| As an important personal identification document,ID card is widely used in Internet self-service systems.Therefore,reliable authentication is an important measure to ensure account and data security in the Internet environment.The paper conducts dual identity authentication research on the relevant information provided by ID card images,which has good theoretical and practical value.Firstly,the paper sorts out the current research status of face recognition and face verification,and then verifies the actual requirements of operator information authenticity for self-service system.The paper proposes a dual authentication scheme based on ID card information from the perspectives of human face verification and human witness comparison,and implements the prototype system.The program directly collects basic information through ID card images,improves the efficiency of information collection,and ensures the security of the system by using two-factor authentication of comparison between live verification and human certification.The main work of this paper is as follows:(1)In order to replace the face collection problem under the camera with photos in identity authentication,the paper proposes a biometric verification scheme combining texture feature face verification and blink detection using Haar wavelet analysis.The scheme first uses Haar wavelet analysis technology to distinguish between real people and photos.At the same time,it uses face key points to detect and calculate the aspect ratio of the eyes to determine whether it is a real person.The first re-verification completed ensures that the operator is a real living face and improves safety.(2)Face verification based on the comparison of ID card images and real-time face images.Based on the experimental analysis of traditional eigenface and convolutional neural network methods,a scheme for face verification based on VGG16 deep convolutional neural network is proposed.The face of the scene is compared with the face on the ID card to determine whether it is the same face.The completed second re-authentication ensures that the real living face of the operation has the correct identity,improving the robustness and stability of the human witness comparison.(3)Aiming at the identity authentication requirements in self-service applications,the paper implements a dual authentication scheme based on ID card image information,and gives a dual authentication prototype system.The system is based on ID card image design and realizes automatic extraction of relevant information.At the same time,it integrates the research of face biometric verification and deep face learning-based face verification method.The prototype system mainly includes modules such as automatic extraction of ID card image information,face biometric verification and face verification of face images,which can be conveniently accessed into self-service related applications. |
