Work place: Dept. of Electronic Engineering, University of Nigeria Nsukka, Enugu State Nigeria
E-mail: ijeoma.ezika@unn.edu.ng
Website:
Research Interests: Engineering, Data Structures and Algorithms, Computer Architecture and Organization, Computer Vision, Computational Learning Theory, Computational Engineering, Computational Science and Engineering
Biography
Ijeoma J.F. Ezika is a Lecturer in the Department of Electronic Engineering, University of Nigeria, Nsukka. Her current research interests are in the areas of computer vision, machine learning, data analytics and engineering education
By Samuel Ezichi Ijeoma J.F. Ezika Ogechukwu N. Iloanusi
DOI: https://doi.org/10.5815/ijigsp.2021.06.05, Pub. Date: 8 Dec. 2021
Soft biometrics is not a unique trait in itself, but it is valuable in enhancing the performance of unique traits used in biometric recognition systems. In this paper, we perform a comparative analysis of soft biometric traits and fusion schemes for improving face recognition systems. Specifically, we present an analysis of the performance of such systems as a function of the fusion strategy used and the soft biometric feature employed. We outline the strengths and weaknesses of the biometric feature employed in fused face and soft biometric systems. The analysis presented in this work is significantly important and different from existing works as the performance profiles of a wider variety of soft biometric traits are compared over major metrics of permanence, ease of collection and distinctiveness.
[...] Read more.Subscribe to receive issue release notifications and newsletters from MECS Press journals