Human Identification via Face Recognition: Comparative Study
Dr.R.Satya Prasad, Muzhir Shaban Al-Ani, Salwa Mohammed Nejres · IOSR Journal of Computer Engineering · 2017
Biometric recognition becomes an attractive issue of processing, and have a vast amount of real applications.According to wide spread of Internet and the new aspect of Internet of things, millions of human characteristics are flying around the world.So the security and identification become a big challenge and most of the individual needs to apply secure system to save their information.Face recognition is a narrow field of biometric recognition but there are vast amount of works and projects concerned this field.These works used various technique to reach high accuracy of recognition.The big challenge of face recognition is how to improve the recognition rate according to the implemented system.The main aim of this approach is concern with the scientific survey of the last two years to generate a simple comparative study of techniques, methods, challenges and benefits of applying these approaches.Mean Square Error (MSE) and Peak Signal to Noise Ratio (PSNR) are measured for all the tested methods 2D-WHT, 2D-DCT, 2D-DWT, 2D-FFT and M-2D-DWT in which the results obtained from M-2D-DWT gives better values of MSE and PSNR.