casa 16(9): e2

Research Article

Face recognition based on LDA in manifold subspace

Download233 downloads
  • @ARTICLE{10.4108/eai.2-5-2016.151209,
        author={Hung Phuoc Truong and Tue-Minh Dinh Vo and Thai Hoang Le},
        title={Face recognition based on LDA in manifold subspace},
        journal={EAI Endorsed Transactions on Context-aware Systems and Applications},
        volume={3},
        number={9},
        publisher={EAI},
        journal_a={CASA},
        year={2016},
        month={5},
        keywords={face recognition, manifold learning, semi-supervised, discriminative},
        doi={10.4108/eai.2-5-2016.151209}
    }
    
  • Hung Phuoc Truong
    Tue-Minh Dinh Vo
    Thai Hoang Le
    Year: 2016
    Face recognition based on LDA in manifold subspace
    CASA
    EAI
    DOI: 10.4108/eai.2-5-2016.151209
Hung Phuoc Truong, Tue-Minh Dinh Vo, Thai Hoang Le1,*
  • 1: Faculty of Information Technology, University of Science – Vietnam National University Ho Chi Minh city, Vietnam
*Contact email: lhthai@fit.hcmus.edu.vn

Abstract

Although LDA has many successes in dimensionality reduction and data separation, it also has disadvantages, especially the small sample size problem in training data because the "within-class scatter" matrix may not be accurately estimated. Moreover, this algorithm can only operate correctly with labeled data in supervised learning. In practice, data collection is very huge and labeling data requires high-cost, thus the combination of a part of labeled data and unlabeled data for this algorithm in Manifold subspace is a novelty research. This paper reports a study that propose a semi-supervised method called DSLM, which aims at overcoming all these limitations. The proposed method ensures that the discriminative information of labeled data and the intrinsic geometric structure of data are mapped to new optimal subspace. Results are obtained from the experiments and compared to several related methods showing the effectiveness of our proposed method.