the basis matrix. In the updating process, the proposed algorithm set the smallest value to zero by projecting the basis vectors onto the closest non-negative vector with the expected sparseness. The experimental results have illustrated that the proposed algorithm can achieve higher reconstruction quality and effectiveness compared with the other algorithms.

• 出版日期2018
• 单位南昌航空大学