Implementation Of Human Face Detection Algorithm Based On Complexion

With the development of computer science,face detection has become one of the hot research directions.In this context,according to the skin color cha racteristics of the face,this paper designs a face detection algorithm based on skin color,which can quickly identify and detect the face.First,the input of the image containing the face area is compensated to make up for the uneven distribution of light caused by light or shadow.The basic idea is to arrange the brightness of all pixel values in the input image according to the order of magnitude from high to low,extract a percentage of pixels,and determine whether the number of these pixels is greater than a threshold,and use their brightness value as a reference white.Secondly,the face skin color model is established.In the YCr Cb format,the color of face area is obviously different from the background area.After two values,the face area and the background area are distinguished,and the two value model of face color is established.Again,the morphology of the corrosion and expansion process of binarization image,the purpose of using this technology can not only refine the image,but also make the outline of the face region connected smoothly,then the median filter,further to remove small areas of non face regions,more in line with human face area of concern.Finally,the extraction of image binarization for facial region contours,accordin g to the "standard",to find out the target pixel where the contours of the object;according to the characteristics of "tracking standard" and "detection" pixels,find other pixels on the target object determine the face region.Contour tracking with a single contour line,then the geometric contour selection,if geometric area face contour in the image is larger than a threshold and the face width and length of the geometric rules in certain proportion,can determine the face region.According to the contour tracking and geometric rules,the number of faces and the number of faces are marked.Face detection algorithm using Microsoft Visual Studio 2008 development environment based on C language,the test results show that the detection method of face detection based on skin color has a higher rate of this,whether it is a single face image,including at least 8 people or above the face image,with more than 70.37% accuracy.And the detection algorithm to adapt to a variety of complex scenes,especially in the test image multi face,not light,measured the attitude change,personnel appendage occlusion effect,has very strong applicability.