【摘要】 讨论了一种彩色静态图像的人脸检测法,先使用光照补偿对图像进行预处理,再采用YCbCr肤色模型检测出图像的肤色部分,然后对肤色部分进行膨胀、腐蚀,去除非人脸区域,最后使用人工神经网络对候选人脸区域进行验证,最终定位出人脸区域。实验证明了该方法对不同光照环境和复杂背景的图片有较好的适应性,提高了检测速度与精度。更多还原

【Abstract】 An algorithm for face detection in color static images is discussed.The algorithm works by first doing lighting compensation on input image,then it detected the skin regions by the YCrCb skin model.Third,it eliminated the regions which are not belonged to the real face region by dilation and erosion.Finally,it used the neural network to evaluate the face candidates and then located the real face region.The experiments proved that the adaptability of this algorithm to those images of different lighting environment and complex background is good.It heightens the speed and the accurate degree of the detection.更多还原