基于双肤色模型和改进的SNoW算法的人脸检测方法

摘要

Aiming at the effect of environment and light on facial skin color extraction, a new face detection method based on double skin color model and improved SNoW algorithm is proposed.Firstly, the image is preprocessed by adaptive illumination compensation method to reduce the color deviation caused by illumination variation.Then, the face information of the preprocessed image is extracted by using the skin color model of YCbCr and HIS.Finally, the improved SNoW algorithm is used to detect the face position.Experimental results show that the proposed method can detect and locate human faces accurately, and improve the detection efficiency and the robustness of the algorithm to complex backgrounds, illumination and expression in complex background and illumination.%针对环境与光照对面部肤色提取的影响,提出一种基于双肤色模型和改进的SNoW算法相结合的人脸检测方法.首先,利用自适应光照补偿方法对图像进行预处理,减少光照变化导致的色彩偏差;然后,综合利用YCbCr和HSI肤色模型提取预处理后图像中的人脸信息;最后,结合改进的SNoW算法检测人脸位置.实验结果表明,在复杂背景和光照变化的情况下,该方法可以准确检测和定位人脸,提升了检测效率和提高了算法对于复杂背景、光照、表情的鲁棒性.