Design Of Dynamic Atendance System Based On Biometrics

Attendance management refers to the investigation and management of employees' attendance by enterprises and institutions.The dynamic attendance system based on biometrics can avoid the problems of forgery,loss,and forgetting of traditional attendance methods such as manual check-in,magnetic card,IC card,etc.,providing a basis for employee assessment and salary evaluation.In this thesis,a multi-modal biometric recognition algorithm is used to implement a dynamic attendance checking system based on biometric recognition.First of all,in view of the defects of stability and accuracy in single biometric recognition,the complementarity between multi-modal biometrics is used to improve the accuracy of biometric recognition.Based on the study of biometric recognition theory and their own characteristics.In view of the characteristics of face recognition,gait recognition,and feature fusion,the system used face recognition,gait recognition based on improved convolutional neural network,and multi-modal biometric recognition schemes fused at the feature layer,and completed the algorithm research,through improving the convolutional layer of neural network to extract the facial features and gait features.These two biological features were fused to generate new biological features,and then the biological features were extracted through the convolutional layer and the pooling layer.And finally the recognition results were output in the softmax layer of the convolutional neural network.After simulation verification,the algorithm used improved the accuracy and stability of the system recognition.Secondly,based on the needs of modern enterprises and institutions for the time and attendance system,a dynamic time and attendance system scheme design was carried out,and it was determined to be composed of a multi-modal biometric identification subsystem and time and attendance management subsystem,and the hardware selection and construction of the system were completed.Finally,based on the above algorithm,the dynamic time and attendance system based on biometrics recognition and mobile time and attendance APP software are designed.According to the recognition results and corresponding strategies,the attendance of the attendance personnel was determined,and the environment test in the laboratory can achieve the predetermined features.The verification results show that the algorithm of face recognition and gait recognition fusion in this thesis is better than the single biometric recognition method in the feature layer.The dynamic attendance system designed in this thesis can meet the dynamic attendance requirements of enterprises and institutions.