Introductory Chapter: On Biometrics with Iris

*Muhammad Sarfraz and Nourah Alfialy*

### **1. Introduction**

Biometrics is the systematic study of measuring and analyzing biological data for the purpose of validation or identification. Biometrics refers to specific physiological and/or behavioral (extrinsic and/or intrinsic respectively) characteristics that are uniquely related to a person [1, 2]. The biometric systems use unique human physiological and anatomical properties to define details. Such systems effectively help to overcome the security issues affecting the conventional methods of personal authentication. In the smart world today, the importance of technological solutions in biometrics is growing. Specifically, automation culture has desired to design and launch automated systems for highly reliable and accurate human authentication and identification.

Biometrics has been deployed successfully in various fields of real life. Numerous methods, techniques, and systems on biometrics serve sciences, security, military, medical area, and human identification. There are various kinds of biometrics being used, these include Fingerprint, Face, Speaker/Voice, Infrared thermogram (facial, hand, or hand vein), Gait, Keystroke, Odor, Ear, Hand geometry, Retina, Iris, Palmprint, Signature, DNA, Knuckle crease, Brain/EEG, Heart sound/ECG. Specifically, in the past decade, iris recognition technology has become the most popular biometric technology for human authentication and recognition due to its stability and uniqueness in its structure. The iris has a unique structure that remains stable throughout a person's life. Iris recognition is one of the authentication methods that uses high-resolution assisted pattern recognition technology. The general method of the iris recognition system includes image acquisition, segmentation, feature extraction, matching, and classification [3].

Iris has become a very effective recording of its superior properties, such as reliability and accuracy. In recent years, a good amount of research is made regarding the evolution of biometric-based on the iris. This presented iris recognition as a very clear and effective concept. There is a need to highlight and analyze the work done by different authors related to iris studies, methodologies, and practices. A detailed comparative study could particularly provide an overview for the readers.

The idea of iris recognition goes back to an eighteenth-century Paris prison, where police distinguished criminals by examining the color of their irises. Daugman [4] was the first to develop the basic algorithms that now form the basis of all current commercial iris recognition systems, having been commissioned by Flom and Safir [5, 6] to conduct extensive and comprehensive research to implement automatic iris recognition. In 1987, Flom and Safir acquired a non-applied concept of an automated iris biometric system. A report was published by Johnston in 1992 without any experimental results [5, 6].

The motivation behind this work is to study biometric iris recognition specifically because it provides one of the most stable biometric signals to recognize distinct tissues that form prematurely and remain constant throughout life unless there is an eye injury.

The aim of this chapter is to contribute in a comprehensive survey about the difference between the existing biometrics techniques. An important aspect of biometric technology is to evaluate its performance. The performance of any biometric authentication technology can be measured by various parameters. Compared to other vital features, such as the face, fingerprint, and voice, the iris patterns are more stable and reliable [7–9]. The reason behind this is that iris recognition algorithms require pre-processing of the input image to obtain better data quality by tracking different feature points of the iris. Biometrics using a feature is so unique that the chance of any two people having the same features is very rare [10]. Identification of a person based on recognition of the iris of the eye gives one of the most reliable results. Iris tissue features provide unique high-dimensional information that explains why iris recognition-based verification has the lowest false acceptance rate among all types of identity verification systems [1, 11–13].

Iris recognition has been used in many countries with the purpose of identifying millions of people around the world. This technology is comfortable to use and difficult to rig. Many authentication programs, including border crossings without a passport, national identity, etc., have adopted this technology for its benefits [14]. For the purpose of human recognition, the iris biometric recognition system has proven its importance. The biometric iris recognition systems are easy to use and create a hassle-free security environment. Iris scanners can be used to protect high-value websites by blocking the access of unwanted visitors. Commercial and governmental institutions in all fields have recognized the benefits of this system and have embarked on implementing validation systems based on iris recognition in a major way [15]. Iris recognition is one of the best-protected methods of authentication and recognition. Iris recognition accuracy is very promising. The false acceptance rate, as well as the rejection rate, is very low. A special grayscale camera is used to take an iris pattern within 10–40 cm from the camera [14, 16–18]. An appropriate methodology is used to define the irises of the image, and if it is present, a grid of curves covering the iris is created and the iris symbol is generated based on the opacity of the points. It is affected by two things—first, the general opacity of the image, and secondly, the changes in the size of the iris. The comparison of two irises can be computed through the knock distance based on the difference in the number of bits and it is very fast [4, 19]. Also, the template matching technique can be used, and it uses statistical calculation to match the stored iris template and the obtained iris template. Iris recognition is applied in the following areas: border control, passports, ID cards, and other government purposes, database access, login authentication, aviation security, hospital security, access control to buildings, areas, homes, and security of restricted prisons [6, 20, 21].

For convenience, it is desired to know the basic concepts and terminologies we are going to use throughout this chapter. There are as follows:


the effect of changing the size of the pupil. Once the outer and inner circles of the iris are localized, these values are taken as input to the Daugman's Rubber-sheet model" [23].


This chapter has been organized in various sections. Section 1 gives a brief introduction about biometric iris recognition, motivation to work in this study, the importance of this study, basic concepts and terminologies to be used, and the organization of this study. Section 2 consists of a literature survey and a comparative study of the existing different methods used in biometric iris recognition. It also gives information about different methods used in the extraction of the features of iris image datasets and data analysis. Finally, Section 3 gives the new directions for the future. It suggests some recommendations for community, government, industry, etc. Then the overall conclusion of the study is done in this chapter. It concludes with the discussion on future trends as well.

### **2. Literature survey**

Although many papers have been published in this field in the past years, twenty-six papers have been selected and presented to understand the iris recognition techniques available in the literature. These articles have shown differences between each other in one way or another. In this chapter, a review is presented focusing on all four phases, i.e., segmentation, normalization, extraction, and template algorithms of the iris recognition technology from Daugman's initial work in 1993 to some recent work.

Daugman [10] developed a feature extraction process based on information from a group of 2D Gabor filter. He created a file 256 bytes by specifying the local phase angle according to outputs of the real and imaginary parts of the filtered image, compare the percentage of mismatched bits between a pair of Iris representations via the XOR operator, and the choice of a separation point in the space of the Hamming distance.

On the contrary, Wildes system took advantage of the Laplacian pyramid, which was built with four different precision levels. Generate the iris symbol [14, 15]. Also, it explained a normalized correlation based on goodness-of-match values and Fisher's linear discriminant for pattern matching. Both iris recognition systems use of bandpass image decompositions to get multi-scale information.

Lim et al. [21] proposed an iris recognition system. It includes a compact representation scheme for iris patterns by the 2D wavelet transform. This method is used for initializing weight vectors and determining winners for recognition in a competitive learning method. Flom and Safir [6] had earned a patent in Iris

Recognition System, which gives a generalized concept in using iris as a biometric system but does not describe any implemented algorithm.

In the process of recognizing the iris of the eye, conversion is necessary. An iris image acquired in a convenient symbol can easily manipulate it. Hence, we will take a quick look at the process of feature extraction and representation of modern wonderful works and papers. Iris recognition is the procedure of comparing known and unknown irises to prove that it is from the same person or not [11].

Today, many approaches, techniques, and systems are used to match iris and solve related problems. This section is focused on analyzing and categorizing different author's work in the iris recognition area. **Table 1** provides a summary of various papers in the current literature. First column determines the Reference of the papers by author names and year of publication. Second column gives the summary of the work in the corresponding paper, and the third column describes the implemented approaches used to solve iris recognition issues. The author names and the year of publication have been used as an identifier for the rest of the tables in the chapter showing other details of the referred literature.

The main point of biometrics technology is to evaluate their performance and accuracy. It can be measured by the various parameters such as False Accept Rate (FAR), False Reject Rate (FRR) and Crossover Rate (CER) or Equal Error Rate (EER). An identity claims wrongly rejected is called False Rejection and a false identity claims wrongly accepted is known as False Acceptance. In order to make limited entry to authorized users FAR and FRR are used. False Rejection Rate (FRR) measures the probability of rejecting an authorized user incorrectly as an invalid user [16].

**Table 2** shows the accuracy and performance in percentage. It also mentions the identification and verification measures. Identification and verification are matching techniques for Iris recognition. In the verification, the person enrolls his Iris to the system and the template is stored in the database. Every time the person accesses the system, he has entered his iris to verify himself. It's a one-to-one relationship where the input Iris is compared with the stored one. On the other hand, identification is one to many relationships because the human Iris is matched with the Irises in 7the database to determine who is that person [11]. While the performance measures used for identification depend on the accuracy, recognition rate, rank K, etc., the performance measures for verification are False Match Rate (FMR), False Non-Match Rate (FNMR), False Accept Rate (FAR), False Rejection Rate (FRR), and Equal Error Rate (EER). The researchers in [3] describe the meaning of the authentication parameters. FAR happens when the system recognizes person erroneous. But when the system rejects entry to approve person that means the FRR is happening. FMR is the amount of fraud assessments with threshold value "T" divided by the total quantity of fraud similarities. FNMR is the quantity with unaffected comparisons with threshold value "T" divided by the total quantity of open comparisons. Last one is EER; it describes the error rate of the system.

In Lim et al. [21], eye images captured at a distance with the help of a CCD camera. Then, in the acquired image, iris is segmented. Initially, it is done by detecting the pupil using the center point detection method followed by edge detection method by finding virtual circles. An analysis was made in the pre-processing stage, with 6000 data to identify the causes of failure at this stage. Data involved images both with Lens, without, and with glasses. In the normalization stage a 45060 bit iris image part was obtained. Gabor transforms and Haar wavelet transforms, which are two different methods, were used to analyze and extract the features from the segmented iris image.






**Table 1.**

*An overview of literature.*

In Daugman [11], proposed an approach that is an improvement to his previous work. This approach is working with the noise disturbances that occur while acquiring an iris image of a human eye. Also, an algorithm was introduced for detecting the eyelids, which involves arcuate edges with spline parameter, instead of circular edges in the Integro differential operator.

In Wilde [14, 15] tried a different approach, in which inner and outer iris boundary is computed with the help of a gradient-based binary edge map followed by circular



#### **Table 2.**

*An overview of literature for Accuracy and performance.*

Hough transform. Wilde used around 60 human irises captured from 40 subjects in his experiment. Also, he has done a comparative study with Daugman's work in his paper. This method proved to provide higher accuracy rate when tested in CASIA database.

These algorithms can provide rotation, translation and size invariant result. Simulation results of these algorithm prove to provide a higher correct accept and reject rate. Results were tested using CASIA database, UBIRIS, UPOL, MMU and a database provided by Institute of Automation for 2005 Biometrics Authentication competition.


### *Introductory Chapter: On Biometrics with Iris DOI: http://dx.doi.org/10.5772/intechopen.105134*


#### **Table 3.**

*An overview of literature for data used.*

The experimental parts of the author's [1–25] are shown in **Table 3**. It explains the type of applications and kind of Databases used. Then, it shows the number of data used in the study.

There is a need to overlook for the data images together with their resolution and format. **Table 4** describes the number of data used in the application, the number of images resulted, their resolutions and formats.

In [10], Daugman described how iris recognition is being used to check visitors coming to the United Arab Emirates (UAE) against a watch-list of people who are denied entry to this country. The UAE database contains around 632,500 different iris images. In all comparison, no false matches were found with Hamming distances below about 0.26. Daugman reports that "to date, some 47,000 persons have been caught trying to enter the UAE under false travel documents, by this iris recognition


### *Introductory Chapter: On Biometrics with Iris DOI: http://dx.doi.org/10.5772/intechopen.105134*


#### **Table 4.**

*An overview of literature for data images with resolution and format.*




### *Introductory Chapter: On Biometrics with Iris DOI: http://dx.doi.org/10.5772/intechopen.105134*



#### **Table 5.**

*An overview of literature for their applications and advantageous features.*

system" [4, 17]. There are similar reports for various kinds of applications and methodologies. **Table 5** describes the implemented application type and the reason for using it by mentioning the advantages and disadvantages of the proposed methods.

### **3. Conclusion**

Biometrics means the automatic identification of a person based on his behavioral and/or physiological unique characteristics. Iris biometrics is an efficient, safe, costeffective, easy-to-use technique for identity verification. This study provides detailed information related to iris recognition techniques. Several author's works, related to iris recognition technology, are discussed, compared and analyzed. A detailed analysis of various studies is made. Various methods are taken into account to extract features of the iris such as wavelet beam analysis and static measurement feature transformation. The main focus is on iris as biometrics feature for the secure authentication and uniqueness of human identification around the world. The iris is one of the biophysiological features that are very reliable in identification systems. It is used in multimodal biometrics and in conjunction with cryptography. It is also considered one of the fairest biometrics of the face. However, it has been found that the localization of the iris is affected by tissue. When not properly interpreted, commercial iris-based biometrics systems provide inaccurate results while identifying humans. Moreover, it is important that the iris-based identification systems work with both ideal and imperfect iris images, otherwise safety will be at stake.

As a future work, there is a scope to improve the problems related to iris recognition, specially, the issues related to the capturing Iris by the sensors. One of the innovations is the touchless Iris sensors, which will be sufficient for various difficult situations including COVID-19 in the current time and age. It will decree the need to touch the devices. This technique is needed to show its reliability and efficacy as an alternative to regular sensors. Relying on an iris recognition in a different government domain is also recommended. Implementing iris recognition technology is not only useful for Government, but other organizations and communities can also think and may benefit by applying iris recognition techniques to identify and verify.

It has also been noted that iris-based biometric systems tend to present erroneous results in uncooperative settings. Another important idea is that the iris can be used for mobile phone communications with smart devices. Revocable biometrics is useful for strong security in the event of attacks. There are direct and indirect attacks on multimodal biometrics that must be overcome. More research is needed to know that attacks like these cannot break the security of biometric systems. With these ideas in mind, in the future, people can focus on designing ATMs with iris recognition in the banking industry.

There is also a need of the time to concentrate on using real apps to support the generation of tiny iris codes for cell phones and PDAs. In this chapter, an attempt is made to provide an insight into different iris recognition methods. Technology survey provides a platform for developing new technologies in this field as a future work.

## **Author details**

Muhammad Sarfraz\* and Nourah Alfialy Department of Information Science, College of Life Sciences, Kuwait University, Sabah AlSalem University City, Shadadiya, Kuwait

\*Address all correspondence to: prof.m.sarfraz@gmail.com

© 2022 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

*Introductory Chapter: On Biometrics with Iris DOI: http://dx.doi.org/10.5772/intechopen.105134*

### **References**

[1] Roy S, Bandyopadhyay SK. Iris recognition using biometric techniques. International Journal of Applied Research 2017;**3**(5):822-823

[2] El-Abed M, Giot R, Hemery B, Rosenberger C. Evaluation of biometric systems: A study of users' acceptance and satisfaction. International Journal of Biometrics, Inderscience. 2012. pp.1-27. 10.1504/IJBM.2012.047644. hal-00984024

[3] Choudhary D, Tiwari S, Singh AK. A survey: Feature extraction methods for iris recognition. International Journal of Electronics Communication and Computer Technology. 2012;**2**(6): 275-279

[4] Daugman JG. High confidence recognition of persons by a test of statistical independence. IEEE Transaction on PAMI. 1993;**15**(11): 1148-1161

[5] Rakesh T, Khogare MG. Survey of biometric recognition system for Iris. International Journal of Emerging Technology and Advanced Engineering. 2012;**2**(6):272-276

[6] Bowyer KW, Hollingsworth K, Flynn PJ. Image understanding for iris biometrics, a survey. Computer Vision and Image Understanding. 2008;**110**(2): 281-307

[7] Jin-Hyuk H, Eun-Kyung Y, Sung-Bae C. A review of performance evaluation for biometrics systems. International Journal of Image and Graphics. 2005;**5** (3):501-536

[8] Manisha KN. Cancelable Biometrics: A comprehensive survey 2020;**53**:3403– 3446. DOI: 10.1007/s10462-019-09767-8 [9] Himanshu S. Personal identification using iris recognition system, A Review. International Journal of Engineering and Applications (IJERA). 2013;**3**:449-453

[10] Daugman J, Dowing C. Epigenetic randomness, complexity, and singularity of human iris patterns. Biological Sciences. 2001:1737-1740

[11] Daugman J. How Iris recognition works. IEEE Transaction on Circuits and Systems for Video Technology. 2004; **14**(1):21-30

[12] Bramhananda Reddy MV, Goutham V. Iris Technology: A review on iris based biometric systems for unique human identification. International Journal of Research— Granthaalayah. 2018;**6**(1):80-90. DOI: 10.5281/zenodo.1162210

[13] Phadke S. The importance of a biometric authentication system. The SIJ Transactions on Computer Science Engineering & its Applications (CSEA). 2013;**1**(4):128-132

[14] Wildes R, Asmuth J, Green G, Hsu S, Kolczynski R, Matey J, et al. A system for automated iris recognition. Sarasota, FL: Proceedings IEEE Workshop on Applications of Computer Vision; 1994. pp. 121-128

[15] Wildes R. Iris recognition: An emerging biometric technology. Proceedings of the IEEE. 1997;**85**(9): 1348-1363

[16] Sabhanayagam T, Prasanna VV, Senthamaraikannan K. A comprehensive survey on various biometric systems. International Journal of Applied Engineering Research. 2018;**13**(5): 2276-2297

[17] Daugman. Recognizing persons by their Iris patterns. In: Biometrics: Personal Identification in Networked Society. Kluwer; 1998. pp. 103-121

[18] Sheela SV, Vijaya PA. Iris recognition methods survey. IJCA. 2010; **3**(5):19-25

[19] Ganorkar SR, Ghatol AA. Iris recognition: An emerging biometric technology. In: Proceedings of international Conference ICSCI 2007. Hyderabad; 2007. pp. 596-600

[20] Sanjay R, Ganorkar AA, Ghatol. Iris recognition, an emerging biometric technology. In: Proceedings of 6th WSEAS International Conference on Signal Processing, Robotics and Automation, Greece. 2007. pp. 91-96

[21] Lim S, Lee K, Byeon O, Kim T. Efficient iris recognition through improvement of feature vector and classifier. ETRI Journal. 2001;**23**(2): 61-70

[22] Anil JK, Ross A, Prabhakar S. An introduction to biometric recognition. IEEE Transactions on Circuits and Systems for Video Technology. 2004; **14**(1):4-20

[23] Arrawatia S, Kishore B, Mitra P. Critical literature survey on iris biometric recognition. International Journal of Scientific Research in Science and Technology IJSRST. 2017;**3**(6): 600-605

[24] Rui Z, Yan Z. A survey on biometric authentication: Toward secure and privacy-preserving identification. IEEE Access. December 2018;**99**:1-1. DOI: 10.1109/ACCESS.2018.2889996

[25] Williams G. IRIS recognition technology. IEEE AES System Magazine. 1997:23-29

### **Chapter 2**
