Matlab Programming on Iris Biometric Security Using Feature Extraction Process

doi:10.21203/rs.3.rs-2215161/v1

Download PDF

Research Article

Matlab Programming on Iris Biometric Security Using Feature Extraction Process

https://doi.org/10.21203/rs.3.rs-2215161/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

A Biometric system provides automatic identification of an individual based on a unique feature or characteristics possessed by the individual. Iris recognition is one of the important biometrics recognition system that identify people based on the iris. Among, the biometric system iris recognition is the best and most reliable and accurate. The aim is to identify a person in real time, with high efficiency and accuracy. In this paper, iris recognition algorithm is implemented via; hybrid techniques and feature extraction process by using MATLAB coding. In hybrid technique, iris recognition is implemented using histogram equalization, wavelet techniques and some of the best edge detection, feature extraction. And Extracting a shape and texture from Characterize images using the feature extraction process.

Iris Recognition

Biometric Identification

Hybrid techniques

wavelet techniques and Feature extraction process

Today’s e-security are in critical to finding accurate, secure and cost-effective alternatives to passwords, personal identification number (PINS) as financial loss increase dramatically year by year. There is no security process in computer-based fraud such as computer hacking and identify theft. To overcome this problem the Biometrics identification system takes a best place in e-security. Because, Biometric solutions only gives the unique characteristics of an individual person that cannot be transferred.

The Biometrics which refers to identifying an individual by his or her physiological (or) behavioral characteristics is capability to distinguish between authorized user and an imposter. so, that only biometric identification systems are most reliable, capable and accurate and it have become a role of the large and effective in the area of information technology. And commonly used biometric features include Iris, face, finger print, speech, voice, hand geometry, retinal identification and body order identification.

IRIS Recognition is a method of biometric authentication, based on extraction of an iris of individual eyes. Each individual has a unique iris, and the variation exists even identical twins and between the left and right eye of the same person.so, that is the reason behind the iris recognition it gives one of the most reliable results. Iris texture and shape feature provides a unique high dimensional information that explains why iris recognition-based verification has a lowest false acceptance rate among all types of biometric verification system.

Iris recognition is most promising biometric system. The iris recognition is that the high precision of the identification and verification in among other biometrics. Because it gives 99% of result comparing with other biometrics and it has the less false acceptance rate. The success of the iris recognition depends mainly on two factors.1. Image acquisition and 2. Iris recognition algorithm. The iris is the thin circular diaphragm, which lies between the cornea and the lens of the human eye. Therefore, function of the iris is to regulate the light which enters into the pupil and this will be done by sphincter and the dilator muscles, which is only adjust the size of the muscle. The typical diameter of the iris is 12mm, pupil size is varying from 10–80% of the iris diameter and, the iris is in color of brown and blue, rings the dark pupil of the human eye.

The amount of data which will be measured during a single iris is far greater than the other biometrics such as fingerprint, face, voice, speech recognition etc. Henceforth, accuracy of iris is much greater than DNA. The iris recognition system is considered as an accurate identification technology because of its stableness. The iris patterns completely shaped and continue to be identical also does not change throughout for lifetime. The iris recognition system is most promising and less false acceptance rate because the same person’ left and right irises are different so it gives the most security and high dimensional of accurate property. Therefore, false acceptance rate of the iris is 1 in 1.2 million, statically, better than the average of fingerprint. And the iris scanning is very fast and false acceptance is very less and it does not take longer than one or two seconds. And the use of the contact lens and glasses has little effect on the representation of the iris and hence it does not interfere with the recognition system.

Image processing technique is used to extract the unique iris pattern from a digitized image of the eye, and encode it into a biometric template, which can be stored in the database. When an individual wish to identified by iris recognition system, their eye is first photographed and then template created for their iris region. This template is then compared with the stored template which is within database until both an identical template is found and therefore the person is identified else it's unidentified.

The iris is high protected and non-invasive and ideal for handling applications requiring management of large user groups. Although small and some problematic to an image, where the iris has a greater mathematical advantage of its pattern variation among different persons.

The preprocessing is the common name for operations with images at a lowest level of abstraction both input and output of the intensity images. Preprocessing is a method mainly given the enhanced output of the input image. The development of the image is done by preprocessing method without the distortions. In this step is performed to minimize noise and resizing the original image. Also, the image gets smoothed after this process. Preprocessing which includes many steps for an example noise removal, converting into a gray scale image, normalization, localization and etc. Image preprocessing methods uses to considerably redundancy in images reduces the noise and improve (or) increase number of the pixels of the dataset.

In this paper the preprocessing includes the converting into grayscale image from the raw image and noise removal of the input image.

Converting into gray scale image

In real time the iris images consist of primary colors it is very difficult to get the result and it has complex computational process as well. Because primary colors have its own property and if we concentrate on that, the process goes too big and difficult and it takes much time. To overcome, this issue the gray scale images has been uses in the preprocessing method. Fig.1 shows converting an real image into a gray image, it gives the range of shades of gray without apparent color. This method is very much important in the preprocessing because the computational step of the gray scale image is simple and easy to use for further process in image processing. After the conversion the image consists only two colors (Black and white). Where it may represent in the form of binary (0 and 1) and decimal (0 to 255) values.

The gray scale image is depending on the amount of primary colors involving in the input image. And output of gray image is equal to the input of the primary image. so, there is no more information is loss in this method. Only the image is converted into a gray scale image.

The black color is represented in the gray scale is the darkest possible shade, which is the total absence of transmitted or reflected light and the white color is defined in form of the light possible shade, which is total presence of transmitted or reflected light.

Comparing with primary colors the black and white is represented by R=G=B = 0 or R=G=B =00000000 and R=G=B =1 or R=G=B =11111111 for a 8bit gray scale image. This method is also known as black and white image.

The three main parameters are defined in the gray scale image that is saturation, hue, brightness. In each pixel the saturation and hue are equal to 0 but the brightness is only the parameter can vary from min of 0(black) to max of 1(white).

The 2dimensional discrete wavelet transform is employed to extract the required features or information from the improved image. the most purpose of the 2D DWT is to pick and extract the important features from image. In Fig.3 the feature vector is made which consists of the ordered sequences of the features extracted from the varied representation of the iris image; which can do by the wavelet technique. We applied 2D DWT technique during this and there are many wavelet masks, additionally I used just one type which is harr wavelet transform masks which provides the adequate leads to this method. Using the wave menu command in MATLAB function the improved image is given the output image by using the harr masks with the extent of 4. Another important uses of the 2D DWT is to take the (LL) sub band for the storage process.

The histogram equalization is the method of technique for adjusting image intensities to enhance contrast and there is no theory behind this. Image enhancement is a technique in image processing that uses in preparing an image for a particular application. The resultant of this process gives out the high-performance technique in one application might not be useful in another application. Generally, there are two approaches in image enhancement technique that is spatial and frequency domain.

In this paper, only the very common technique HE was used. It enhances overall contrast of an image by transforming an original image into a uniform histogram. The HE is used to enhance the image and gives the output image in the form of graphical representation. From this enhancement we can get some accurate portion of the image.

In modern days, the field of digital image processing the image enhancement plays a vital role. Therefore, a several enhancement techniques has been proposed and used in image processing to improve the quality of the image. Up to date, there is no best method is to enhance the image than the histogram equalization. This method is very simple and effective method among other enhancement method. This work based on the histogram equalization in fig.4 , the output of the images explains the detail information of the iris image. Where the stretched iris image which gives the better resolution than normal iris image. It will helpful for the further process and we can easily identify the difference in particular and specified image.

The feature extraction is the process which gives the better and accurate process to extract the features from the images in fig5. These details were helpful to the persons secured information. And this process is best process in image processing it gives the better resolution and accuracy. Especially extracting a texture and shape image from this method is very much helpful for the security systems. Because the shape of the iris does not change throughout our lifetime. From this the false acceptance rate of the iris recognition is very less and the security is better other than the biometrics.

The image boundaries are carried out discontinuous in brightness by Edge detection technique. Most commonly edge detection is used for image segmentation and data extraction in area such as computer vision, machine vision, image processing the result and output are in the form of binary. The edge detection is processed on the preprocessed image the resulting image in the form of binary. Consider an iris edge detection it provides a map of eyelid. edge candidates from this we can improve the speed and robustness of the algorithm and the computational steps are very simple. In this method only the edges playing a vital role we do not concentrate on the whole image. If the edge detection step is ignored in this process the output of the image which consist of so many errors because of the eyelid, pupil and the connections between the eyelids/iris and iris/sclera boundaries.

The different types of the edge detection make to recognize the iris feature. In this paper we used four different types of the edge detection method in fig.6 that is canny, sobal, prewitt and Roberts. And it is clear that the obtained results of the prewitt, sobal, Roberts are approximately similar, but the results of a leads to give different details, this is due to high power of canny mask. And the edge detection only calculates the local maxima of the gradient of I.

canny edge detection

The most common method of using the edge detection is canny edge detection in image processing. The canny edge detection is finding only the maximum level of image of the gradient I. where the gradient of the I is represented by the Gaussian filter and it has two threshold points to detect the strong and weak edges. The canny edge detection algorithm runs on 5 steps which explains below:

Smoothing, Finding gradient , Non-maximum suppression , Double thresholding , Edge tracking by hysteresis.

The canny edge detection is used to find the wide range of the edges by using this algorithm. Canny edge detection is a technique to extract the useful structural information from the different types of the objects and dramatically reduce the amount of data. And it has the low error rate and the detection should be accurate.

Biometrics recognition is an automatic recognition of an individual based on the behavioral and physiological features. Iris recognition have been proven to be very effective and secured and the information’s are protected very well. And there are no hacking and mal activities in this security systems so in modern days iris recognition is used most of the application. In this paper the biometric iris recognition system is implemented via the hybrid techniques such as wavelet and the feature extraction and the edge detection techniques. This approach leads the high level and precis ionized security protection other than the biometric recognition system.

Ethical Approval

No human or animal iris are used for this studies.

Competing interests

No, I declare that the authors have no competing interests as defined by Springer, or other interests that might be perceived to influence the results and/or discussion reported in this paper.

Authors' contributions

M.E have done the whole work. P.P, B.P and M.A has helped to improve text and prepare figures for manuscript. M.E have written the paper. All authors reviewed the manuscript.

Funding

There is no funding for this work.

Availability of data and materials

Selectively used from iris dataset.

PoojaKawale,JagrutiGhatole,ShwetaThote,Gauri Pohokar, PallaviDupare ,KarunaGechudeAsst Prof: YuwrajV.Waghmare”Matlab Based Iris Pattern RecognitionSystem” International Journal of Advanced Research in Computer Engineering & Technology (IJARCET) Volume 4 Issue 1, January 2015
SanpachaiH, Settapong Malisuwan “A Study of Image Enhancement for Iris Recognition”Journal of Industrial and Intelligent Information” Vol. 3, No. 1, March 2015
Harsha R1, Dr. K. Ramesha2”DWT Based Feature Extraction for Iris Recognition”International Journal of Advanced Research in Computer and Communication Engineering” Vol. 4, Issue 5, May 2015
MayankVatsa, Student Member, IEEE, Richa Singh, Student Member, IEEE, and AfzelNoore, Member, IEEE”Improving Iris Recognition Performance Using Segmentation, Quality Enhancement,Match Score Fusion, and Indexing” Ieee Transactions On Systems, Man, And Cybernetics—Part B: Cybernetics
VanajaRoselin.E.Chirchi, Dr.L.M.Waghmare, E.R.Chirchi”Iris Biometric Recognition for Person Identification in Security Systems”International Journal of Computer Applications (0975 – 8887) Volume 24– No.9, June 2011
JohnDaugman,PhD, OBE” How Iris RecognitionWorks”University of Cambridge, The Computer Laboratory, CambridgeCB23QG,U.K. www.CL.cam.ac.uk/users/jgd1000/
Binsu C. Kovoor1, Supriya M.H.2 and K. Poulose Jacob3”Iris Biometric Recognition System Employing Canny Operator”Sundarapandian et al. (Eds) : CSE, CICS, DBDM, AIFL, SCOM – 2013 pp. 65–74, 2013. © CS & IT-CSCP 2013
Khattab M. Ali Alheeti “Biometric Iris Recognition Based on HybridTechnique”International Journal on Soft Computing ( IJSC ) Vol.2, No.4, November 2011
Ayra G. Panganiban, Noel B. Linsangan, and Felicito S. Caluy “Implementation of Wavelet Transform-Based Algorithmfor Iris Recognition System”International Journal of Information and Electronics Engineering, Vol. 2, No. 3, May 2012
Rupinder Saini , Narinder Rana “Comparison Of Various Biometric Methods”International Journal of Advances in Science and Technology (IJAST)Vol 2 Issue I (March 2014)
Neha Kak, Rishi Gupta, SanchitMahajan”Iris Recognition System”(IJACSA) International Journal of Advanced Computer Science and Applications, Vol. 1, No. 1, 2010

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Matlab Programming on Iris Biometric Security Using Feature Extraction Process

Status:

Version 1

Abstract

Figures

Introduction

Iris Recognition

Image Preprocessing

2d Dwt

Histogram Equalization

Feature Extraction

EDGE DETECTION

Conclusion

Declarations

References

Additional Declarations

Status:

Version 1