Occluded Face Detection Based on Adaboost Technology

Abstract

This paper proposes an occluded face detection technology based on the Adaboost algorithm. In this paper, we select moving regions for detection using a background subtraction method. The upper half and lower half parts of human face are detected respectively in moving regions by facial detector which was trained based on Adaboost algorithm and Haar features. Our experimental results indicate the occluded face detection can be built in the Adaboost algorithm for detect human faces in front of ATM machine effectively and efficiently.