Extraction of face regions in color image

We propose a method to automatically locate the frontal human face regions under a complex background in a color image. The method starts from clustering the color space using the histogram of the detected skin pixels, where the YCrCb color model and HSV color model are used to detect the skin area. Then an opening morphological operation is used to remove the small objects in the background area. Some rules are used to select the candidate face regions according to the shape and size information. Finally, the face regions can be located by using a backpropagation neural network to detect the appearance of two eyes in the candidate face regions. Experimental results show that the proposed approach is efficient.