基于二重BP神经网络的静态人脸检测算法的研究

AN ALGORITHM OF STATIC HUMAN FACE DETECTION BASED ON DUPLICATE BP NEW WORK

提出了一种基于BP神经网络的静态人脸检测算法。首先 ,通过形态学闭运算提取包含人眼区域的二值图像 ,对二值图像进行基于连通性的标注 ,标注区域会聚形成候选人眼点。对待选人眼点进行筛选后 ,送入BP神经网络进行判断 ,得出人眼点。最后人眼点两两组合形成待选人脸 ,送入第二个BP网络进行判断。

An algorithm of static human face detection based on BP network is proposed. Firstly, the binary image containing human eye regions is extracted by morphological operation of closing, after which, the binary image just obtained is labeled based on connection analysis. Then, the labeled regions are assembled into candidates of human eye dots . After the dots are chosen , they are sent into BP network for the obtaining of the eye dots. Each pair of the eye dots is used to form the candidate of human faces. Finally, the candidates are sent into the second BP network for the judgment of the face.