Face Detection And Recognition Algorithms For Outdoor Environment

Abstract

Face Detection and Recognition based research are the hottest and active research area of Computer Technology. It is the biometric identification of a human face, which matches the image against a library of known faces and one of the biometric individual identification for different purpose. Many studies are done on Face Detection and Recognition for long period of times. The accuracy of recognition under different Position, Illumination and Expression which called PIE under this study is the yet not fully solved .Face DCT based Face Detection and Recognition algorithms is designed for this study in order to fill the gap in Position ,illumination and Expression with respect to the accuracy. The proposed solution is used Visual Studio 2015 with the integration of OpenCV IDE. The study used 1000 image from the collection of real-times system through camera and From ESSEC-EU Face Database from External Gallery. The image of different size are equalized and Training and Testing Process are follow it. The study also used 20 different expression at different expression. This increase the number of face label and it increase the probability of Face Recognition at different Position and expression. Haar cascade classifier is used to identify either face or not. The proposed solution have high rate of accuracy when compared to existing related work research with respect to Position, Illumination and expression as well as it takes low training and Testing time. By using DCT algorithms with Haar Cascade Classifier the accuracy of recognized face is an average rate of 98%.It I also identified that the image input from real world through camera and from external database are also determine the rate of accuracy.