Japanese / English

Detail of Publication

Text Language Japanese
Authors Masakazu Iwamura, Shinichiro Omachi,and Hirotomo Aso
Title Estimation of True Mahalanobis Distance from Eigenvectors of Sample Covariance Matrix
Journal Trans. IEICE
Vol. J86-D-II
No. 1
Pages pp.22-31
Reviewed or not Reviewed
Month & Year January 2003
Abstract In statistical pattern recognition, parameters of distributions are usually estimated from training sample vectors. However, estimated parameters contain estimation errors and the errors cause bad influence on recognition performance when the sample size is not sufficient. In this paper, we present a method to estimate the true Mahalanobis distance from sample eigenvectors (eigenvectors of sample covariance matrix) by considering the errors of eigenvectors. Recognition experiments show that by applying the proposed method, the true Mahalanobis distance can be estimated even if the sample size is small, and better recognition accuracy is achieved.
URL http://search.ieice.org/bin/summary.php?id=j86-d2_1_22&category=D&year=2003&lang=J&abst=
Back to list