Brown, DaneBradshaw, K2017-06-072017-06-072016-09Brown, D. and Bradshaw, K. 2016. A dynamically weighted multi-modal biometric security system. Southern Africa Telecommunication Networks and Applications Conference (SATNAC), 4-7 September 2016, George, p. 254-258978-0-620-72418-0http://www.satnac.org.za/proceedings/2016/SATNAC%202016%20Proceedings%20Final.pdfhttp://hdl.handle.net/10204/9118Southern Africa Telecommunication Networks and Applications Conference, 4-7 September 2016, GeorgeThe face, fingerprint and palmprint feature vectors are automatically extracted and dynamically selected for fusion at the feature-level, toward an improved human identification accuracy. The feature-level has a higher potential accuracy than the match score-level. However, leveraging this potential requires a new approach. This work demonstrates a novel dynamic weighting algorithm for improved image-based biometric feature-fusion. A comparison is performed on uni-modal, bi-modal, tri-modal and proposed dynamic approaches. The proposed dynamic approach yields a high genuine acceptance rate of 99.25% genuine acceptance rate at a false acceptance rate of 1% on challenging datasets and big impostor datasets.enFace feature vectorsFingerprint feature vectorsPalmprint feature vectorsBiometric security systemsConference PresentationBrown, D., & Bradshaw, K. (2016). A dynamically weighted multi-modal biometric security system. www.satnac.org.za. http://hdl.handle.net/10204/9118Brown, Dane, and K Bradshaw. "A dynamically weighted multi-modal biometric security system." (2016): http://hdl.handle.net/10204/9118Brown D, Bradshaw K, A dynamically weighted multi-modal biometric security system; www.satnac.org.za; 2016. http://hdl.handle.net/10204/9118 .TY - Conference Presentation AU - Brown, Dane AU - Bradshaw, K AB - The face, fingerprint and palmprint feature vectors are automatically extracted and dynamically selected for fusion at the feature-level, toward an improved human identification accuracy. The feature-level has a higher potential accuracy than the match score-level. However, leveraging this potential requires a new approach. This work demonstrates a novel dynamic weighting algorithm for improved image-based biometric feature-fusion. A comparison is performed on uni-modal, bi-modal, tri-modal and proposed dynamic approaches. The proposed dynamic approach yields a high genuine acceptance rate of 99.25% genuine acceptance rate at a false acceptance rate of 1% on challenging datasets and big impostor datasets. DA - 2016-09 DB - ResearchSpace DP - CSIR KW - Face feature vectors KW - Fingerprint feature vectors KW - Palmprint feature vectors KW - Biometric security systems LK - https://researchspace.csir.co.za PY - 2016 SM - 978-0-620-72418-0 T1 - A dynamically weighted multi-modal biometric security system TI - A dynamically weighted multi-modal biometric security system UR - http://hdl.handle.net/10204/9118 ER -