Abstract | ||
---|---|---|
To further improve the effectiveness and reliability of sparse representation classification, a multiple sparse representation classification based on weighted residuals is proposed in this paper. To overcome the deficiency of single feature identification of SRC (sparse representation classification), we propose extracting more features to represent samples. To enhance the performance of the conventional method SRC, we propose using the normalized weighted l2-norm of sparse representation coefficients. Experiments show that the effectiveness of our proposed method WR_MSRC (multiple sparse representation classification approach based on weighted residuals) can be improved considerably. |
Year | DOI | Venue |
---|---|---|
2013 | 10.1109/ICNC.2013.6818121 | ICNC |
Keywords | Field | DocType |
normalized weighted l2-norm,src,pattern classification,weighted residuals,feature extraction,sparse representation coefficients,sparse representation classification,wr_msrc method,sparse representation,databases,mathematical model,face recognition,classification algorithms,face | Mathematical optimization,Normalization (statistics),Pattern recognition,Computer science,Sparse approximation,Artificial intelligence,Machine learning | Conference |
Citations | PageRank | References |
0 | 0.34 | 3 |
Authors | ||
4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Ganglong Duan | 1 | 9 | 2.35 |
Ni Li | 2 | 0 | 0.34 |
Zhishi Wang | 3 | 11 | 2.33 |
Jianan Huangfu | 4 | 0 | 0.34 |