Science Current Events | Science News |

New Notre Dame research raises questions about iris recognition systems

July 13, 2012

Since the early days of iris recognition technologies, it has been assumed that the iris was a "stable" biometric over a person's lifetime - "one enrollment for life." However, new research from University of Notre Dame researchers has found that iris biometric enrollment is susceptible to an aging process that causes recognition performance to degrade slowly over time.

"The biometric community has long accepted that there is no 'template aging effect' for iris recognition, meaning that once you are enrolled in an iris recognition system, your chances of experiencing a false non-match error remain constant over time," Kevin Bowyer, Notre Dame's Schubmel-Prein Family Chair in Computer Science and Engineering, said. "This was sometimes expressed as 'a single enrollment for life.' Our experimental results show that, in fact, the false non-match rate increases over time, which means that the single enrollment for life idea is wrong.

"The false match rate is how often the system says that two images are a match when in truth they are from different persons. The false non-match rate is how often the system says that two images are not a match when in truth they are from the same person."

Bowyer noted that there are several reasons the misconceptions about iris biometric stability has persisted.

"One reason is that because it was believed from the early days of iris recognition that there was no template aging effect, nobody bothered to look for the effect," he said. "Also, only recently have research groups had access to image datasets acquired for the same people over a period of several years. Recently, another biometric research group (from Clarkson University and West Virginia University) has also published a study that finds an iris template aging effect."

In their study, Bowyer and Notre Dame undergraduate Sam Fenker analyzed a large dataset with more images acquired over a longer period of time. For one group of people in their dataset, they were able to analyze a year-to-year change over three successive years.

Bowyer points out that iris recognition is already used in various airports and border crossings, including London airports, Schiphol 9Amsterdam) airport, and border entry in the United Arab Emirates. And probably the highest profile and largest application of iris biometrics currently underway is the Unique ID program in India, which has enrolled more people that live in the United Kingdom.

Despite the results of the study, Bowyer does not see them as a "negative" for iris recognition technologies.

"I do not see this as a major problem for security systems going forward," he said. "Once you have admitted that there is a template effect and have set up your system to handle it appropriately in some way, it is no longer a big deal. One possibility is setting up a reenrollment interval. Another possibility is some type of 'rolling re-enrollment,' in which a person is automatically re-enrolled each time they are recognized. And, in the long run, researchers may develop new approaches that are 'aging-resistant.' The iris template aging effect will only be a problem for those who for some reason refuse to believe that it exists."

Bowyer and Fenker recently presented their research paper at the IEEE Computer Society Biometrics Workshop.

University of Notre Dame

Related Iris Recognition Current Events and Iris Recognition News Articles

New NIST biometric data standard adds DNA, footmarks and enhanced fingerprint descriptions
The National Institute of Standards and Technology (NIST) published a revised biometric standard in November, 2011, that vastly expands the type and amount of information that forensic scientists can share across their international networks to identify victims or solve crimes.

Twins are intriguing research subjects for Notre Dame biometircs researchers
Each year in August, the aptly named town of Twinsburg, Ohio, is the site of the largest official gathering of twins in the world. Open to all multiples - identical and fraternal twins, triplets and quads from newborns to octogenarians - the weekend's events include food, live entertainment, a golf tournament, and a twins' parade.

Light at the speed of a bicycle and much more
The speed of light, 300 million metres per second, was long thought an immutable constant and has defined our understanding of matter and energy but recent research in the area of optics and photonics is proving that we can manipulate light to some ingenious and hugely lucrative ends.

Knobbly kneed ID
Forget LED thumb-pad identification devices, complex retinal laser scanning, or even computerized iris recognition, the way forward for biometric validation is a quick X-ray snapshot of a person's knees, according to a report published in the International Journal of Biometrics.

First large-scale evaluation of iris recognition under way
The National Institute of Standards and Technology (NIST) has announced that it is running the Iris Challenge Evaluation (ICE), the first large-scale evaluation of iris recognition.
More Iris Recognition Current Events and Iris Recognition News Articles

Handbook of Iris Recognition (Advances in Computer Vision and Pattern Recognition)

Handbook of Iris Recognition (Advances in Computer Vision and Pattern Recognition)
by Mark J. Burge (Editor), Kevin W. Bowyer (Editor)

This authoritative collection introduces the reader to the state of the art in iris recognition technology. Topics and features: with a Foreword by the “father of iris recognition,” Professor John Daugman of Cambridge University; presents work from an international selection of preeminent researchers, reflecting the uses of iris recognition in many different social contexts; provides viewpoints from researchers in government, industry and academia, highlighting how iris recognition is both a thriving industry and an active research area; surveys previous developments in the field, and covers topics ranging from the low-level (e.g., physics of iris image acquisition) to the high level (e.g., alternative non-Daugman approaches to iris matching); introduces many active and open areas of...

Iris Recognition

Iris Recognition
by Zyad Thalji (Author)

The purpose of this Project is to develop Human Identification system that avoids high computational costs, features extraction and curse of dimensionality. The developed system can make a great impact in law enforcement, surveillance, and commercial application such as ATM card verification. This project provides a new methods for Iris recognition with color image; first method separated color image in three gray scale images red value ,green value, and blue value, Second method determined the center pupil and outer circle, the third method is the new algorithm avoid eyelashes and eyelid by taking a sub Iris image, fourth method merge the three sub image in one array, finally use Fourier Fast Transfer (FFT) with markov model to detected Iris print.

Iris Analysis for Biometric Recognition Systems (SpringerBriefs in Applied Sciences and Technology)

Iris Analysis for Biometric Recognition Systems (SpringerBriefs in Applied Sciences and Technology)
by Rajesh M. Bodade (Author), Sanjay Talbar (Author)

The book presents three most significant areas in Biometrics and Pattern Recognition. A step-by-step approach for design and implementation of Dual Tree Complex Wavelet Transform (DTCWT) plus Rotated Complex Wavelet Filters (RCWF) is discussed in detail. In addition to the above, the book provides detailed analysis of iris images and two methods of iris segmentation. It also discusses simplified study of some subspace-based methods and distance measures for iris recognition backed by empirical studies and statistical success verifications.

Iris Recognition System: Implementation of iris recognition system with different circle algorithm in iris boundary localization.

Iris Recognition System: Implementation of iris recognition system with different circle algorithm in iris boundary localization.
by Md. Hedayetul Islam Shovon (Author), Mohammad Munzurul Islam (Author)

In biometrics, one of the most important type of physical identification that is based on the personal and unique characteristics of the iris - the colored ring around the pupil of an eye.The whole iris recognition system using Dougman’s method is implemented here with the extended implementation of the system with four circle algorithm in iris segmentation stage to make an iris recognition system with efficient use. For efficient iris recognition system we need the best tools in every stage it encountered. This book concentrates on iris localization for that reason. Because if it smooth the way of the segmentation stage for iris recognition, the whole system’s output will be successful.

Iris recognition system: Digital image processing

Iris recognition system: Digital image processing
by Younus Javed (Author), Sadia Minhas (Author)

The iris recognition system utilizes image processing and computer vision in order to identify human beings. Most effective algorithms are employed to gather suitable patterns from an iris image. Then, mathematical analysis is carried out for collecting required features using efficient image enhancement techniques and feature extraction methodologies. The extracted features through six different schemes are compared exhaustively for matching with the iris images of individuals already stored in the database. The low cost of development of the proposed iris recognition system encourages its application in majority of the organizations and establishments. The experimental work presented in this book provides detailed and critical analysis using three experiments. Various components of the...

Iris Recognition: Matrix Index Distance Method: Basic, Solving Procedure and Discussion

Iris Recognition: Matrix Index Distance Method: Basic, Solving Procedure and Discussion
by Shubhashis Kumar Shil (Author), Iftekhar Ibney Gias Remon (Author)

Iris recognition has received increasing attention recently in the world. A biometric system provides automatic identification of an individual based on a unique feature or characteristic possessed by the individual. Iris recognition is regarded as themost reliable and accurate biometric identification system available. Most commercialiris recognition systems use patented algorithms developed by Daugman, and these algorithms are able to produce perfect recognition rates. However, published results have usually been produced under favorable conditions, and there have been no independent trials of the technology. This research presents an iris recognition system in order to verify both theuniqueness of the human iris and also its performance as a biometric identification andalso focuses on...

Iris Recognition In Eigen Space: A Book On Image Processing

Iris Recognition In Eigen Space: A Book On Image Processing
by Abhijit Das (Author)

This work proposes novel approaches for iris recognition in the context of biometric identification of a person. The iris is a portion of the inner eye of an individual and contains an abstract and randomly generated texture pattern arising from orientation of complex tissues within this region. This random pattern can provide a unique identifier of a person if a mathematical model can be built to represent and compare it. The manuscript consist of two approaches .In first approach the iris images are mapped to Eigen-space and a robust iris code signature is generated from different camera snapshots of the same eye to incorporate tonal and lighting variations. The covariance matrix of the images is computed and they are converted to a row vector, to enable real-time identification the...

Iris Recognition: An Identification Biometric system: Make your own biometric system using human iris

Iris Recognition: An Identification Biometric system: Make your own biometric system using human iris
by Abdul Basit (Author)

This book reviews different methods used for iris recognition system. Special emphasis is given to most difficult step in iris biometrics system i.e. iris segmentation. Different methods for four different available databases have been given in detail. Figures will give readers a complete understanding of the concept. Further more, instead of phase, identification is based on bit-plane slicing of the normalized iris.

Iris Recognition Based New Technique for Person Identification: DTCWT Based Feature Extraction

Iris Recognition Based New Technique for Person Identification: DTCWT Based Feature Extraction
by S P Narote (Author), R S Holambe (Author), L M Waghmare (Author)

This book explores the use of iris recognition for person identification. Description of different methods of iris recognition is presented and discussed. The image acquisition system is developed and the benchmark iris databases, SCOE-Iris v1.0 have been explained. New methods for iris liveness detection are presented. In iris images the irregular boundaries of pupil are segmented using histogram and edge detection based method. The detail experimentations were carried out using different mother wavelets: Haar, Daubechies, Coiflet, Symlet and Biorthogonal. In addition a new robust algorithm for iris feature extraction utilizing dual tree complex wavelet transform has been presented and evaluated in terms of its speed, accuracy and efficiency. The technique developed ...

Iris Recognition: A Machine Learning Approach

Iris Recognition: A Machine Learning Approach
by Kaushik Roy (Author), Prabir Bhattacharya (Author)

The increasing demand for enhanced security in the daily life has directed the improvement of the reliable and intelligent personal identification system based on biometrics. Iris recognition has been regarded as one of the most reliable biometrics technologies in recent years. In this book, an iris recognition scheme is presented as a biometrically based technology for person identification using Multi-Class Support Vector Machines (SVM). The Multi- Objectives Genetic Algorithms (MOGA) is used to select the most significant features in order to increase the matching accuracy. The traditional SVM is modified into an asymmetrical SVM to treat the cases of False Accept and False Reject differently and also to control the unbalanced data of a specific class with respect to the...

© 2016