Feature Extraction and Analysis in Multimodal Biometric Authentication using Lu Factorization with Kronecker Algebra
Y Suresh
Y Suresh, Assistant Professor, Department of Information Technology, Prasad V. Polturi Siddhartha Institute of Technology, Vijayawada, Andhra Pradesh India.
Manuscript received on 24 August 2019. | Revised Manuscript received on 06 September 2019. | Manuscript published on 30 September 2019. | PP: 3837-3839 | Volume-8 Issue-11, September 2019. | Retrieval Number: K22650981119/2019©BEIESP | DOI: 10.35940/ijitee.K2265.0981119
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Pattern recognition is one of the current and advanced technologies that focus on analysis and construction of pattern is a complex work. For recognition of patterns Vector logic gives good strategies. This paper focuses on pattern recognition in multimodal authentication system by using vector logic. A framework has been proposed to provide more security in biometric aspect. Initially, features are extracted through PCA from the normalized biometric imaginaries, and then using LU factorization key components are extracted. By using convolution kernel methods such as Khatri Rao an application of Kronecker product weights are computed for different key sizes. In the same way verification process is implemented and verified with MSE. This framework gives better result for chosen threshold value.
Keywords: Khatri-Rao Product, Kronecker Product, LU, MSE, PCA.
Scope of the Article: Authentication, Authorization, Accounting