基于Haar特征和改进的AdaBoost算法的人脸图像识别  

Recognition of Face Image Based on Haar Features and Improved Adaboost Algorithm

作　　者：贺瑜飞 HE Yu-fei(School of Mathematics and Statistics,Yulin University,Yulin 719000,China)

机构地区：[1]榆林学院数学与统计学院

出　　处：《榆林学院学报》2019年第6期69-70,75,共3页Journal of Yulin University

基　　金：榆林学院高层次人才科研启动基金项目(3GK39)

摘　　要：针对在复杂背景的情况下,AdaBoost算法主要存在识别率较低,速度慢的问题,提出了一种基于改进的AdaBoost算法和Haar特征的方法。首先进行的是用扩展的Haar特征进行特征值的计算,然后描述了boosted分类器,并指出其存在的问题,接着描绘了改进的AdaBoost算法。多分辨率搜索进行了算法优化,试验结果表明,改进的AdaBoost算法和Haar特征提高了人脸检测的实效性,漏检率降低,并且识别率有很大的提高。To solve the problem of low recognition rate of the traditional AdaBoost algorithm in identifying face images under a complex background,we chose a Haar features and improved AdaBoost algorithm for human face image recognition.Firstly,we used the extended Haar features to calculate the Haar value,described the boosted classifiers,and pointed out the problems.Secondly,we described the AdaBoost algorithm and used multi-resolution search to improve the algorithm.The experimental results showed that the improved algorithm increased the real-time of the face detection,lowered the missing rate,and greatly improved the recognition rate.

关 键 词：HAAR特征 特征模板 ADABOOST算法 

分 类 号：TM71[电气工程—电力系统及自动化]