Design And Development Of Face Detection And Face Recognition System Based On Basic Algorithm

Face recognition technology has the advantages of convenient collection of human face features,low cost of equipment,no direct contact with users and other advantages.In recent years,it has been rapid development,has a huge application space and investment value.It is foreseeable that the 21 st century will inevitably become a "brush face era".However,because the technology is not yet mature,still need a lot of researchers study in this area.But for a new algorithm,how to quickly and easily evaluate its effectiveness and rationality has become a need to be addressed problem.The evaluation of a new algorithm often needs to be compared with the previously proposed algorithm.Therefore,this paper designs a human face detection and recognition algorithm evaluation system,through the "recognition accuracy" and "running time" two evaluation indicators to evaluate the new algorithm.At the same time,the paper is inspired by the nearest neighbor classification algorithm,innovatively improved WNN algorithm which is classified by "weight",and use the system to evaluate the algorithm,the new algorithm WNN has further research value.In addition to,the user can also use the system to select the appropriate algorithm for face detection and face recognition according to different practical application environment.At the same time,for the researchers who have just started to study face recognition technology,the open source model of the system is very beneficial to them.Moreover,by adding a camera,the system can be expanded and applied in the field of the access control system.In order to design and develop the system,the paper mainly studies two aspects: In the face detection,the Adaboost algorithm based on Haar feature and the face detection algorithm based on skin color model are studied.In the face recognition algorithm,the algorithm is divided into three steps: image preprocessing algorithm,image feature extraction algorithm,classifier design,and each step to study three or more sub-algorithms.After studying the basic principles and algorithms of face detection technology and face recognition technology,this paper analyzes the foreground demand,equipment demand and function requirement of the system.According to these requirements,this paper designs the overall framework of the system.It also designs the sub-modules for face detection and face recognition,including color space conversion module,skin color model selection module,mathematical morphology processing module,image pre-processing module,image feature extraction module,classifier selection module,face database selection module and so on,then showing the key code in the system.Finally,the system uses the GUI in the Matlab to design an interface,the algorithm detection and recognition of the results were displayed.