Article: Improving face recognition using deep autoencoders and feature fusion Journal: International Journal of Biometrics (IJBM) 2023 Vol.15 No.1 pp.40 - 58 Abstract: Uncontrolled environments are the main challenges of real face recognition systems, recent success of deep learning and features fusion has led to various performance improvements. This paper proposes a novel scheme called feature autoencoder (FAE), where an autoencoder model is not trained directly from the raw facial images, rather it uses a fusion of features constructed by Gabor filter, local binary pattern and local phase quantisation. For each feature, a linear discriminant analysis is applied to reduce its high dimensionality and a limited adaptive histogram equalisation process is employed for contrast enhancement. The proposed scheme has been evaluated using known datasets such as AR, ORL and YALE, and the experimental results carried out on these databases have been compared using three classifiers: k-nearest neighbour, multiclass support vector machine and softmax classifier, demonstrating the effectiveness of proposed approach and parameters. The experimental results obtained and compared with recent and similar approaches on six databases: ORL, YALE, AR, extended YALE B, CMU PIE, and LFWcrop, suggest that the proposed technique outperforms similar techniques. The recognition rates got from them are 100%, 100%, 99.66%, 99.40%, 97.31%, and 90.68% respectively. Inderscience Publishers - linking academia, business and industry through research

You can view the full text of this article for free using the link below.

Title: Improving face recognition using deep autoencoders and feature fusion

Authors: Ali Khider; Rafik Djemili; Ahmed Bouridane; Richard Jiang

Addresses: PIMIS Lab, Department of Electronics and Telecommunications, Université 8 Mai 1945 Guelma, B.P. 401, Guelma, Algeria ' PIMIS Lab, Department of Electronics and Telecommunications, Université 8 Mai 1945 Guelma, B.P. 401, Guelma, Algeria ' Centre for Data Analytics and Cybersecurity, University of Sharjah, Sharjah, UAE ' School of Computing and Communication, Lancaster University, Lancaster, UK

Abstract: Uncontrolled environments are the main challenges of real face recognition systems, recent success of deep learning and features fusion has led to various performance improvements. This paper proposes a novel scheme called feature autoencoder (FAE), where an autoencoder model is not trained directly from the raw facial images, rather it uses a fusion of features constructed by Gabor filter, local binary pattern and local phase quantisation. For each feature, a linear discriminant analysis is applied to reduce its high dimensionality and a limited adaptive histogram equalisation process is employed for contrast enhancement. The proposed scheme has been evaluated using known datasets such as AR, ORL and YALE, and the experimental results carried out on these databases have been compared using three classifiers: k-nearest neighbour, multiclass support vector machine and softmax classifier, demonstrating the effectiveness of proposed approach and parameters. The experimental results obtained and compared with recent and similar approaches on six databases: ORL, YALE, AR, extended YALE B, CMU PIE, and LFWcrop, suggest that the proposed technique outperforms similar techniques. The recognition rates got from them are 100%, 100%, 99.66%, 99.40%, 97.31%, and 90.68% respectively.

Keywords: uncontrolled environments; face recognition; deep learning; sparse; autoencoder; feature extraction; fusion.

DOI: 10.1504/IJBM.2023.127730

International Journal of Biometrics, 2023 Vol.15 No.1, pp.40 - 58

Received: 08 Jun 2021
Accepted: 15 Sep 2021
Published online: 15 Dec 2022 *

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

Title: Improving face recognition using deep autoencoders and feature fusion

Keep up-to-date