Machine Learning Performance on Face Expression Recognition using Filtered Backprojection in DCT-PCA Domain
Abstract
An accurate and robust transformed face descriptor that exploits the capabilities of filtered backprojection applied on Discrete Cosine Transform (DCT) and kernel Principal Component Analysis (PCA) methods is proposed. The method is invariant to rotation, variations in facial expression and illumination. Filtered backprojection constructs transform parameters from a set of projections through an image enhancing feature patterns that provide an initialization for subsequent DCT computations. DCT discards high-frequency coefficients that form least significant data to retain a subset of lower frequency coefficients visually significant in the image. The resulting coefficient features are mapped to lower dimensional space using PCA which extracts principal components that form the basis for the neural network classifier. Experiments were carried on JAFEE database and computed results compared with PCA and DCT approach. The results demonstrate significant improvements in results compared to other approaches.
Keywords: Filtered backprojection, DCT, PCA, Neural Network.
Download Full-Text
ABOUT THE AUTHORS
Ongalo Pheobe
Ongalo Pheobe. Received B.Com. Degree from Bharadhidasan government college for women in 1997 India, and Master of Computer Application(MCA) in 2001 from Pondicherry university India. Since 2003 she has been a lecturer in the department of Computer Science, Egerton University, Kenya. She is currently working towards her Ph.D. degree at the School of Information Science and Engineering, Central South University, Changsha, China. Her research interests include multimedia, neural networks, image processing and pattern recognition.
Huang Dongjun
Huang Dongjun, is a professor of Central South University and a doctoral tutor. He currently heads the Department of Computer Engineering, College of Information Science and Engineering. He obtained a master\'s degree in computer science and technology in May 1996.He completed a PHD degree from Central South University in 2004. In 2007 to 2008 he was at British University of Glasgow graphics and computer vision research group as a visiting scholar. He is committed to teaching and research development, his area of interest includes networking, distributed computing, multimedia systems and applications. Over the last 10 years he has presided over the completion of the National Natural Science Foundation of the State and the school-enterprise cooperation projects 8. He got a Provincial Science and Technology Progress Award (ranked first), made two software copyright, and has published more than 40 academic articles in the \"Journal of Software\" Electronic Journal and IEEE CVPR well-known publications and conferences, EI, 20. Taught Multimedia Technology and Application \"courses for undergraduates and graduate students. Teaching Achievement of Central South University (ranked first), Directive graduated design Outstanding Thesis for school, second prize two, has won a first prize Teaching Quality Excellence Award.
Richard Rimiru
Richard Rimiru received his B.Sc. degree from the Department of Mathematics and Computer Science, Jomo Kenyatta University of Agriculture and Technology, Nairobi, Kenya, in 1999; received his M.Sc. degree from the Department of Computer Science, National University of Science and Technology, Bulawayo, Zimbabwe, in 2002. He is currently working towards his Ph.D. degree at the School of Information Science and Engineering, Central South University, Changsha, China. His main research areas are artificial intelligence especially in bio-inspired computing, pattern recognition and image processing.
Ongalo Pheobe
Ongalo Pheobe. Received B.Com. Degree from Bharadhidasan government college for women in 1997 India, and Master of Computer Application(MCA) in 2001 from Pondicherry university India. Since 2003 she has been a lecturer in the department of Computer Science, Egerton University, Kenya. She is currently working towards her Ph.D. degree at the School of Information Science and Engineering, Central South University, Changsha, China. Her research interests include multimedia, neural networks, image processing and pattern recognition.
Huang Dongjun
Huang Dongjun, is a professor of Central South University and a doctoral tutor. He currently heads the Department of Computer Engineering, College of Information Science and Engineering. He obtained a master\'s degree in computer science and technology in May 1996.He completed a PHD degree from Central South University in 2004. In 2007 to 2008 he was at British University of Glasgow graphics and computer vision research group as a visiting scholar. He is committed to teaching and research development, his area of interest includes networking, distributed computing, multimedia systems and applications. Over the last 10 years he has presided over the completion of the National Natural Science Foundation of the State and the school-enterprise cooperation projects 8. He got a Provincial Science and Technology Progress Award (ranked first), made two software copyright, and has published more than 40 academic articles in the \"Journal of Software\" Electronic Journal and IEEE CVPR well-known publications and conferences, EI, 20. Taught Multimedia Technology and Application \"courses for undergraduates and graduate students. Teaching Achievement of Central South University (ranked first), Directive graduated design Outstanding Thesis for school, second prize two, has won a first prize Teaching Quality Excellence Award.
Richard Rimiru
Richard Rimiru received his B.Sc. degree from the Department of Mathematics and Computer Science, Jomo Kenyatta University of Agriculture and Technology, Nairobi, Kenya, in 1999; received his M.Sc. degree from the Department of Computer Science, National University of Science and Technology, Bulawayo, Zimbabwe, in 2002. He is currently working towards his Ph.D. degree at the School of Information Science and Engineering, Central South University, Changsha, China. His main research areas are artificial intelligence especially in bio-inspired computing, pattern recognition and image processing.
