Research And Implement Of Face Detection Algorithm

Biology recognition technology, a kind of technology which can identify individual status through connecting with High-tech means such as optics,acoustics,biosensor and biostatistics as well as using people's inherent physiological characteristics (such as fingerprint,face,iris and so on)and behavioral characteristics(such as handwriting,voice,walking posture and so on).Human Face Detection and Recognition is the most popular branch in biological characteristic recognition. It is also a hot research topic in many areas such as picture process, pattern recognition, computer vision and so on. It is widely used by public security department, offenders search programs, safety department, dynamic monitoring identification, Bank Cryptosystem and many other fields.Human Face Detection is the Preliminary work of recognition, the theory and Implementation methods of Human Face Detection can also be used by other objects identifications, such as Vehicle detection, pedestrian detection and so on. This article researches it deeply. The main work and achievements of this thesis lie in the following several aspects:1.The thesis summarizes Biology Recognition Technology and its developing direction;it mainly describes Human Face Recognition, Application background of Detection Technology, significance of the research,domestic and international research situation as well as the difficulties of Human Face Detection.2.The thesis summarizes different ways of Human Face Detection, some classic algorithm ways of Human Face Detection as well as the database it used.3.On the base of 2DPCA algorithm, the thesis put forward an improved way of algorithm for Human Face Detection.4.The thesis put forward a Human Face Detection algorithm on the base of Template Matching and improved 2DPCA. After rough classification of human face template which construct two eyes-face template firstly and then detailed classification through the improved 2DPCA method, this method is not only discard many non-human face areas, but also improve the speed of next detection, finally, merge all the areas detected and then get the human face area.5.By programme we can realize the classify method that is divided into rough classification and detailed classification and is called two layers detective face method,which can detect image of single face and multi-faces and analyze the data of the experiment.