Authentication with Iris Recognition Based on A 3-Tier Security Analysis Approach

 Audit controls are made using the tools such as ID, magnetic card, password, pin code to enable people to access to areas requiring access permission. This situation with increasing the number of security measures forces people to remember more than one password. In addition, it is becoming compulsory for a person to have more than one type of card in order to be able to identify himself / herself. Increasingly reliable and practical detachment of such measures has increased the interest of researchers in biometrics systems, which is the recognition method of self-identification by using their own structural features. The aim in this project is to authenticate with one of these biometric systems, iris recognition. Iris screening is one of the most reliable biometric scans. There is no need for physical contact between the user and the scanner. Being able to use even with glasses, easy integration into systems and being one of the most reliable designs of iris has been the main factors in choosing iris definition. In the project done, the security level is aimed to be authenticated in a short time and correct match with the algorithm produced based on the increased three layer security analysis. These layers in the project; eye color in the first layer, ratio of the area of the iris to the area of the eyeball in the second layer, and tissue analysis in the third layer. The difference between this project and other work that has been worked on before is that the authentication process is performed with a different algorithm approach by increasing the number of security layers. Thus, it is aimed to reach reality in a safer and shorter time. At this time, only studies of iris texture have been carried out in the examination of iris. Other factors have not been evaluated in studies. In this study, eye color and the ratio of the iris area to the eyeball are examined by adding the account. After these factors are correctly matched, iris tissue is examined. The project has two databases, one to record real-time data, and the other to contain data from the CASIA database. The real-time data base is created with the images we have obtained from different people with the piece of hardware we have developed. By using different image processing algorithms, images in these databases are processed and an iris code is created to see the key task. Snapshots taken from the live eye are compared through the created iris code to the database data to find out if there is a match. It is determined that the analysis of the three security layers is done in a short time and with high success rate. The version_01 of the CASIA database is used in the study. In this database, 50 images of eyes taken from different angles at different times were worked on. In the real time database there are 25 images. The success rates of both databases are calculated separately. The obligation of individuals must carry lots of card with them and memorizing many password in order to introduce themselves will be avoid with this work. It will provide savings time and financial gains to institutions and individuals. There is not much study about identification of iris in the country, and necessary software in this area is generally supplied from abroad. With this work, it is aimed to increase the interest of researchers in this field and to eliminate this deficiency in the country.

Authentication with Iris Recognition Based on A 3-Tier Security Analysis Approach

Audit controls are made using the tools such as ID, magnetic card, password, pin code to enable people to access to areas requiring access permission. This situation with increasing the number of security measures forces people to remember more than one password. In addition, it is becoming compulsory for a person to have more than one type of card in order to be able to identify himself / herself. Increasingly reliable and practical detachment of such measures has increased the interest of researchers in biometrics systems, which is the recognition method of self-identification by using their own structural features. The aim in this project is to authenticate with one of these biometric systems, iris recognition. Iris screening is one of the most reliable biometric scans. There is no need for physical contact between the user and the scanner. Being able to use even with glasses, easy integration into systems and being one of the most reliable designs of iris has been the main factors in choosing iris definition. In the project done, the security level is aimed to be authenticated in a short time and correct match with the algorithm produced based on the increased three layer security analysis. These layers in the project; eye color in the first layer, ratio of the area of the iris to the area of the eyeball in the second layer, and tissue analysis in the third layer. The difference between this project and other work that has been worked on before is that the authentication process is performed with a different algorithm approach by increasing the number of security layers. Thus, it is aimed to reach reality in a safer and shorter time. At this time, only studies of iris texture have been carried out in the examination of iris. Other factors have not been evaluated in studies. In this study, eye color and the ratio of the iris area to the eyeball are examined by adding the account. After these factors are correctly matched, iris tissue is examined. The project has two databases, one to record real-time data, and the other to contain data from the CASIA database. The real-time data base is created with the images we have obtained from different people with the piece of hardware we have developed. By using different image processing algorithms, images in these databases are processed and an iris code is created to see the key task. Snapshots taken from the live eye are compared through the created iris code to the database data to find out if there is a match. It is determined that the analysis of the three security layers is done in a short time and with high success rate. The version_01 of the CASIA database is used in the study. In this database, 50 images of eyes taken from different angles at different times were worked on. In the real time database there are 25 images. The success rates of both databases are calculated separately. The obligation of individuals must carry lots of card with them and memorizing many password in order to introduce themselves will be avoid with this work. It will provide savings time and financial gains to institutions and individuals. There is not much study about identification of iris in the country, and necessary software in this area is generally supplied from abroad. With this work, it is aimed to increase the interest of researchers in this field and to eliminate this deficiency in the country.

___

  • Abatea A. F, Barrab S., Gallob L., Narducci F. “Kurtosis and skewness at pixel level as input for SOM networks to iris recognition on mobile devices” 2017.02.
  • Akçay M., Çetinkaya H. H., Kampüslerde Uygulanan Yeni Biyometrik Sistemler, Akademik Bilişim, Malatya-Türkiye, (2011).
  • Çakır A, Volkan A, Akbulut F. T, “Iris Tanıma Sistemleri ve Uygulama Alanları” 3-4-
  • Erişti E. “Görüntü İşlemede Yeni Bir Soluk, OPENCV” 223-224 Akademik Bilişim, Muğla (2010)
  • Gürkan G, Akan A (2015) “Doku Analiz Yaklaşımına Dayanan Yeni Bir İris Tanıma Yöntemi A New Iris Recognition Method Based on Texture Analysis” 26-28
  • Jain A.K., Ross A., and Prabhakar S., "An Introduction to Biometric Recognition," IEEE Trans. on Circuits and Systems for Video Technology, 14(1):4-20, 2004.
  • Kamal H., Ujwalla G., Yogesh G. (2015) “Neural Network Approach to Iris Recognition in Noisy Environment” International Conference on Information Security & Privacy (ICISP2015), 11-12 December 2015, Nagpur, INDIA
  • Tekyıldız A., Güllü M.K, Urhan O (2015) “Irıs Recognıtıon System Based On Fast Irıs Localızatıon And Phase Correlatıon Mathchıng “İşaret Ve Görüntü İşleme Lab. (Kulis), 4-5.
  • Yang H., Konstantinos S., Gareth H. “A novel iris weight map method for less constrained iris recognition based on bit stability and discriminability” February 2017, Pages 168–180
  • Yuanning L, Fei H., Xiaodong Z., Zhen L., Ying C., (2015) “The Improved Characteristics of Bionic Gabor Representations by Combining with SIFT Key-points for Iris Recognition” Journal of Bionic Engineering 12 (2015) 504–517
Bilge International Journal of Science and Technology Research-Cover
  • ISSN: 2651-401X
  • Yayın Aralığı: Yılda 2 Sayı
  • Başlangıç: 2017
  • Yayıncı: Kutbilge Akademisyenler Derneği