Face Detection Based On Near Infrared Images

Face Detection is to determine whether there is someone in a given image, if yes, then figure out the position and scope of the people' s face. The face detection demands a lot of calculating work which is the most important pre-processing of face recognition. How to detect the faces rapidly and effectively is an important project for the face detection.The light environment is a big challenge to the traditional face recognize systems which based on visible light images. Due to the change of light, lots of face recognize light productions are largely limited in reality applications. When the environment light of recognize time is different from that of login time, the capability of face recognize would largely reduced. It's also the biggest reason that restrict the application of face recognize to users.To overcome the infection of environment light, both academe and interrelated enterprise have made large amount of research and technological development. But most of their works are trying to improve the visible light face recognize system in existence , so that they can reduce the infection of environment light. Though they have made some improvements, it's too little. It's the near-infrared technology that can conquer the infection of change of light.The main work of the paper is to summarize of the existing face detection techniques. It consult and improve the previously researches, bring forward to adopt Quoit filter to get eyes features, so that to detect faces. It uses Template Matching to validate, and prove the validity of this arithmetic by experiments. The algorithm includes two parts: first is the rough face detection; second is the face validation. At the stage of rough face detection, adopt the Quoit filter to get eye features ,and then get the candidate face images. At the stage of face validation, uses Template Matching to validate.The thesis made a detailed introduction of rough face detection and face validation, and have experiments on the test set . The experiment proves the validity of this method.