Dimensionality Reduction via Locality Preserving Reconstructive Discriminant Analysis

Chen Yi


 In this paper, a new dimension reduction technique called locality preserving reconstructive discriminant analysis (LPRDA) is proposed. LPRDA aims to find the projection axes onto which structure information is preserved, meanwhile in the low dimensional subspace, the intra-class reconstruction is minimized and the inter-class reconstruction error is maximized. Then LPRDA is applied to solve face recognition problem on the ORL, extended YALE-B face databases and the finger knuckle recognition problem on the PolyU finger knuckle print database. The experimental results indicate LPRDA has better performance than others.

Full Text:



  • There are currently no refbacks.

Creative Commons License
This work is licensed under a Creative Commons Attribution 3.0 License.

JES ©: World Science Publisher United States