Title: An intelligent technique for uniquely recognising face and finger image using learning vector quantisation (LVQ)-based template key generation

Authors: V. Arulkumar; P. Vivekanandan

Addresses: Department of Computer Science and Engineering, Sri Eshwar College of Engineering, Coimbatore, Tamilnadu, India ' Department of Computer Science and Engineering, Park College of Engineering and Technology, Coimbatore, Tamilnadu, India

Abstract: In current days, to identify, verify and detect the humans by using the bio recognition based on multimodal biometric was speedily developed which focus to necessitate the security keys in the authentication process of industries/organisation. The main goal of this organisation system is to integrate the instability biometric feature of the users and it is also used to formulate the key unimaginable to an unauthorised person. In this proposed work, in multimodal biometrics methods, particularly face and fingerprint based novel pattern matching is described. There are three modules to be used for exact feature extraction; multimodal biometric pattern generation and pattern matching are used to achieve better result for this proposed work. At first, from the face images and finger prints the minutiae points, face features and some other features are extracted. Consequently, by using a density-based score level fusion, the extracted features are merged together at match score level to create the multi-biometric pattern. After that, learning vector quantisation is used to perform the pattern matching. The assessment process was performed using large scale subjects to enter the system proving the ability of the proposed system over the state of art.

Keywords: face; finger image; multimodal biometric pattern generation; feature extraction; density-based score level fusion; learning vector quantisation; LVQ; key generation.

DOI: 10.1504/IJBET.2018.089951

International Journal of Biomedical Engineering and Technology, 2018 Vol.26 No.3/4, pp.237 - 249

Available online: 15 Feb 2018 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article