Face Recognition Using Various Illumination Normalization Techniques With Fuzzy K Nearest Neighbour Classifier

The face recognition problem is made difficult by the great variability in head rotation and tilt, lighting intensity and angle, facial expression, aging, partial occlusion (e.g. Wearing Hats, scarves, glasses etc.), etc. In this paper I apply the various preprocessing techniques for illumination normalization for face images prior to face recognition accuracy testing. The aim is to find how these illumination normalization techniques reduces the computational complexity i.e. by either reducing the size of the image or by using the reduced feature set and how these techniques improves the face recognition rate. In this paper the illumination normalization techniques include: Log transformations, Power law transformations, Contrast stretching, Convolution filter, Histogram equalization and Homomorphic filter. Here K Means clustering algorithm is used to cluster the pixels in face image. Binary threshold is applied in the clusters. The proposed work is to compare the performance of various illumination normalization techniques with Homomorphic filters by computing the face recognition accuracy rate and find which illumination normalization technique produces good results with K means cluster using Fuzzy K nearest neighbour classifier. Face recognition accuracy is tested using the ORL face database.