Canonical Correlated Kernel PCA Method for Face Recognition

Journal Title: International Journal on Computer Science and Engineering - Year 2012, Vol 4, Issue 8

Abstract

Practical face recognition systems are sometimes confronted with low-resolution face images. To address this problem, a super-resolution method that uses nonlinear mappings to infer coherent features that favor higher recognition of the nearest neighbor (NN) classifiers for recognition of single LR face image is presented. Canonical correlation analysis is applied to establish the coherent subspaces between the principal component analyses (PCA) based features of high-resolution (HR) and LR face images. The obtained features from PCA are not good enough for dimensionality reduction and computational complexity when large set of databases are taken into consideration. To overcome that problem Kernel PCA is introduced. Then, a nonlinear mapping between HR/LR features can be built by radial basis functions (RBFs) with lower regression errors in the coherent feature space than in the KPCA feature space. Thus, we can compute super-resolved coherent features corresponding to an input LR image according to the trained RBF model efficiently and accurately. And, face identity can be obtained by feeding these super-resolved features to a simple NN classifier. Extensive experiments on the Yale database show that the proposed method outperforms the state-of-the-art face recognition algorithms for single LR image in terms of both recognition rate and robustness to facial variations of pose and expression.

Authors and Affiliations

N. NagaMounica , Ch. Ganapathy Reddy

Keywords

Related Articles

Reclaiming Individuality of Mysterious Passage

Authorship attribution, the science of inferring characteristics of author from characteristics of documents written by that author become an urgent need to find the original author of anonymous text. In this paper, a no...

Optimal Feed Forward MLPArchitecture for Off-Line Cursive Numeral Recognition

The purpose of this work is to analyze the performance of back-propagation feed-forward algorithm using various different activation functions for the neurons of hidden and output layer and varying the number of neurons...

An Approach to Automatic Generation of Test Cases Based on Use Cases in the Requirements Phase

The main aim of this paper is to generate test cases from the use cases. In the real-time scenario we have to face several issues like inaccuracy, ambiguity, and incompleteness in requirements this is because the require...

Survey of Attacks on Mobile AdhocWireless Networks

Security has always been a key issue with wireless networks since there are no physical boundaries. Experience has shown numerous vulnerabilities to a variety of attacks even when security measures are in place. In the c...

Results evaluation of max rule, min rule and product rule in score fusion multibiometric systems

this paper discusses about unibiometric systems, multibiometric systems, product rule, max rule and min rule of score level fusion. Score level fusion is used to generate scores of a person. Min max normalization scheme...

Download PDF file
  • EP ID EP130152
  • DOI -
  • Views 121
  • Downloads 0

How To Cite

N. NagaMounica, Ch. Ganapathy Reddy (2012). Canonical Correlated Kernel PCA Method for Face Recognition. International Journal on Computer Science and Engineering, 4(8), 1454-1460. https://europub.co.uk/articles/-A-130152