Group-based Security Research Biometric Authentication And Shared Secret Can Not Completely Revoked

Identity Authentication is the process of users’identification in some systems. The essence of authentication is to restrict access to unauthorized users access to some secret resources, that is, the user access restrictions. Some authentication methods which we are common are user name and password, IC card authentication, dynamic token authentication, USB Key authentication, SMS password authentication and so on. Because of its importance the authentication technology has become a hot research scholars at home and abroad, usually the authentication only on the identity of each person identification and verification, as long as the personal information with predetermined information found by the authentication. Group secure authentication is n authorized users and at least k users present at the same time, submit their information and recover the information of other users, the only way to authenticate through the system. If there are fewer than k users, even if k-1users present at the same time, submit k-1copies of secret information, they are not certified by the system security group identity. Mainly used in the criminal investigation, defense, finance, e-government, e-commerce, education, transportation, medical and other industries.The main work of this paper is divided into five parts:(1) The collection and sorting related to the face image processing;(2) The completed of face image database image pre-processing of the human face, which is normalized;(3) Extracting the histogram of each normalized face image sequence, and to obtain all the biometric hash coding of face images of persons.(4) The use of Mixed-Separation Algorithm performed to achieve the human face image hash coding random mixing and separation, generated and distributed a hash key to each user;(5) Any combination of greater than or equal to the threshold value can submit their respective face image biometric hash code and the hash key, can recover the other user’s face image hash coding and hash keys, if constructed image hash coding with the face of all users in the group authentication codes are in accord with which we are prior to construct, then the combination can access the authentication system.In this paper, the method of the group authentication is proposed by facial biometric hash coding non-perfect secret sharing to achieve, non-perfect secret sharing is achieved by the mixed-separation algorithm. Difficult to crack the robustness of the face image and key characteristics, and different people face image is very different, so the method can protect secret documents well.In this paper, the internationally accepted human face database Yale Face Database B verify the effectiveness and feasibility of the algorithm, the database consists of10individual face image, each image size is640x480, all face images normalized size is240×200, and then calculate the hash code of the face image of each user and randomly mixed, then separated from the codebook according to a predefined random hash key, that is the shared key for each user. The experimental results show that a combination of any value greater than the threshold (threshold set in this paper is four) by submitting their face image hash coding and shared key, and the recovery of other users face image hash coding and shared key authentication, otherwise not have access to secret documents by the security group.