【摘要】 提出一种基于主成分分析的人脸检测方法。由于RGB各波段之间有高的相关性,所以不适用来进行场景分割和分析。通过对彩色图像RGB波段数据进行PC变换使得数据差异最大化。特征值大的 PC 波段包含最大量的数据差异,较小的特征值包含较少的数据信息和较多的噪声。对第二波段图像数据进行二值化,定位人脸的初始位置。对候选区域进行判断,抛弃不满足人脸长宽比例的区域,从而对图像中的人脸精确定位,该方法速度快,定位准确。更多还原

【Abstract】 This paper presents an algorithm to detect human face which is based on principle component analysis (PCA). Because the RGB data bands are often highly correlated, the Principal Component (PC) Transformation is used to produce uncorrelated output bands. This is done by finding a new set of orthogonal axes that have their origin at the data mean and that are rotated so the data variance is maximized. PC Bands with large eigenvalues contain the largest amounts of data variance. Bands with lower eigenvalues contain less data information and more noise. In detecting face, it is best to output only this band with the second largest eigenvalue. Combining the ratio of width and length, we get the accurate location of face. The experiment results show that the algorithm is fast and accurate.更多还原