Robust face recognition using curvelet transform

Authors:
Gyanendra K. Verma;Shitala Prasad;Gohel Bakul
Affiliations:
Indian Institute of Information, Technology, Allahabad India;Indian Institute of Information, Technology, Allahabad India;National Institute for Mathematical Science, Daejeon, South Korea
Venue:
Proceedings of the 2011 International Conference on Communication, Computing & Security
Year:
2011

Citing 3
Cited 0

Face Recognition Based on Curvefaces

ICNC '07 Proceedings of the Third International Conference on Natural Computation - Volume 02
Face Recognition Using PCA and DCT

ICMENS '09 Proceedings of the 2009 Fifth International Conference on MEMS NANO, and Smart Systems
Face recognition by curvelet based feature extraction

ICIAR'07 Proceedings of the 4th international conference on Image Analysis and Recognition

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, we perform the face recognition using curvelet transform. In literature, multi resolution analyses of image through wavelet and Gabor transform have been quite exploited successfully for pattern recognition and so far, for face recognition. In contrast to wavelet transform, curvelet transform very efficiently approximate the curved edges with very few coefficients in addition to space-frequency information. Edges information is very important to pattern recognition task in image, hence for face recognition. Various statistical features from curvelet coefficient have used and evaluated with Indian face dataset of IIT Kanpur having 60 multi-pose different objects, using machine learning approach. The results achieved are very promising with 92.0% accuracy for 500 images. Some popular existing methods based on wavelet transform are also evaluated for comparison in this study.