Title :
Effective Feature Selection for Face Recognition Based on Correspondence Analysis and Trained Artificial Neural Network
Author :
Pazoki, Zohreh ; Farokhi, Fardad
Author_Institution :
Sci. Assoc. of Electr. & Electron. Eng., Islamic Azad Univ. Central Tehran Branch, Tehran, Iran
Abstract :
This paper presents a face recognition method based on correspondence analysis (CA) and trained artificial neural network. In this algorithm, features are extracted using CA, then these features are fed to Multi layer Perceptron (MLP)network for classification and finally, after training the network, effective features are selected with UTA algorithm. The obtained experimental results indicate high average accuracy (98%) and the minimum run time of the algorithm as well.
Keywords :
face recognition; feature extraction; image classification; multilayer perceptrons; correspondence analysis; face recognition; feature extraction; feature selection; image classification; multilayer perceptron network; trained artificial neural network; Accuracy; Algorithm design and analysis; Artificial neural networks; Classification algorithms; Face; Face recognition; Feature extraction; Correspondence Analysis (CA); Multilayer perceptron (MLP) neural network; UTA algorithm;
Conference_Titel :
Signal-Image Technology and Internet-Based Systems (SITIS), 2010 Sixth International Conference on
Conference_Location :
Kuala Lumpur
Print_ISBN :
978-1-4244-9527-6
Electronic_ISBN :
978-0-7695-4319-2
DOI :
10.1109/SITIS.2010.23
