The Research About Face Detection Method And Eyes Location Arithmetic Based On Skin Color

Face detection is original from face recognition, which is the key section of face recognition system. In recent years, face detection has been attached importance to by researchers as an absolute task with the development of electronic commerce. Nowadays, the application background of face detection has exceeded the range of face recognition system, and it has important applied value in Content-Based Image Retrieval. Digital Video Processing and Visual Supervision And Control etc.In this paper, the following research work is done in allusion to the status quo and difficulties of face detection research: (l)The methods that skin information is used to reduce the search range firstly are discussed.The skin color is an important characteristic which is different from other sceneries, the first problem that is the face segmentation based on the skin color to be solved is to choose the color space. There are some methods in showing the color. In this paper, we carried on the simple introduction and analysis according to the skin pick-up patterns in common color spaces in latest methods of face detection, and Cb, Cr ellipse clustering method of Anil K · Jain is improved base on its shortcoming ,by the experiments, we find the improved method is better than other methods in distilling the person's facial features. (2) Eyes Location Based on Skin color. Geometry Characteristics and GraynessThe eye is the most important facial feature in face detection. , a reliable and fast method for locating the eye precisely in an image is vital to many practical applications. A quick efficient and steady eyes detection algorithm based on skin color geometry characteristics and grayness is brought forward. This algorithm adopts a coarse strategy to a fine strategy: First, the face candidate is detected with the help of skin color information; Secondly, eyes are roughly located with the help of geometry characteristics that are the distance of the two eyes and the angle between the central line of the two eyes and the x axis. Last, the eyes are accurately located with the help of eye grayness. The grayness of eyes has mutation, and the horizontal characteristic is more obvious than the vertical characteristic. The complexity of each block which is roughly located in the second step is computed and leaved in a array. The biggest two in the array are the eyes. This algorithm is proved to be efficient and robust to the image in different background and size. (3)A face detector based on differential image and PCABased on differential image and PCA a new-method for face detection is proposed which modified the Eigenface technique. According to the color of skin, the probably fields where face may be are detected. Then for finding the exact positions of faces a finer matching is performed in these detected fields and a Mosaic template is used to suppress the false face. Considering the correlation between the pixels in face image, the PCA method is performed both on the raw image and its horizontal differential image and vertical differential image for feature extraction. And then the stability of the face detection is improved.