Sub-Image Homomorphic Filtering Technique for Improving Facial Identification under Difficult Illumination Conditions

In this paper we will propose a simple modification of standard homomorphic filtering technique and thus significantly improve face recognition performance on images with difficult illumination conditions. We will also give a detailed theoretical description of a homomorphic filter and compare our proposed method to common illumination compensation techniques used in face recognition literature. The comparisons will be performed on standard grayscale FERET database and this will, in addition, be the first evaluation of homomorphic filter on this database. Results will show that our method yields significantly better identification results than standard illumination compensation methods currently used in face recognition.