复杂背景下基于模糊集的快速人脸检测

针对如何在复杂背景下快速定位人脸并对其进行确认的问题,提出了一种基于模糊集的快速人脸检测方法.该方法利用肤色模型对图像进行肤色检测,获得肤色分割区域后,并利用数学形态学算子对分割区域进行处理并获得人脸候选区域,再结合模糊集理论,融合人脸的特征进行模糊事件的判断,确定人脸候选区域是否包含真正的人脸.实验结果表明,该算法具有较强的鲁棒性和高的正确检测率,并能满足实时应用的要求.

Fast face detection based on fuzzy sets in complex background

Focusing on how to locate face and affirm it in complex background, a fast face detection algorithm based on fuzzy sets is proposed. Skin model is firstly used to detect skin color in image, so as to obtain skin segment regions. Then mathematic morphologic operator is used to deal with the skin segment regions to get the face like regions, and based on the theory of fuzzy sets and several characters of the face, a fuzzy membership function is constructed to judge whether the face like region concludes the face. The experimental results show that this algorithm has a high robust and retrieval rate. And it can meet the need of real time application .