Abstract
In spite of the benefits of biometricbased authentication systems, there are few concerns raised because of the sensitivity of biometric data to outliers, low performance caused due to intraclass variations, and privacy invasion caused by information leakage. To address these issues, we propose a hybrid fusion framework where only the protected modalities are combined to fulfill the requirement of secrecy and performance improvement. This paper presents a method to integrate cancelable modalities utilizing MeanClosure Weighting (MCW) score level and DempsterShafer (DS) theory based decision level fusion for iris and fingerprint to mitigate the limitations in the individual score or decision fusion mechanisms. The proposed hybrid fusion scheme incorporates the similarity scores from different matchers corresponding to each protected modality. The individual scores obtained from different matchers for each modality are combined using MCW score fusion method. The MCW technique achieves the optimal weight for each matcher involved in the score computation. Further, DS theory is applied to the induced scores to output the final decision. The rigorous experimental evaluations on three virtual databases indicate that the proposed hybrid fusion framework outperforms over the component level or individual fusion methods (score level and decision level fusion). As a result, we achieve (48%, 66%), (72%, 86%) and (49%, 38%) of performance improvement over unimodal cancelable iris and unimodal cancelable fingerprint verification systems for Virtual_A, Virtual_B, and Virtual_C databases, respectively. Also, the proposed method is robust enough to the variability of scores and outliers satisfying the requirement of secure authentication.
This is a preview of subscription content, log in to check access.
References
 1.
Bengio S, Mariéthoz J (2004) A statistical significance test for person authentication. In: Proceedings of Odyssey 2004: The Speaker and Language Recognition Workshop, pp 237–244
 2.
Camlikaya E, Kholmatov A, Yanikoglu B (2008) Multibiometric templates using fingerprint and voice. In: Biometric technology for human identification V. vol. 6944, p 69440i. International society for optics and photonics
 3.
Canuto AM, Pintro F, XavierJunior JC (2013) Investigating fusion approaches in multibiometric cancellable recognition. Expert Syst Appl 40(6):1971–1980
 4.
CASIA V3Interval database: Institute of automation, chinese academy of sciences(CASIA), CASIA iris image database version 3.0 (2010), http://www.cbsr.ia.ac.cn/Databases.htm
 5.
Chenghua Y, Qixiang C (2009) An improved algorithm for dempstershafer theory of evidence. In: 2009 International conference on electronic commerce and business intelligence, pp 475–478
 6.
Chin Y, Ong T, Teoh A, Goh K (2014) Integrated biometrics template protection technique based on fingerprint and palmprint featurelevel fusion. Information Fusion 18:161–174
 7.
Cimato S, Gamassi M, Piuri V, Sassi R, Scotti F (2008) Privacyaware biometrics: Design and implementation of a multimodal verification system. In: 2008 Annual computer security applications conference (ACSAC), pp 130–139
 8.
Daugman J (2002) How iris recognition works. In: International conference on image processing. vol. 1, pp 33–36
 9.
Dwivedi R, Dey S (2018a) Securing fingerprint template using noninvertible ridge feature transformation. Unpublished manuscript
 10.
Dwivedi R, Dey S (2018b) Scorelevel fusion for cancelable multibiometric verification. Pattern Recognition Letters pp –
 11.
Dwivedi R, Dey S, Singh R, Prasad A (2017) A privacypreserving cancelable iris template generation scheme using decimal encoding and lookup table mapping. Computers & Security 65:373–386
 12.
FVC2002: Fingerprint Verification Competition (2002). https://biolab.csr.unibo.it/FVCOnGoing/UI/Form/Home.aspxp
 13.
Grover J, Hanmandlu M (2015) Hybrid fusion of score level and adaptive fuzzy decision level fusions for the fingerknuckleprint based authentication. Appl Soft Comput 31:1–13
 14.
He M, Horng SJ, Fan P, Run RS, Chen RJ, Lai JL, Khan MK, Sentosa KO (2010) Performance evaluation of score level fusion in multimodal biometric systems. Pattern Recogn 43(5):1789–1800
 15.
ISO/IEC 197951:2006 Information technology – Biometric performance testing and reporting – Part 1: Principles and framework: ISO/IEC 197951:2006 information technology – biometric performance testing and reporting – part 1: Principles and framework (2006). https://www.iso.org/standard/41447.html
 16.
ISO/IEC 238237:2012 information technology – vocabulary – part 37: biometrics 2012: ISO/IEC 238237:2012 information technology – vocabulary – part 37: biometrics 2012 (2012). https://www.iso.org/standard/55194.html
 17.
Joshi T, Dey S, Samanta D (2009) Multimodal biometrics: state of the art in fusion techniques. International Journal of Biometrics 1(4):393–417
 18.
Kabir W, Ahmad MO, Swamy MNS (2018) Normalization and weighting techniques based on genuineimpostor score fusion in multibiometric systems. IEEE Trans Inf Forensics Secur 13(8):1989–2000
 19.
Kelkboom EJC, Zhou X, Breebaart J, Veldhuis RNJ, Busch C (2009) Multialgorithm fusion with template protection. In: 2009 IEEE 3rd international conference on biometrics: Theory, applications, and systems, pp 1–8
 20.
Kolomvatsos K, Anagnostopoulos C, Hadjiefthymiades S (2015) Contextual reasoning under uncertainty in sensor data stream monitoring. International Journal of Monitoring and Surveillance Technologies Research 3 (1):1–19
 21.
Kolomvatsos K, Anagnostopoulos C, Hadjiefthymiades S (2017a) Data fusion and type2 fuzzy inference in contextual data stream monitoring. IEEE Trans Syst Man Cybern Syst Hum 47(8):1839–1853
 22.
Kolomvatsos K, Anagnostopoulos C, Hadjiefthymiades S (2017b) Distributed localized contextual event reasoning under uncertainty. IEEE Internet of Things Journal 4(1):183–191
 23.
Kumar A, Kumar A (2016) Adaptive management of multimodal biometrics fusion using ant colony optimization. Information Fusion 32:49–63
 24.
Li SZ, Jain AK (2009) Encyclopedia of Biometrics, vol. 1 Springer Science & Business Media
 25.
Maiorana E, Hine GE, Campisi P (2015) Hillclimbing attacks on multibiometrics recognition systems. IEEE Trans Inf Forensics Secur 10(5):900–915
 26.
Masek L (2003) Recognition of human iris patterns for biometric identification. Tech. rep., University of Western Australia, Crawley
 27.
Mezai L, Hachouf F (2015) Scorelevel fusion of face and voice using particle swarm optimization and belief functions. IEEE Transactions on HumanMachine Systems 45(6):761–772
 28.
Multimedia University version1.0: Multimedia university version1.0, MMU1 iris image database (2014 (Accessed)).www.cs.princeton.edu/~andyz/downloads/MMUIrisDatabase.zip
 29.
Nandakumar K, Jain AK (Sept 2015) Biometric template protection: Bridging the performance gap between theory and practice. IEEE Signal Proc Mag 32(5):88–100
 30.
Nguyen K, Denman S, Sridharan S, Fookes C (2015) Scorelevel multibiometric fusion based on dempster shafer theory incorporating uncertainty factors. IEEE Transactions on HumanMachine Systems 45(1):132–140
 31.
Othman A, Ross A (2013) On mixing fingerprints. IEEE Trans Inf Forensics Secur 8(1):260–267
 32.
Paul PP, Gavrilova M (2012) Multimodal cancelable biometrics. In: 2012 IEEE 11th international conference on cognitive informatics and cognitive computing, pp 43–49
 33.
Prabhakar S, Jain AK (2002) Decisionlevel fusion in fingerprint verification. Pattern Recogn 35(4):861–874
 34.
Provan GM (1990) A logicbased analysis of dempstershafer theory. Int J Approx Reason 4(5):451–495
 35.
Rathgeb C, GomezBarrero M, Busch C, Galbally J, Fierrez J (2015) Towards cancelable multibiometrics based on bloom filters: a case study on feature level fusion of face and iris. In: 3rd international workshop on biometrics and forensics (IWBF 2015), pp 1–6
 36.
Ross AA, Nandakumar K, Jain AK (2006) Handbook of multibiometrics. Springer, New York
 37.
Sadhya D, Singh SK (2018) Construction of a bayesian decision theorybased secure multimodal fusion framework for soft biometric traits. IET Biometrics 7(3):251–259
 38.
Shafer G (1976) A mathematical theory of evidence, vol. 42, Princeton University Press, Princeton
 39.
Simoens K, Bringer J, Chabanne H, Seys S (2012) A framework for analyzing template security and privacy in biometric authentication systems. IEEE Trans Inf Forensics Secur 7(2):833– 841
 40.
Smets P (2002) Decision making in a context where uncertainty is represented by belief functions, pp. 17–61. PhysicaVerlag HD, Heidelberg
 41.
Sutcu Y, Li Q, Memon N (2007) Secure biometric templates from fingerprintface features. In: 2007 IEEE Conference on computer vision and pattern recognition, pp 1–6
 42.
Talreja V, Valenti MC, Nasrabadi NM (2017) Multibiometric secure system based on deep learning. In: 2017 IEEE Global conference on signal and information processing (globalSIP), pp 298–302
 43.
Tao Q, Veldhuis R (2008) Hybrid fusion for biometrics: Combining scorelevel and decisionlevel fusion. In: 2008 IEEE Computer society conference on computer vision and pattern recognition workshops, pp 1–6
 44.
Wong WJ, Teoh AB, Wong MD, Kho YH (2013) Enhanced multiline code for minutiaebased fingerprint template protection. Pattern Recogn Lett 34(11):1221–1229
 45.
Yang W, Wang S, Hu J, Zheng G, Valli C (2018) A fingerprint and fingervein based cancelable multibiometric system. Pattern Recogn 78:242–251
Acknowledgements
The authors are thankful to SERB (ECR/2017/000027), Department of Science & Technology, Govt. of India for providing financial support. Also, we would like to acknowledge Indian Institute of Technology Indore for providing the laboratory support and research facilities to carry out this research.
Author information
Affiliations
Corresponding author
Appendices
Appendix A: Preliminaries on DempsterShafer theory of evidence
In Bayesian theory, the probabilities are assigned to each individual proposition from a set of mutually exclusive propositions. Alternatively, DS theory assigns masses to each combination of events. Unlike DS theory, the probability theory is unable to discriminate between ignorance and uncertainty due to sketchy information. Fundamentally, the Bayesian theory departs DS theory in the aspect of handling ignorance. DS theory does not assign belief to ignorance or to a falsified hypothesis. The mass is assigned particularly to the subsets for which we seek to assign belief. This implies neither belief nor disbelief for the evidence to a certain value. Hence we have utilized DS theory in our work.
Consider, 𝜃 be a finite set of all possible hypotheses known as a frame of discernment. The power set 2^{𝜃} contains all subsets of 𝜃 including a null set (ϕ) and itself. Each subset in the power set is referred as a focal element and assigned a value in between [0, 1] on the basis of their evidence. A value of 1 corresponds to total belief and 0 for no belief. In general, the assigned value is named as basic belief assignment (BBA). In DS theory [33], BBA is assigned to each subset i.e. hypothesis also known as the mass of the individual proposition,
If \(\theta =\left \{ A,B \right \} \ \text {then} \ 2^{\theta }=\left \{\varnothing ,A,B,\theta \right \}\). The mass function fulfills the following criteria:
where \(\varnothing \) represents the empty set. The measure of belief is defined by the function bel : 2^{𝜃} → [0, 1],
The bel can also be formally defined as:
This means the degree of belief x for the classifier Y at time t when w_{0} ∈ A. Here, E_{Y,t} represents the evidential information known to classifier Y at time t. For ease in representation, we use bel(A) instead of \( bel_{Y,t}^{\theta ,\Re }\left [ E_{Y,t} \right ]\left (w_{0}\in A \right )\). Next, plausibility (pl) is measured as:
If 𝜃 defines the set of all possible hypotheses, then the level of uncertainty is denoted by m (𝜃). In a hypothesis, beliefs and disbeliefs may not sum to 1 and may attain 0 value. A value of 0 signifies no evidence present for the hypothesis. The DS theory based aggregation involves the following steps:

The measure of belief is evaluated based on the facts from the different sources of information. As compared to Bayesian theory, the masses are not distributed among classes.

Dempster rule of combination is applied to aggregate belief measure obtained from the available information and facts.
For different sources, (1, 2,⋯ ,N), Dempster’s rule of combination is described in (16):
where A,B_{1},…,B_{N} ⊆ 𝜃, and
where K denotes the conflict present between evidences; 1K is the normalization factor.
A.1 Updation of masses
In a majority of the scenarios, mass updating is required if any new evidence or belief is encountered. Suppose, E ⊂ 𝜃 and E_{d} be the evidence not present in E. If this new evidence provides the exact value of E_{d}, then bel(A) is updated based on the following condition rule:
After the computation of the masses, the classification is performed onto the training set. One of the aggregation rules is applied to evaluate total conflicting mass. Next, the winnertakeall assignment is utilized to compute A(k), which is defined in (19):
where M + 1 represents is the total number of classes including the class of rejection and A_{M+ 1} = 𝜃.
Rights and permissions
About this article
Cite this article
Dwivedi, R., Dey, S. A novel hybrid score level and decision level fusion scheme for cancelable multibiometric verification. Appl Intell 49, 1016–1035 (2019). https://doi.org/10.1007/s1048901813112
Published:
Issue Date:
Keywords
 Biometric
 Multibiometric system
 Verification
 fusion
 Decision level fusion
 Security
 Privacy