Research Of Face Detection Based On Skin Model And AdaBoost Training Algorithm

At the moment, with the rapid development of computer information technologyand the widely use of intelligent Human-Computer Interaction system, the facedetection technology had been the widely used in the computer vision, digital imagemanagement and pattern recognition domain. Face detection is the first step of facerecognition, face detection is to detect the whether the faces in the images, if havemany faces in the image, then detected the faces and determine the position andposture, because of the influence the illumination and brightness make the facedetection has been a challenge studied task.Firstly we described the based conception of face detection technique.Intensively studied the based face detection theory. Through the lots of theexperiments included that our proposed the combine the AdaBoost algorithm and skinmodel has the best detection rate and misclassify rate. The experiments show that ourface detection technique in this paper is reasonable.Our work based on AdaBoost algorithm and face skin model. We adopted theseparated chrominance and luminance color space, through our experiments showsthat skin color information is a universal face detection method.Our proposed method has been aimed at the illuminate and brightness problemsstudied that combining skin color and AdaBoost algorithm is much more effective,meanwhile can detect face with high detection rate and low false acceptance rate.Through the most experiments we can conclude our method effectively applied to thecase of multi-face and multi-pose images.