Analysis of Recognition Accuracy Using Curvelet Tranform

This paper describes a comparative analysis of recognition accuracy using feature extraction algorithm. A feature extraction algorithm is introduced for face recognition, Principle Component Analysis (PCA),Linear Discriminant Analysis(LDA) , Independent Component Analysis(ICA) and Nonnegative matrix factorization (NMF) based on curvelet transform. Mostly recognition system is capable to perform three subtasks face detection, feature extraction and classification. Digital curvelet transform is an even better method due to its directional properties, Curvelet transform is multiresolution analysis method to improve directional elements with anisotropy and better ability to represent edges and other singularities along the curves. This paper aims to compare different face recognition techniques based on curvelet transform for improving the performance of recognition accuracy. All these algorithms are based on ORL database using MATLAB. The achievability of these algorithms for human face identification is presented through new study. Face recognition algorithms are used in security control, forensic application, and access control at automatic teller machines, passport verification etc. Keywords— Face recognition, Feature Extraction, PCA, LDA, ICA, NMF, Curvelet Transform.