Certain improvement in preprocessing fingerprint image using artificial neural network

Biometrics is the science of measuring an individual’s physical properties. Biometric systems are being used as high level security technologies that provide identification and verification of human characteristics for security proposes. Biometric is characterized based on the feature that is analyz...

Full description

Bibliographic Details
Main Authors: Paulraj M. Pandiyan, Mohd. Yunus Hamid, Azali Saudi, Chin Kim On
Format: Proceedings
Language:English
English
Published: Allied Publishers Pvt. Ltd 2005
Subjects:
Online Access:https://eprints.ums.edu.my/id/eprint/31135/1/Certain%20improvement%20in%20preprocessing%20fingerprint%20image%20using%20artificial%20neural%20network-ABSTRACT.pdf
https://eprints.ums.edu.my/id/eprint/31135/2/Certain%20improvement%20in%20preprocessing%20fingerprint%20image%20using%20artificial%20neural%20network.pdf
_version_ 1825714464763150336
author Paulraj M. Pandiyan
Mohd. Yunus Hamid
Azali Saudi
Chin Kim On
author_facet Paulraj M. Pandiyan
Mohd. Yunus Hamid
Azali Saudi
Chin Kim On
author_sort Paulraj M. Pandiyan
collection UMS
description Biometrics is the science of measuring an individual’s physical properties. Biometric systems are being used as high level security technologies that provide identification and verification of human characteristics for security proposes. Biometric is characterized based on the feature that is analyzed. Presently, fingerprint biometric is the most widely adopted biometric technologies in the industry. A number of methods are already available in the literature to identify the fingerprints. The general steps in preprocessing the fingerprint image recognition system consists of image capturing, enhancement, binarization, filtering, and image thinning process. In order to obtain the minutiae features from the image, the image must be thinned properly. The recognition rate of fingerprints minutiae depend on the method of thinning the images. In this paper, a simple algorithm is proposed to thin the fingerprint image and the results are compared with the existing methods. Simple neural network models are also developed to thin the fingerprint images.
first_indexed 2024-03-06T03:12:12Z
format Proceedings
id ums.eprints-31135
institution Universiti Malaysia Sabah
language English
English
last_indexed 2024-03-06T03:12:12Z
publishDate 2005
publisher Allied Publishers Pvt. Ltd
record_format dspace
spelling ums.eprints-311352021-11-17T06:35:44Z https://eprints.ums.edu.my/id/eprint/31135/ Certain improvement in preprocessing fingerprint image using artificial neural network Paulraj M. Pandiyan Mohd. Yunus Hamid Azali Saudi Chin Kim On TA1501-1820 Applied optics. Photonics TK7800-8360 Electronics Biometrics is the science of measuring an individual’s physical properties. Biometric systems are being used as high level security technologies that provide identification and verification of human characteristics for security proposes. Biometric is characterized based on the feature that is analyzed. Presently, fingerprint biometric is the most widely adopted biometric technologies in the industry. A number of methods are already available in the literature to identify the fingerprints. The general steps in preprocessing the fingerprint image recognition system consists of image capturing, enhancement, binarization, filtering, and image thinning process. In order to obtain the minutiae features from the image, the image must be thinned properly. The recognition rate of fingerprints minutiae depend on the method of thinning the images. In this paper, a simple algorithm is proposed to thin the fingerprint image and the results are compared with the existing methods. Simple neural network models are also developed to thin the fingerprint images. Allied Publishers Pvt. Ltd 2005 Proceedings PeerReviewed text en https://eprints.ums.edu.my/id/eprint/31135/1/Certain%20improvement%20in%20preprocessing%20fingerprint%20image%20using%20artificial%20neural%20network-ABSTRACT.pdf text en https://eprints.ums.edu.my/id/eprint/31135/2/Certain%20improvement%20in%20preprocessing%20fingerprint%20image%20using%20artificial%20neural%20network.pdf Paulraj M. Pandiyan and Mohd. Yunus Hamid and Azali Saudi and Chin Kim On (2005) Certain improvement in preprocessing fingerprint image using artificial neural network. https://www.researchgate.net/publication/351267849_Certain_Improvement_In_Preprocessing_Fingerprint_Image_Using_Artificial_Neural_Network
spellingShingle TA1501-1820 Applied optics. Photonics
TK7800-8360 Electronics
Paulraj M. Pandiyan
Mohd. Yunus Hamid
Azali Saudi
Chin Kim On
Certain improvement in preprocessing fingerprint image using artificial neural network
title Certain improvement in preprocessing fingerprint image using artificial neural network
title_full Certain improvement in preprocessing fingerprint image using artificial neural network
title_fullStr Certain improvement in preprocessing fingerprint image using artificial neural network
title_full_unstemmed Certain improvement in preprocessing fingerprint image using artificial neural network
title_short Certain improvement in preprocessing fingerprint image using artificial neural network
title_sort certain improvement in preprocessing fingerprint image using artificial neural network
topic TA1501-1820 Applied optics. Photonics
TK7800-8360 Electronics
url https://eprints.ums.edu.my/id/eprint/31135/1/Certain%20improvement%20in%20preprocessing%20fingerprint%20image%20using%20artificial%20neural%20network-ABSTRACT.pdf
https://eprints.ums.edu.my/id/eprint/31135/2/Certain%20improvement%20in%20preprocessing%20fingerprint%20image%20using%20artificial%20neural%20network.pdf
work_keys_str_mv AT paulrajmpandiyan certainimprovementinpreprocessingfingerprintimageusingartificialneuralnetwork
AT mohdyunushamid certainimprovementinpreprocessingfingerprintimageusingartificialneuralnetwork
AT azalisaudi certainimprovementinpreprocessingfingerprintimageusingartificialneuralnetwork
AT chinkimon certainimprovementinpreprocessingfingerprintimageusingartificialneuralnetwork