ResearchSpace

An integrated approach to fingerprint indexing using spectral clustering based on minutiae points

Show simple item record

dc.contributor.author Mngenge, NA
dc.contributor.author Mthembu, L
dc.contributor.author Nelwamondo, Fulufhelo V
dc.contributor.author Ngejane, Hombakazi C
dc.date.accessioned 2016-07-11T10:42:06Z
dc.date.available 2016-07-11T10:42:06Z
dc.date.issued 2015-07
dc.identifier.citation Mngenge, N.A Mthembu, L. Nelwamondo, F.V. and Ngejane, C.H. 2015. An integrated approach to fingerprint indexing using spectral clustering based on minutiae points. In: Science and Information Conference 2015, 28-30 July, 2015, London, UK. en_US
dc.identifier.isbn 978-1-4799-8547-0
dc.identifier.uri 10.1109/SAI.2015.7237300
dc.identifier.uri http://hdl.handle.net/10204/8612
dc.identifier.uri https://ieeexplore.ieee.org/document/7237300
dc.description Science and Information Conference 2015, 28-30 July, 2015, London, UK. Due to copyright restrictions, the attached PDF file only contains the abstract of the full text item. For access to the full text item, please consult the publisher's website en_US
dc.description.abstract Fingerprint indexing is an efficient approach that improves matching performance significantly in Automated Fingerprint Recognition Systems (AFRSs). Fingerprints are currently the most highly reliable and widely biometrics trait for identification and 1 - 1 matching. Hence, it would be very desirable to optimize them for identification and 1 - 1 matching applications. This work proposes an indexing approach based on minutiae points to reduce database search space. This is motivated by the fact that predefined classes (Left Loop, Right Loop, Whorl, Tented Arch, Plain Arch) are not always equally distributed in the search space i.e. some classes are more dominant than others. In such cases, a matching module can take hours to find an exact match. We solve this problem by constructing a rotational, scale and translation (RST) invariant fingerprint descriptor based on minutiae points. The proposed RST invariant descriptor dimensions are then reduced and passed to a spectral clustering algorithm which automatically creates 50 classes. Each of these 50 classes are then represented with a B+-Tree data structure for fast indexing. The keys used in each cluster are distances of feature vectors from the center of the cluster where they belong. Instead of searching a query to only a predicted cluster we also proposed to search for it in other clusters by employing triangle inequality rule. The system proposed is 81.4443% accurate on the NIST 4 special database. The results we got are promising because NIST 4 special database contains a lot of partial fingerprint. en_US
dc.language.iso en en_US
dc.publisher IEEE en_US
dc.relation.ispartofseries Workflow;16728
dc.subject Fingerprints en_US
dc.subject Indexing en_US
dc.subject Spectral Clustering en_US
dc.subject B+-Trees en_US
dc.subject Continuous Classification en_US
dc.title An integrated approach to fingerprint indexing using spectral clustering based on minutiae points en_US
dc.type Conference Presentation en_US
dc.identifier.apacitation Mngenge, N., Mthembu, L., Nelwamondo, F. V., & Ngejane, H. C. (2015). An integrated approach to fingerprint indexing using spectral clustering based on minutiae points. IEEE. http://hdl.handle.net/10204/8612 en_ZA
dc.identifier.chicagocitation Mngenge, NA, L Mthembu, Fulufhelo V Nelwamondo, and Hombakazi C Ngejane. "An integrated approach to fingerprint indexing using spectral clustering based on minutiae points." (2015): http://hdl.handle.net/10204/8612 en_ZA
dc.identifier.vancouvercitation Mngenge N, Mthembu L, Nelwamondo FV, Ngejane HC, An integrated approach to fingerprint indexing using spectral clustering based on minutiae points; IEEE; 2015. http://hdl.handle.net/10204/8612 . en_ZA
dc.identifier.ris TY - Conference Presentation AU - Mngenge, NA AU - Mthembu, L AU - Nelwamondo, Fulufhelo V AU - Ngejane, Hombakazi C AB - Fingerprint indexing is an efficient approach that improves matching performance significantly in Automated Fingerprint Recognition Systems (AFRSs). Fingerprints are currently the most highly reliable and widely biometrics trait for identification and 1 - 1 matching. Hence, it would be very desirable to optimize them for identification and 1 - 1 matching applications. This work proposes an indexing approach based on minutiae points to reduce database search space. This is motivated by the fact that predefined classes (Left Loop, Right Loop, Whorl, Tented Arch, Plain Arch) are not always equally distributed in the search space i.e. some classes are more dominant than others. In such cases, a matching module can take hours to find an exact match. We solve this problem by constructing a rotational, scale and translation (RST) invariant fingerprint descriptor based on minutiae points. The proposed RST invariant descriptor dimensions are then reduced and passed to a spectral clustering algorithm which automatically creates 50 classes. Each of these 50 classes are then represented with a B+-Tree data structure for fast indexing. The keys used in each cluster are distances of feature vectors from the center of the cluster where they belong. Instead of searching a query to only a predicted cluster we also proposed to search for it in other clusters by employing triangle inequality rule. The system proposed is 81.4443% accurate on the NIST 4 special database. The results we got are promising because NIST 4 special database contains a lot of partial fingerprint. DA - 2015-07 DB - ResearchSpace DP - CSIR KW - Fingerprints KW - Indexing KW - Spectral Clustering KW - B+-Trees KW - Continuous Classification LK - https://researchspace.csir.co.za PY - 2015 SM - 978-1-4799-8547-0 T1 - An integrated approach to fingerprint indexing using spectral clustering based on minutiae points TI - An integrated approach to fingerprint indexing using spectral clustering based on minutiae points UR - http://hdl.handle.net/10204/8612 ER - en_ZA


Files in this item

This item appears in the following Collection(s)

Show simple item record