Literature review: Iris Segmentation Approaches for Iris Recognition Systems

Ms. Sruthi

چکیده

INTRODUCTION In the recent years, drastic improvements have been accomplished in the areas like iris recognition, automated iris segmentation, edge detection, boundary detection etc. Iris recognition is a biometric recognition technology that utilizes the pattern recognition techniques based on the high quality images of iris. An iris recognition system mainly uses infrared or else the visible light. The systems which are based on near infrared light (NIR) are very common because NIR will not generate reflections that makes iris recognition complex. But, NIR images lacks pigment coloration information, thus recognition algorithms must entirely depend on the patterns that are unrelated to color. Iris recognition which is based on visible light raise up pigmentation, thus the recognition systems will be able to exploit the color patterns, which make identification much easier. This is because; the pigmentation patterns contain lots of information that can be utilized for recognition. But the visible light reflections in these types of systems can result in a extensive amount of noise in the gathered images. Typical iris recognition system consists of mainly three modules. They are image acquisition, pre-processing stage as well as feature extraction and encoding. Image acquisition is a module which involves the capturing of iris images with the help of sensors. Pre-processing module provides the determination of the boundary of iris within the eye image, and then extracts the iris portion from the image in order to facilitate its processing. It involves the stages like iris segmentation, iris normalization, image enhancement etc. The performance of the system has been analyzed in the feature extraction and encoding stage. All these stages involve their own developments. Edge detection systems like boundary as well as the edge detection have been provided with much advancement. Also, major improvements have been made in the field of iris segmentation, which is very much important in the area of iris recognition.. In this paper, some of the methods involving iris segmentation have been analyzed. Approaches like Integrodifferential operator [1], Hough transform [2] constitute a major part in the iris recognition techniques. Many other iris recognition as well as the iris segmentation approaches [3,4,5,6,7,8] has been introduced. Each one has its own advantage as well as the disadvantage. Major drawbacks involved in all the papers is that, the segmentation accuracy has not been achieved yet. Thus, there is a strong need to develop a new segmentation approach that is more reliable as …