Publication

Abstract : Background/Objectives: Providing security to systems is one of the major challenges faced in day-to-day life. Biometrics plays a vital role in ensuring security. Out of the different existing recognition systems available - namely face, finger, gait, retina and so on, sclera recognition system gives out better performance. Sclera is the white part of the eye, which is unique and consistent in nature because of which it is chosen for recognition. In this paper, we analyzed the existing sclera recognition system with both human and animal eye images. Methods/Statistical Analysis: In this paper, we compared the performance of the algorithm with both human and animal eye images. The animals we considered for the algorithm analysis include deer, buffalo and lion. Human eyes are the most observable due to the presence of more sclera area. The blood vessel patterns present in the sclera region are stable over lifetime and unique person by person, thus making it appropriate for identification. The results show that the algorithm works better for human eye. Results/Findings: In this paper, we compared the extracted input vessel structure with that in the database and verify whether the input is authorized or not. It is observed that the system underperforms for animal eye images, whereas it shows an acceptable performance with human eye images. Conclusion/Application: The results prove that the algorithm works satisfactorily for identifying human. It can be incorporated with other recognition systems to build up a multimodal biometric system that can provide better security than the existing systems.