A Face Detection Method Based on Sliding Window and Support Vector Machine

Abstract

Face detection is a biometric technology based on human face features for identity authentication. With the development of e-commerce and other applications, face recognition has become the most potential means of biometric authentication. Classical face recognition is based on statistical methods, but the accuracy of this method is not high. In this paper, a face detection method based on the sliding window and support vector machine is proposed. Firstly, the image is divided into blocks, and the HOG features of the target image are extracted. Then the support vector machine model is trained through the data sets of human face and non-face. The support vector machine model can detect whether the target area belongs to the face area or not. Finally, the whole face area is detected by the sliding window model. Experiments verify the effectiveness of the proposed method.