Please use this identifier to cite or link to this item: http://hdl.handle.net/10603/18614
Title: DEVELOPMENT OF ARTIFICIAL NEURAL NETWORKS FOR FINGERPRINT RECOGNITION
Researcher: GUHAN. P
Guide(s): Dr. Purushothaman. S
Keywords: 
Upload Date: 24-May-2014
University: Vels University
Completed Date: 25-04-2014
Abstract: The fingerprint recognition system is considered to most important newlinebiometric system in addition to other biometrics recognition systems. The newlinefingerprint recognition problem can be fingerprint verification and newlinefingerprint identification. newlineFingerprint verification refers to authenticity of a person by his newlinefingerprint. The user provides fingerprint together with identity newlineinformation. In the verification process template is retrieved based on the newlineidentification provided and matching is performed. newlinefingerprints based upon the unspecified conditions. In the identification of newlinefingerprint, the process matches fingerprints with the fingerprint database newlinefor similarity. newlineA good fingerprint is required for the best verification and newlineidentification tasks. Many approaches are available for fingerprint newlineverification and identification. One method is based on minutia, newlinerepresenting the fingerprint by its local features, like terminations and newlinebifurcations. The other method is based on image processing. In this newlinemethod, matching is based on the features of the image. newlineThis research work uses decomposition of fingerprint image by newlineusing wavelet method. The wavelet type used is db-1, and coiflet. The newlinefingerprint image is decomposed to five levels. In each level of newlinedecomposition, the fingerprint image is split into four parts namely: newlineapproximation matrix, vertical matrix, horizontal matrix and a diagonal newlinematrix. The subsequent level of decomposition uses an approximation of newlinethe previous level for further decomposition. newlineStatistical features are calculated at each level of decomposition newlineusing all the 4 coefficient matrices. The statistical features are used as newlineinputs for training the artificial neural network (ANN) and Fuzzy logic newlinealgorithms. newline newlineThe purpose of using ANN in the research work is because, existing newlinemethods are working based on statistical parameters. The purpose of newlineusing ANN for fingerprint recognition is due to the following reasons: newline1. The working concepts of ANN are based on statistics like using newlinelinear summation betw
Pagination: 
URI: http://hdl.handle.net/10603/18614
Appears in Departments:School of Computing Sciences

Files in This Item:
File Description SizeFormat 
01_title.pdfAttached File345.18 kBAdobe PDFView/Open
03_abstract.pdf18.96 kBAdobe PDFView/Open
04_declaration.pdf7.44 kBAdobe PDFView/Open
05_acknowledgement.pdf7.73 kBAdobe PDFView/Open
06_contents.pdf18.68 kBAdobe PDFView/Open
07_list of tables.pdf7.05 kBAdobe PDFView/Open
08_list of figures.pdf18.06 kBAdobe PDFView/Open
09_abbreviations.pdf10.06 kBAdobe PDFView/Open
10_chapter-1.pdf393.88 kBAdobe PDFView/Open
11_chapter-2.pdf460.41 kBAdobe PDFView/Open
12_chapter-3.pdf633.45 kBAdobe PDFView/Open
13_chapter-4.pdf1.32 MBAdobe PDFView/Open
14_chapter-5.pdf677.9 kBAdobe PDFView/Open
15_chapter-6.pdf379.95 kBAdobe PDFView/Open
16_conclusion.pdf195.93 kBAdobe PDFView/Open
17_biybliography.pdf292.88 kBAdobe PDFView/Open
guhan_paper-1.pdf1.15 MBAdobe PDFView/Open
guhan-paper-2.pdf1.24 MBAdobe PDFView/Open
journalofcomputerscienceijcsisvol.11no.10october2013.pdf7.73 MBAdobe PDFView/Open


Items in Shodhganga are protected by copyright, with all rights reserved, unless otherwise indicated.