The Biometrics System Based on Iris Image Processing: A Review

Recently, it reported that the cases of cybercrime had increased by about 13% in 2017, especially in data fraud crime, due to the poor security system. One of the solutions to this problem is by implementing a biometric system as a security platform. Biometric is a method of personal ID authentication by using human recognition using unique human biometric values such as fingerprint, voice, face, or iris. This review paper focused on the iris biometric system since Iris biometric widely used as a system for maintaining data security, such as ATM, cellular phone, etc. Moreover, the biometric has very high sensitivity and accuracy for recognition than the other. In this paper, we will review some experiments of iris recognition that was done by researchers. As a result, we will present some comparisons among several studies, including in feature extraction technique, performance, and time computation required on the matching process. From the paper review, we can analyze that a good result of iris recognition depends on the selection of the right features and the needed time computation. The use of Haar Wavelet (texture analysis) with three different iris databases showed the best performance, such as MMU (ACC=96.44%, time=0.34s), UTIRIS (ACC=96.21%, time-0.68s), and IITD (ACC=97.14%, time=0.64s).