Face Recognition using Deep Neural Network Across Variationsin Pose and Illumination
S. Meenakshi1, M. Siva Jothi2, D. Murugan3
1S. Meenakshi, Research Scholar, Department of Computer Science & Engineering, Manonmaniam Sundaranar University, Abishekapatti, Tirunelveli (Tamil Nadu), India.
2Dr. M. Siva Jothi, Associate Professor, Department of Computer Science, Sri Parasakthi College for Women, Courtallam (Tamil Nadu), India.
3Dr. D. Murugan, Professor and Head, Department of Computer Science & Engineering, Manonmaniam Sundaranar University, Abishekapatti, Tirunelveli (Tamil Nadu), India.
Manuscript received on 03 June 2019 | Revised Manuscript received on 28 June 2019 | Manuscript Published on 04 July 2019 | PP: 289-292 | Volume-8 Issue-1S4 June 2019 | Retrieval Number: A10500681S419/2019©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Face recognition is an active area of research in computer vision. It is one of the fastest growing biometric systems over the past years. Several research efforts have been carried out to recognize the face images using different techniques ranging from appearance-basedmethod, feature based method and hybrid method with different results. However, face recognition is a challenging task due to the facial expressions, occlusions, variations in pose and illumination variation etc. To handle the pose and illumination variation in face images, this paper developed an architecture for face recognition using deep neural network. Convolutional neural network is trained to recognize the face images. The developed method is tested on ORL database by varying feature maps to find the best architecture. Results showed that the proposed method with architecture 15-90-150 provide better results compared to the state of art methods. It is also proved that the proposed model is robust to pose and illumination variation.
Keywords: Biometric Authentication, Convolutional Neural Network, Face Recognition.
Scope of the Article: Pattern Recognition