在人脸探测过程中 ,双眼的探测非常重要 ,这是因为人的眼睛是人脸上最容易探测到的特征部位。通过双眼探测还可以对人脸进行定位 ,使人脸的尺寸归一化。探讨了一种基于二值化图像中黑斑几何形状分析的双眼探测方法。因为人眼的形状比较接近椭圆 ,故在研究黑斑的几何性质时 ,重点研究黑斑的椭圆性质。所谓黑斑的椭圆性质是指按照黑斑的面积和二阶矩计算出的椭圆参数。椭圆的主要参数有椭圆的面积、质心、长短轴的长度和方向角 (椭圆长轴与水平方向的夹角 )。为了描述黑斑与椭圆的近似程度 ,定义了椭圆硬度的概念。根据眼睛黑斑椭圆参数的测试数据建立了探测眼睛的数学模型 ,并给出了根据此模型得到的双眼探测的实验结果 In the face detection process, the detection of both eyes is very important, because the human eye is the most easily detected feature on the human face. Through binocular exploration can also locate the face, face size to normalize. This paper discusses a binocular detection method based on the analysis of the black spot geometry in binary images. Because the human eye is relatively close to the shape of the ellipse, so in the study of the geometric properties of dark spots, the focus of the elliptical nature of the dark spots. The so-called dark ellipse refers to the ellipse in accordance with the size of the dark spots and the second moment calculated elliptic parameters. The main parameters of the ellipse are the area of ​​the ellipse, the center of mass, the length and the direction angle of the long and short axes (the angle between the long axis of the ellipse and the horizontal direction). In order to describe the approximate degree of dark spots and ellipses, the concept of ellipticity is defined. Based on the test data of the ellipse parameters of the eyes, a mathematical model for detecting the eyes is set up and the experimental results of the binocular detection obtained according to the model are given