A face detector based on color and texture
Mohammad Reza Mahmoodi, Sayed Masoud Sayedi · 2014
Face detection is one of the most important parts of biometrics and face analysis science. Numerous methods and algorithms have been developed in recent years; however, there is a sensible gap between the current detection rate and the ideal one yet. In this paper, a novel multi-stage face detection method is proposed which can remarkably detect faces in different images with different illumination conditions, variety of poses and disparate sizes. The idea is to utilize a preprocessing step to filter many non-face windows by means of a skin segmentation procedure in order to boost the speed of the detection and also utilize the color information as much as possible. Subsequently, candidate windows are fed to a Local Hierarchical Pattern (LHP) generator unit where a new texture pattern is produced. Based on this pattern, a kernel probability map is calculated for each window, and by summing probabilities of all kernels and comparing it with a predefined threshold, decision is made about content of the window. This algorithm not only effectively eliminates many non-face regions, but also it is capable of detecting faces with relatively acceptable rates.