Browse Prior Art Database

An Efficient and Robust Iris Recognition System for Access Control

IP.com Disclosure Number: IPCOM000175267D
Published in the IP.com Journal: Volume 8 Issue 10B (2008-10-23)
Included in the Prior Art Database: 2008-Oct-23
Document File: 2 page(s) / 285K

Publishing Venue

Siemens

Related People

Juergen Carstens: CONTACT

Abstract

In the field of biometrics and personal identification, iris recognition is one of the major methods used. There are many technical solutions for the design of iris recognition systems. In general, the iris recognition can be divided into three main steps: At first, image acquisition needs to be performed which has great impact on the image quality. Secondly, feature extraction is executed which comprises different local steps such as localization, segmentation, normalization and feature encoding. Lastly, the recognition and identification needs to be done. Each of the above mentioned steps involves further micro steps that eventually are responsible for making the iris recognition system successful and robust. In Figure 1, a block diagram depicts the main steps of the iris recognition system. According to this figure, the iris recognition system is divided into three steps; each step contains its own technical challenges influencing the performance of the system. If the acquisition captures images of good quality, the performance of the second step will be improved. In this second step, the most important micro step is the iris segmentation. Again, if the iris segmentation part is successfully executed, the verification/identification will return good results.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 52% of the total text.

Page 1 of 2

An Efficient and Robust Iris Recognition System for Access Control

Idea: Santosh Singh, IN-Bangalore

In the field of biometrics and personal identification, iris recognition is one of the major methods used. There are many technical solutions for the design of iris recognition systems. In general, the iris recognition can be divided into three main steps: At first, image acquisition needs to be performed which has great impact on the image quality. Secondly, feature extraction is executed which comprises different local steps such as localization, segmentation, normalization and feature encoding. Lastly, the recognition and identification needs to be done.

Each of the above mentioned steps involves further micro steps that eventually are responsible for making the iris recognition system successful and robust. In Figure 1, a block diagram depicts the main steps of the iris recognition system. According to this figure, the iris recognition system is divided into three steps; each step contains its own technical challenges influencing the performance of the system. If the acquisition captures images of good quality, the performance of the second step will be improved. In this second step, the most important micro step is the iris segmentation. Again, if the iris segmentation part is successfully executed, the verification/identification will return good results.

The image processing hardware mainly dominates the technical challenges involved with image acquisition of the eye. The other steps are managed by the image processing software. Based on the available knowledge, it is difficult to quantify the performance of the available iris recognition system in the market. However, based on some experiments it has been observed that the iris segmentation is one of the most difficult steps. Once the segmentation fails, the whole performance of the iris recognition system will fail.

In the following, an iris recognition system is proposed that avoids the complicated iris segmentation step. The flow graph depicted in Figure 2 reflects the complete iris recognition system in which most of the intermediate steps are shown. With this iris recognition system, the segmentation of the iris region is avoided.

As a first step, a sequence of inp...