PCA-ANN Face Recognition System based on Photometric Normalization Techniques

The human face is the main focus of attention in social interaction, and is also the major key in conveying identity and emotion of a person. It has the appealing characteristic of not being intrusive as compared with other biometric techniques. The research works on face recognition started in the 1960s with the pioneering work of Bledsoe and Kanade, who introduced the first automated face recognition system (Zhao et al, 2003). From that onwards, the research on face recognition has widespread and become one of the most interesting research area in vision system, image analysis, pattern recognition and biometric technology. Recently, research on face recognition has received attention and interest from the scientific community as well as from the general public. Face recognition has become a major issue in many security, credit card verification, and criminal identification applications due to its applicability as a biometric system in commercial and security applications to prevent unauthorized access or fraudulent use of Automated Teller Machines (ATMs), cellular phones, smart cards, desktop personal computers, workstations and computer networks. Face recognition has been used by law enforcement agencies for finding criminals, by government agencies for fraud and homeland security, and by financial institutions for ATM and check-cashing security to protect customers against identity theft and fraudulent transactions. By using the face recognition, a picture identity, bankcard or Personal Identification Number (PIN) is no longer needed to verify a customer's identity. Face recognition is also applicable in areas other than security oriented applications such as computer entertainment and customized computer-human interaction applications that can be found in products such as cars, aids for disabled people, or buildings. The interest for face recognition will most likely increase even more in the future due to the increased penetration of technologies, such as digital cameras and the internet, and a larger demand for different security schemes. Face recognition systems (FRS) are still in their infancy. The current FRS still experiencing low accuracy rates due to factors such as illumination, orientation and other disturbances. The quality of a face image has also a big impact on the performance of the FRS. If the illumination on the face image is too high, the face image will be too bright; however, if the illumination is too low, the face image will be too dark. The variation on the illumination will greatly affect the quality of the face image and reduce the performance of the FRS. Thus, it is crucial to improve the quality of the face image in order to have a better O pe n A cc es s D at ab as e w w w .in te ch w eb .o rg