Title: Self-adaptive local Fisher discriminant analysis for semi-supervised image recognition

Authors: Zhonghua Liu; Jingyan Wang; Jiaju Man; Yongping Li; Xinge You; Chao Wang

Addresses: Shanghai Institute of Applied Physics, Chinese Academy of Science, 2019 Jialuo Road, Jiading District, Shanghai 201800, China ' Shanghai Institute of Applied Physics, Chinese Academy of Science, 2019 Jialuo Road, Jiading District, Shanghai 201800, China; HPCSIP Key Laboratory, Ministry of Education, College of Mathematics and Computer Science, Hunan Normal University, Changsha 410081, Hunan, China ' HPCSIP Key Laboratory, Ministry of Education, College of Mathematics and Computer Science, Hunan Normal University, Changsha 410081, Hunan, China ' Shanghai Institute of Applied Physics, Chinese Academy of Science, 2019 Jialuo Road, Jiading District, Shanghai 201800, China ' Department of Electronics and Information Engineering, Huazhong University of Science and Technology, Wuhan, Hubei, China ' OGI School of Science and Engineering, Oregon Health and Science University (OHSU), Beaverton, Oregon, 97006, USA

Abstract: In this paper, we present a self-adaptive semi-supervised dimension reduction framework for image class recognition. Compared with the Semi-supervised Local Fisher Discriminant Analysis (SELF), whose classification performance is significantly influenced by some parameters - Neighbour size k of every labelled sample, Affinity Matrix A = {A_ij} and Trade-off parameter β, our sparse algorithm is developed based on the distribution of the data set, which is much more adaptive to data set themselves. To develop a more tractable and practical approach, we in particular impose neighourhood structure constraint on the labelled samples in the minimum reconstruction criterion and develop a quadratic optimisation technique to approximately estimate the affine matrix used in the Local Fisher Discriminant Analysis (LFDA). We also give a novel approach to estimate the β automatically. Our experiments on semi-supervised face recognition task demonstrate that the proposed method is more robust and efficient in dealing with the semi-supervised problems in face recognition when compared with the related SELF methods

Keywords: semi-supervised learning; image classification; self-adaptive LFDA; local Fisher discriminant analysis; hypergraphs; quadratic programming; image recognition; face recognition; biometrics.

DOI: 10.1504/IJBM.2012.049731

International Journal of Biometrics, 2012 Vol.4 No.4, pp.338 - 356

Received: 22 Aug 2011
Accepted: 04 Oct 2011
Published online: 29 Nov 2014 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article