Bi-directional PCA with assembled matrix distance metric
Wangmeng Zuo, Kuanquan Wang, David Zhang · 2005
Principal component analysis (PCA) has been very successful in image recognition. Recent researches on PCA-based methods are mainly concentrated on two issues, feature extraction and classification. In this paper we propose bi-directional PCA (BDPCA) with assembled matrix distance (AMD) metric to simultaneously deal with these two issues. For feature extraction, we propose a BDPCA approach which can reduce the dimension of the original image matrix in both column and row directions. For classification, we present an AMD metric to calculate the distance between two feature matrices. The results of our experiments show that, BDPCA with AMD metric is very effective in image recognition.