Novel face detection using Gabor filter bank with variable threshold

Face detection is the method of locating human faces in a given image under all lighting conditions, scales and orientations. Face is a unique feature of every person and the same is applicable to pupil, iris and fingerprints which are also unique as well. With the improvement of technology, neural network and processors´ high capacity have resulted in induction of informatics in this area. Automatic face detection and recognition has been drawing the main attention in the recent years. We have proposed here an accurate face detection system that can detect faces under different contrast with hurdles like faces with spectacles, heavy beard and even closed eyes. We use Gabor filter bank with varying threshold for feature extraction and face detection.