Illumination and Expression Invariant Face Recognition System Using Discrete Wavelet and Hybrid Fourier Feature with Multiple Face Model

Face recognition is one of the challenging applications of image processing. It has been actively investigated by the scientific community and has already taken its place in modern consumer software. However, there are still major challenges remaining e.g. variations in pose, lighting and appearance, even with well known face recognition systems. In this paper, we present a Robust face recognition system should posses the ability to recognize identity despite many variations in pose, lighting and appearance. This proposed face recognition system consists of a novel illumination-insensitive preprocessing method, a combined method of feature extraction using discrete wavelet Transform and hybrid fourier transform with multiple face model to improve the recognition rate, and a score fusion scheme. First, in the preprocessing stage, a face image is transformed into an illuminationinsensitivimage, called an ―integral normalized gradient image,‖ by normalizing and integrating the smoothed gradients of a facial image.Then, for feature extraction of complementary classifiers,multipleface models based upon discrete wavelet and hybrid fourier featuresare applied. The wavelet featuresare extracted from wavelet coefficient values with the same size as the original image. Thesecoefficients are used to describe the face image. The hybrid Fouriereatures are extracted from different Fourier domains in differentfrequency bandwidths and then each feature is individually classified by linear discriminant analysis. In addition, multiple face models are enerated by plural normalized face images that have different eye distances. Finally, to combine scores from multiple complementarym classifiers, a log likelihood ratio-based score fusion scheme is applied.