Fast Face Recognition Based on Wavelet Transform on PCA

Abstract

Today the word is moving towards the globalization in engineering techniques, the capacity and techniques established for an identity of individuals using face as a biometric has become more importance. The face extracted leads the many application like photography, security surveillance, database identification etc. This paper includes the comparison of the rate of face recognition using the Principal Component Analysis (PCA) and the PCA using Discrete Wavelet Transforms (DWT). The proposed algorithm uses the concept of DWT for the image compression and PCA for the feature extraction and identification method. The limitations of the only PCA algorithm are a poor recognition speed and complex mathematical calculating load. To eliminate these limitations we are applying the DWT with different decomposition levels, i.e from level 0 to level 3 to facial image by using Daubechies Transform and applying the PCA for feature extraction process. The Euclidean Distance Measures system is used to find the nearest matching features in the whole database. In this paper the the mentioned algorithms are compared with their total recognisation time and the second parameter is the percentage of recognition of a test image. The results shows that the PCA with DWT applied gives higher recognition rate up to 93% than only PCA ,with very less access time.