Linear Dimensionality Reduction via a Heteroscedastic Extension of LDA: The Chernoff Criterion
IEEE Transactions on Pattern Analysis and Machine Intelligence
A new covariance estimate for Bayesian classifiers in biometric recognition
IEEE Transactions on Circuits and Systems for Video Technology
Face recognition using a kernel fractional-step discriminant analysis algorithm
Pattern Recognition
Exploring the boundary region of tolerance rough sets for feature selection
Pattern Recognition
A rank-one update algorithm for fast solving kernel Foley-Sammon optimal discriminant vectors
IEEE Transactions on Neural Networks
Face recognition using heteroscedastic weighted kernel discriminant analysis
ICAPR'05 Proceedings of the Third international conference on Pattern Recognition and Image Analysis - Volume Part II
Extending kernel fisher discriminant analysis with the weighted pairwise chernoff criterion
ECCV'06 Proceedings of the 9th European conference on Computer Vision - Volume Part IV
Using Clustering and Metric Learning to Improve Science Return of Remote Sensed Imagery
ACM Transactions on Intelligent Systems and Technology (TIST)
Hi-index | 0.01 |
We propose an uncorrelated heteroscedastic LDA (UHLDA) technique, which extends the uncorrelated LDA (ULDA) technique by integrating the weighted pairwise Chernoff criterion. The UHLDA can extract discriminatory information present in both the differences between per class means and the differences between per class covariance matrices. Meanwhile, the extracted feature components are statistically uncorrelated the maximum number of which exceeds the limitation of the ULDA. Experimental results demonstrate the promising performance of our proposed technique compared with the ULDA.