Iris recognition under non-ideal conditions (Record no. 6565)

MARC details
000 -LEADER
fixed length control field 03541nam a22001937a 4500
003 - CONTROL NUMBER IDENTIFIER
control field OSt
005 - DATE AND TIME OF LATEST TRANSACTION
control field 20220107122853.0
008 - FIXED-LENGTH DATA ELEMENTS--GENERAL INFORMATION
fixed length control field 190710b xxu||||| |||| 00| 0 eng d
040 ## - CATALOGING SOURCE
Transcribing agency IIITMK
100 ## - MAIN ENTRY--PERSONAL NAME
Personal name Akhila P (41718002)
9 (RLIN) 16153
245 ## - TITLE STATEMENT
Title Iris recognition under non-ideal conditions
300 ## - PHYSICAL DESCRIPTION
Extent MPhil CS 2018-2019
500 ## - GENERAL NOTE
General note A biometric system provides automatic identication of an individual based on unique physiological<br/>or behavioral features or characteristics possessed by an individual. Iris recognition<br/>is such a technology for identifying humans by capturing and analyzing the unique patterns<br/>of the iris in the human eye. Unlike other biometrics such as ngerprints or face, the distinctive<br/>aspect of iris comes from randomly distributed features in the iris. Iris recognition is<br/>regarded as the most reliable and accurate biometric identication system available at present.<br/>Iris recognition/classification under non ideal imaging conditions is a challenging task. Nonideal<br/>conditions include iris images that are off-angled, occluded, blurred, noisy, iris captured<br/>at a distance and so on.<br/>Segmentation of the iris from the eye image is an important task in iris recognition. Iris<br/>segmentation is the process of isolating the iris region from the surrounding structures of the<br/>captured image of the eye. In ideal situations, we can easily segment the iris region from the<br/>eye images. However, in the case of non ideal situations, iris segmentation is a challenging task.<br/>One can never get valid recognition result without the right type of segmentation. Previous<br/>studies have shown that the performance of an iris recognition systems highly dependents on<br/>the performance of its segmentation algorithm. Several iris segmentation algorithms such as<br/>those based on Hough transform, active contour and integro-differential operator have been<br/>developed. However, these segmentation algorithms do not perform well under non-ideal and<br/>less constrained conditions. So, one of the objective of this research is to develop an automatic<br/>segmentation method for extracting the iris region. Towards this in this thesis, we propose<br/>a novel mechanism for segmentation of iris images using a encoder-decoder network. The<br/>proposed method gave an accuracy of 98% on non ideal iris images taken from UBIRIS v2<br/>database.<br/>Classification of iris images acquired under non ideal conditions is another challenge<br/>due to the poor quality of the images. Currently, there are some works on iris recognition/classification<br/>based on CNN. They use one of the pre-trained models to extract the features<br/>and then SVM is used for the classification. SVM is a non parametric model where the complexity<br/>grows as the number of training samples increases. To reduce the complexity and improve<br/>the recognition accuracy, we propose a novel mechanism for iris classication using LSTM<br/>sequence prediction model. LSTM can extract the long-term dependencies of the data features<br/>in the sequence. It is very similar to the recognition by the human visual system. We tested the<br/>proposed method on a UBIRIS v2 database which includes iris images under various non ideal<br/>conditions and compared its performance with existing SVM based approach. The proposed<br/>method gave an accuracy of 81% and outperformed the existing SVM based method (78%<br/>accuracy).
502 ## - DISSERTATION NOTE
Degree type MPhil CS
Name of granting institution 2018-2019
Year degree granted INT
-- Dr Tony Thomas
650 ## - SUBJECT ADDED ENTRY--TOPICAL TERM
Topical term or geographic name entry element IRIS RECOGNITION
9 (RLIN) 16154
650 ## - SUBJECT ADDED ENTRY--TOPICAL TERM
Topical term or geographic name entry element RECURRENT NEURAL NETWORK
9 (RLIN) 16155
650 ## - SUBJECT ADDED ENTRY--TOPICAL TERM
Topical term or geographic name entry element CONVOLUTIONAL NEURAL NETWORK
9 (RLIN) 16156
942 ## - ADDED ENTRY ELEMENTS (KOHA)
Source of classification or shelving scheme Dewey Decimal Classification
Koha item type
Holdings
Withdrawn status Lost status Source of classification or shelving scheme Damaged status Not for loan Collection code Home library Current library Shelving location Date acquired Total Checkouts Barcode Date last seen Price effective from Koha item type
    Dewey Decimal Classification     Non Fiction IIITM-K Kerala University of Digital Sciences, Innovation and Technology Knowledge Centre   10/07/2019   R-1569 10/07/2019 10/07/2019 Project Reports