A face detector based on color and texture

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.