基于L^＊a^＊b^＊肤色模型和数学形态学的人脸检测

提出利用L*a*b*肤色模型和数学形态学相结合的方法,实现彩色图像人脸检测.建立L*a*b*颜色空间的肤色模型,利用该模型提取肤色区域;然后,用数学形态学算子,完成去除噪声和干扰,分割候选目标等处理过程;最后,利用人脸的比例关系框选出人脸区域.实验表明,该方法计算开销较小,易于实现,准确率较高,速度快.

Face Detection Based on L*a*b* Skin Color Model and Morphology

A novel method for face detection in color images is presented by using L*a*b* skin color model and morphology.First,a skin color model is built in the L*a*b* color space,by which skin-like pixels can be signed.Then,morphology operators are used to delete noise and disturbance in an image and connect the objects selected.Finally,the face area can be located depending on the ratio between the face and the eyes.The experimental result shows that the method presented in this paper is simple,effective,robust and fast.