Face Recognition Based On Marking Secret Authentication System Design And Implementation

The electronic secret level marking system is an essential ingredient of securitymechanisms of e-commerce and e-government system. It not only plays important rolein the security of research management, but also is a typical application-levelinformation security technology. Strict and secure identity authentication mechanismselectronic document identifies the security classification system security andreliability.The biometrics identification technology is based on the identification of humanphysiological or behavioral characteristics typically have theuniqueness,stabilitycharacteristics,it is more and more attention and study.Face recognition is the mostcommonly usedbiometric recognition algorithm is a very wide range of biometricand agood performance.Identity verification based on face recognition can be applied to thesecurity classification marking system,construct electronic document securityclassificationmarking system based on face recognition authenticationmodule,whosemain role is: to achieve the function of user permissions and authentication of users ofindividual consistency,for the dense identification system to achieve a more securetwo-factor authentication to provide support.In this study,analyzes the demand for recognition marked secretauthenticationsystem authentication,design and implementauthentication scheme based on facerecognition.The programis based on the AdaBoost face detection algorithm andtwo-dimensional principal component analysis (2D-PCA) feature extraction,first useAdaBoost algorithm to dynamically capture the human face,and then the After imagepreprocessing,feature extraction performed by2D-PCA, and finally with the nearestneighbor classifier face recognition. Verify the design of themodule through the loginmodule is designed for face recognition,face detection algorithm design, based on thecharacteristics ofthe2D-PCA face extraction algorithm design, the design of facerecognition algorithm, the control DLL interface to achieve a good systemimplementation and operating results. On this basis, the face recognition-basedauthentication modulepackaged as dynamic link library, improve the reusability ofthemodule. The practical application results show that: the program can reach a higher recognition rate to meet the needs of thesecurity classification marking system.