摘要

This paper proposes a novel algorithm for image feature extraction and dimension reduction, namely, the bilateral two-dimensional locality preserving projections (B2DLPP). Different from the traditional LPP based approaches, B2DLPP is based on 2D image matrices rather than column vectors so the image matrix does not need to be transformed into a long vector before feature extraction. The advantage arising in this way is that 2D image matrices can be effectively compressed from horizontal and vertical directions and uses F-norm classification measure. It is applied to face recognition where only few training images exist for each subject. Extensive experimental results show that the extraction of image features is computationally more efficient using B2DLPP than traditional LPP on Yale face database B.

全文