Performance Comparison between DCT And FFT Feature Extraction Methods in Face Recognition
In this paper Performance of Discrete Cosine Transform and Fast Fourier Transform methods for feature extraction
in face recognition system are compared with each other. In this system, before using DCT and FFT, Wavelet Translate was
unitized. Euclidean distance is employed to measure the similarity between the test and reference vectors. According to
experimental results on ORL face dataset, the DWT/DCT method shows better performance, compared to using DWT/FFT
Keywords - Face Recognition, Wavelet Transform, DCT, FFT.