摘要
人脸特征提取是人脸检测的关键环节,有效的人脸特征将使得人脸检测更精确。Haar-Like特征作为一种矩形特征,虽然简单、计算迅速,但只能描述特定方向的图形结构。提出的类人脸特征是一种反映人脸灰度分布模型的矩形特征,更加有效地描述了人脸的特征。所提出的人脸检测算法,应用BP神经网络算法训练肤色区域,进行肤色分割。应用类人脸特征的AdaBoost算法进行人脸检测。实验结果表明,该算法可以提高人脸检测的检测率。
Facial feature extraction is the key process of the face detection. Effective features make the face detection more exactitude. Although Haar-Like feature is simple and computed rapidly by integral image, as a rectangle feature, the only orientations available are vertical, horizontal and diagonal. This paper presents a face-like feature that expresses the face gray distribution model, which describes the facial feature more effective. The face detection algorithm of this paper, skin segmentation is done by using BP neural network to train the skin region. Face detection is done by the AdaBoost algorithm with the face-like feature. Experimental results show that this algorithm can improve the detection rate.
出处
《计算机工程与应用》
CSCD
2013年第3期194-197,共4页
Computer Engineering and Applications
