• N.U.Samli
• M.Parisa Beham
• I.Sweetlin
• R.Tamilselvi

Biometric recognition is the use of physiological and traits to identify an individual. Many biometric traits have been developed and are being used to authenticate the person’s identity. Iris recognition systems are widely used and have proved to be efficient at individual recognition with high accuracy and nearly perfect matching. The iris features of two eyes of a same person are not similar making it more secure way of authentication compared to other biometric recognition system. The system is designed to pass only the users into the exam hall those are all verified by iris pattern recognition and block the unauthorized users. In the examination hall we proposed to use the iris pattern verification system to recognize only the authorized user. In the back end pre-processing, feature extraction, database matching are build in the Processor and the matched or mis-matched results will be displayed in the LCD which is front ended hardware level. The iris pattern capture from the subject is involved three phases. After recognition the output will be shown in digital devices comes under hardware part. The hardware part is an external feature which will have all the software inbuilt programmes for recognition. Here different databases namely CASIA, UBIRIS, IITD is used. When a person come in front of the camera(iris scanner) , the camera capture the eye and the software undergoes the techniques and match the sampled image with the total database stored in a memory chip and proved the results respective to the matched database. In Hardware implementation, Beagle board is interfaced with the LCD display to display the result i.e., authorized person/not. The machine learning algorithm’s are compiled with the python language which is easily fused in the beagle board. The objectives of the proposed method is to implement a new type of smart exam hall entry, to reduce the tension about the ID cards and hall ticket for the students which is also reduce the work tension of the supervisor. The time consumption for the document / ID card verification is reduced by this method thereby we also reduce the other unnecessary malpractice.

: Iris recognition, DSIFT algorithm, k-nn classifier.

" Invariant Iris Feature Extraction and Recognition For Smart Exam Hall Entry", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 5, page no.120-126, May-2019, Available :http://www.jetir.org/papers/JETIR1905720.pdf

" Invariant Iris Feature Extraction and Recognition For Smart Exam Hall Entry", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 5, page no. pp120-126, May-2019, Available at : http://www.jetir.org/papers/JETIR1905720.pdf