A new entropy-based algorithm for face localization

Detecting and localizing a face in a single image is the most important part of almost all face recognition systems. Face localization aims to determine the image position of a face for verification purpose of documents such as passport, driving license, ID cards, etc. In this paper an entropy-based method is proposed for detecting the high information region of the image, which may include eyes, mouth, nose, etc. The derived regions in this stage of recognition are sent to feature extraction and classification phase. The method has been tested on the ORL database. The results show the effectiveness and robustness of the proposed method for face detection and localization in presence of white additive Gaussian noise up to 25 dBw. We have achieved localization rate 99.75% for detection of faces in the ORL data set that we had which means 1 miss over 400 ORL faces.