Abstract | ||
---|---|---|
A novel improved linear discriminant analysis (ILDA) method is presented. Comparing with LDA, under the condition of d < c -1, d and c are the dimensionality of feature subspace and the number of classes respectively, ILDA uniformly preserves the class distances of classpairs by rearranging the contribution of each class-pair to the generalized between-class scatter matrix after whitening within-class scatter matrix. Experiment results based on simulating data and measured radar data both show that, under the condition of d < c -1, the features extracted by ILDA are more efficient for multi-class classification than those extracted by LDA. |
Year | DOI | Venue |
---|---|---|
2008 | 10.1109/FSKD.2008.182 | FSKD (2) |
Keywords | Field | DocType |
statistical analysis,matrix algebra,target recognition,radar target recognition,linear discriminant analysis,improved lda method,radar resolution,scatter matrix,high resolution radar automatic,high resolution radar automatic target recognition,linear discrimnant analysis,high resolution,feature extraction,testing,multi class classification,support vector machines,automatic target recognition,scattering matrix,training data,matrix decomposition | Radar,Automatic target recognition,Pattern recognition,Subspace topology,Computer science,Support vector machine,Matrix decomposition,Feature extraction,Artificial intelligence,Linear discriminant analysis,Scatter matrix,Machine learning | Conference |
Volume | ISBN | Citations |
2 | 978-0-7695-3305-6 | 1 |
PageRank | References | Authors |
0.35 | 2 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Jing Liu | 1 | 1 | 0.69 |
Junying Zhang | 2 | 86 | 7.59 |
Feng Zhao | 3 | 7 | 2.13 |