An algorithm for human eye location in complex background based on image processing

In order to reduce the impact of image background and illumination in face locating, this dissertation has put forward a new algorithm to locate human eyes, applying YCbCr model to extract human face region, and then locating eyes correctly according to geometry and pixel features of human eyes. Experimental results show that this algorithm can be applicable in images with different backgrounds and Nonuniform illumination environment. It has proved that it is realtime and accurate.