Please use this identifier to cite or link to this item: https://hdl.handle.net/20.500.12540/166
Title: Efficient log-based iris detection and image sharpness enhancement (l-IDISE) using artificial neural network
Authors: Lakshmi, S. V. 
Sathyamoorthy, K. 
Krishnamoorthy, Sujatha 
Luhach, Ashish Kr. 
Issue Date: 2019
Publisher: The Mattingley Publishing Co., Inc.
Source: Krishnamoorthy, S., Sathyamoorthy, K., Luhach, A. K., & Lakshmi, S. V. (2019). Efficient log-based iris detection and image sharpness enhancement (l-IDISE) using artificial neural network. TEST Engineering & Management, 81, 5137-5145.
Journal: TEST Engineering & Management 
Abstract: Identification of people managed by Biometric, and it is dependent on their organic qualities. Iris Recognition System (IRS) is viewed as the most dependable and exact biometric recognizable proof framework accessible. Conventional individual validation techniques have numerous intuitive imperfections. Biometrics is a compelling innovation to conquer these imperfections. Biometric oversees the Identification of individuals subject to their natural properties. Iris Recognition System (IRS) is seen as the most reliable and exact biometric unmistakable verification structure available. Standard individual approval methodologies have various instinctual methods. Biometrics is a reasonable advancement to overcome these distortions. In this paper, a methodology for IRS has proposed systems for developing and to distinguish the pupil. The noises, for example, eyelashes, and reflections are expelled through the straight thresholding. The 3D log is applied for highlight extraction from a fragmented iris image (IMG).It is seen that the proposed methodology is increasingly proficient for the considered dataset from the test results. It is likewise seen that the proposed methodology sets aside sensible measures of effort to perform iris division, and acknowledgment precision is additionally sensible. Hazy, little resolution IMGwith poor light make a significant test for IRS are the examples for Low-quality iris IMG. This proposed l-IDISE is planned to structure an IRS in three first strides to confirm the objective. (1) Applying IMG handling procedures on the image of an eye for information arrangement. (2) The Back Propagation Artificial Neural Network (BPANN) systems for recognizable proof. Third new IRS calculation for improvement of standardized iris IMG. The Logarithmic Image Processing (LIP) IMG upgrades process. Consequences of numerous biometric stupendous test iris information show noteworthy improvement in the presentation of IR calculations as far as equivalent mistake rates. The BPANN was prevailing in recognizable proof and getting the best outcomes since it achieved Recognition Rate equivalent to 90%.
URI: https://hdl.handle.net/20.500.12540/166
Appears in Collections:Scholarly Publications

Files in This Item:
File Description SizeFormat 
wku_schlrs_publcn_000117.pdf1.39 MBAdobe PDFThumbnail
View/Open
Show full item record

Page view(s)

584
checked on Jun 25, 2021

Download(s)

131
checked on Jun 25, 2021

Google ScholarTM

Check


This item is licensed under a Creative Commons License Creative Commons