Face Recognition using Curvelet and ICA.
Kishor S. Kinage, Sunil G. Bhirud · IPCV · 2010
In recent years, many different image features have been used for face recognition. Wavelet Transform is a popular multiresolution analysis tool in image processing and computer vision. Recent researches on multi-scale analysis, especially the curvelet research, provide good opportunity to extract face image features. The transform was designed to represent edges and other singularities along curves much more efficiently than traditional transforms. In this paper we extract image features of facial images from curvelet transform. We then transformed this feature vector into the basis space of PCA and ICA and compared the performance using Euclidean distance measure as classifier. The results show maximum accuracy of 87.50% and 85.00% for curvelet-ICA and curvelet-PCA respectively. Whereas accuracy using PCA and ICA alone was 82.50 and 73% respectively.