人脸检测：
第一部分，使用Harr-like特征表示人脸，使用“ 积分图”实现特征数值的快速计算;
　　第二部分， 使用Adaboost算法挑选出一些最能代表人脸的矩形特征( 弱分类器)，按照加权投票的方式将弱分类器构造为一个强分类器;
　　第三部分， 将训练得到的若干强分类器串联组成一个级联结构的层叠分类器，级联结构能有效地提高分类器的检测速度。(Face detection:
In the first part, the Harr-like feature is used to represent the human face, and the "integral graph" is used to realize the fast calculation of feature values;
In the second part, the Adaboost algorithm is used to select some rectangular features (weak classifiers) that represent the human face, and the weak classifier is constructed into a strong classifier according to weighted voting;
In the third part, a series of strong classifiers are formed in series to form a cascade classifier with cascading structure, and the cascade structure can effectively improve the detection speed of classifier.)