Date of Publication :30th September 2020

Abstract: Biometric is defined as study of various methods for measurement of physiological and behavioral characteristics that can be considered to identify a person. Biometric identification of a person gained major importance in the world from its application such as access control and security. The iris recognition is the process of recognition of individual by analyzing random pattern of iris. As compared to several other biometrics, iris recognition system is believed to be more reliable, accurate and scalable for person identification. Iris recognition is one of the booming biometric modalities due to its unique characteristics. The iris structure from human eye can be used for biometric authentication and identification at reduced resolution, iris under uncontrolled illumination, iris at a distance, iris off axis, presence of eyelashes, low accuracy. These type of visible wavelength based iris recognition system eliminate the limitation of iris recognition system that require close range iris imaging under infrared illumination which can be hazardous. I prefer image processing technique for overcoming these difficulties. The challenges emerge when the iris images acquired in one domain is matched against the images acquired in different domain. Such cross-domain iris recognition problem includes the cases when the images in one domain represent the sensor-specific iris images or wavelength- specific iris images. Here a new class of bi-spectral iris recognition system that can simultaneously acquire visible and near infra- red images with pixel-to-pixel correspondences is proposed and evaluated. This paper describes the approaches used by other research group around the world in related area. It also presents a brief overview of digital image processing techniques such as image segmentation, normalization, feature extraction, image restoration and image enhancement.

Reference :