Secured E-voting system based on iris identification

Abstract

Security is one of the most important requirements in the design and implementation of electronic services. Access to and the protection of data during transmission and storage are critical issues. In this work, the security of Electronic Voting System (E-Voting system) is proposed to handled and process. Secured E-voting system is proposed, designed and implemented for private purposes. The system is proposed to protect mainly using iris identification as an authentication method.in addition to protection of transmission channel and stored data. Iris recognition is implemented using one of deep learning neural network model. Special purpose camera is used for iris capturing in addition to CASIA and IITD data set, recognition result is evaluated and compared with other related works. Transmitted data is protected using modified AES algorithm. Stored data is protected using some of MYSQL properties. Security of the system is evaluated, Where the recognition rate for the proposed method was, 96.5, 98.1, for all data types, and for cameras samples respectively. Speed and security level of the modified encryption algorithm also evaluated, where the modified algorithm processing time for encryption and decryption less than the time of the original algorithm processing time. Where the encryption time for original and modified algorithm was 82 and 78 milliseconds respectively for image of size 59.6 Kbyte, modified algorithm had passed histogram and NIST test. The evaluation of the entire system security proved that the system had acceptable security. The results were promised.