YASAMAN HEYDARZADEH at al: AN EFFICIENT FACE DETECTION METHOD USING ADABOOST

In order to build a fully automated system that analyzes the information in face image, there is a need for robust and efficient face detection algorithms. One of the fastest and most successful approaches in this field was presented by Viola. In this paper, Viola approach is used as the basis algorithm in order to propose a new efficient face detection system. Using Haar-like features for facial parts and learning these features by AdaBoost algorithm make proposed approach possible to detect occluded faces and increase the detection rate. Moreover, some preprocessing techniques are also used to limit the candidate places which make final algorithm runs faster. The experimental results represent that, the proposed approach has the capability of achieving a detection rate of 94.7% while false positive rate is just 2% .Furthermore, this system can detect faces in image, averagely four times faster than Viola approach.