Face Detection Technology Research Based on AdaBoost Algorithm and Haar Features

Abstract

AbstractFace detection technology is an important branch of digital video processing technology. This chapter proposed a face detection technology based on haar features and AdaBoost algorithm. In this chapter, haar features, integral image, AdaBoost algorithm, and cascade classifier were introduced, features were extracted by haar features, and integral image and AdaBoost algorithm were used to select suitable haar features for facial features; classifier was the classifier finally constructed and used to face detection. It was found out that the common features (a small amount of haar features) play an important role in face detection. To verify this method, experiments on both static pictures and video stream were conducted. Experimental results showed that the model of haar features and AdaBoost algorithm face detection technology had high detection accuracy with more hardware cost while a small number of common features would reduce hardware cost and had greater significance in real-time face detection.KeywordsFace ImageFalse Alarm RateVideo StreamFace DetectionIntegral ImageThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.