Algorithm Research of Face Recognition System Based on Haar

Face detection is to detect the face from the original image, which is the basis of face recognition technology. Face is a kind of natural structural object with quite complex details, which will be challenges to this kind of recognition. Face has the variability of pattern because of the different appearance, expression and skin color, and in general, there may be appendages such as glasses, beard and so on. Face image as a three-dimensional object has shadows caused by light also unavoidable. In this paper, the most mainstream Harr feature classifier algorithm is used, innovatively propose non-human rejection algorithm to reduce the impact of complex environment on detection results and improve detection accuracy.