Chin-Chuan Han, Hong-Yuan Mark Liao, Kuo-Chung Yu and Liang-Hua Chen
An efficient face detection algorithm which can detect multiple faces in cluttered environment is proposed. The proposed system consists of three main steps. In the first step, a morphology-based technique is devised to perform eye-analogue segmentation. Morphological operations are applied to locate eye-analogue pixels in the original image. Then, a labeling process is executed to generate the eye-analogue segments. In the second step, the previously located eye-analogue segments are used as guides to search for potential face regions. The last step of the proposed system is to perform face verification. In this step, every face candidate obtained from the previous step is normalized to a standard size. Then, each of these normalized potential face images is fed into a trained backpropagation neural network for identification. After all the true faces are identified, their corresponding poses are located based on the guidance of optimizing a cost function. The proposed face detection technique may locate multiple faces oriented in any directions. Besides, the morphology-based eye-analogue segmentation process is able to reduce the background part of a cluttered image up to 95\%. This process significantly speeds up the subsequent face detection procedure because only 5-10% regions of the original image are left for further processing. Experiments demonstrate that an approximately 94% success rate is reached and the relative false detection rate is very low.