The Fall of One, the Rise of Many: A Survey on Multi-Biometric Fusion Methods

• Published in: IEEE access Vol. 5; pp. 6247 - 6289
• Main Authors: Dinca, Lavinia Mihaela, Hancke, Gerhard Petrus
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 2017; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Adaptive systems; biometric cryptosystems; biometric key derivation; Biometric sensor; Biometrics; Encryption; Feature extraction; Iris recognition; Measurement; multi-biometric fusion; multi-biometrics; Security; Spoofing; template protection; Upper bound
• ISSN: 2169-3536; 2169-3536
• DOI: 10.1109/ACCESS.2017.2694050

Increasing operational and security demands changed biometrics by shifting the focus from single to multi-biometrics. Multi-biometrics are mandatory in the current context of large international biometric databases and to accommodate new emerging security demands. Our paper is a comprehensive survey on multi-biometrics, covering two important topics related to the multi-biometric field: fusion methods and security. Fusion is a core requirement in multi-biometric systems, being the method used to combine multiple biometric methods into a single system. The fusion section surveys recent multi-biometric schemes categorized from the perspective of fusion method. The security section is a comprehensive review of current issues, such as sensor spoofing, template security, and biometric encryption. New research trends and open challenges are discussed, such as soft, adaptive contextual-based biometrics. Finally, an implementation blueprint for a multi-biometric system is presented in the form of a list of questions to be answered when designing the system.