Face Recognition is an efficient biometric system to identify human beings effectively. In this paper, we propose SWT based Face recognition using FFT features. The face images of various face databases with different sizes are resized to 112Ã—92 to maintain uniformity. The Stationary Wavelet Transform (SWT) is used on face images to obtain low frequency sub-band of size 112Ã—92. The Fast Fourier Transform is applied on SWT subband to extract final absolute FFT features of face images. The Euclidean Distance is used to compare features of database images with test face images to compute performance parameters. The SWT is the shift invariant transformation used on face images of a person with different pose and intensity values to extract initial features for better recognition. The final FFT features of SWT matrix identity effectively similar and different face images of human beings. It is observed that the performance of the proposed method is better compared to existing methods.
Keyword : Biometrics, SWT, FFT, Face Recognition
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.