Image Fusion for Single-Trait Multimodal Biometrics: A Brief Review

• Published in: 2022 IEEE 20th Student Conference on Research and Development, SCOReD 2022
• Main Author: 2-s2.0-85145433170
• Format: Conference paper
• Language: English
• Published: Institute of Electrical and Electronics Engineers Inc. 2022
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85145433170&doi=10.1109%2fSCOReD57082.2022.9974027&partnerID=40&md5=516708f7d62bc834400c4d111d56aaaa

Biometrics is a process of determining an individual's identity based on their personal traits. Most of the available biometrics applications are based on physical traits, such as fingerprint, face and ear. All biometrics traits have their own drawbacks, such as occlusion, illumination, and pose variation. To overcome these issues, image fusion approach has been proposed in several studies to improve the performance of biometrics applications. Image fusion may happen at three levels (i.e. pixel, feature, and score). Implementation of image fusion approach in biometrics is generally called multimodal biometrics. In multimodal biometrics, source images to be fused can be taken from a single trait (same biometrics trait, e.g., thermal and visible face images) or multiple traits (different biometrics traits, e.g., fusion of ear and fingerprint). In addition, multimodal biometrics also can be done based on a single trait, such as fusion of thermal and visible face image or fusion of features extracted from one trait (e.g., LBP and HOG features of the ear). This paper describes and discusses image fusion implementation at the mentioned three levels, using a single biometrics trait. © 2022 IEEE.