Editing Biometrics (section)

==Multimodal biometric system==
Multimodal biometric systems use multiple sensors or biometrics to overcome the limitations of unimodal biometric systems.<ref name="dca">{{Cite journal|url=https://dx.doi.org/10.1109/TIFS.2016.2569061|doi = 10.1109/TIFS.2016.2569061|title = Discriminant Correlation Analysis: Real-Time Feature Level Fusion for Multimodal Biometric Recognition|year = 2016|last1 = Haghighat|first1 = Mohammad|last2 = Abdel-Mottaleb|first2 = Mohamed|last3 = Alhalabi|first3 = Wadee|journal = IEEE Transactions on Information Forensics and Security|volume = 11|issue = 9|pages = 1984–1996|s2cid = 15624506|url-access = subscription}}</ref> For instance iris recognition systems can be compromised by aging irises<ref>{{Cite news | title = Questions Raised About Iris Recognition Systems | url = http://sciencedaily.com/releases/2012/07/120712141938.htm | date = 12 July 2012 | work = Science Daily | url-status = live | archive-url = https://web.archive.org/web/20121022051152/http://www.sciencedaily.com/releases/2012/07/120712141938.htm | archive-date = 22 October 2012}}</ref> and [[electronic fingerprint recognition]] can be worsened by worn-out or cut fingerprints. While unimodal biometric systems are limited by the integrity of their identifier, it is unlikely that several unimodal systems will suffer from identical limitations. Multimodal biometric systems can obtain sets of information from the same marker (i.e., multiple images of an iris, or scans of the same finger) or information from different biometrics (requiring fingerprint scans and, using [[speaker recognition|voice recognition]], a spoken passcode).<ref>{{cite book |title=The Mobile Wave: How Mobile Intelligence Will Change Everything |last=Saylor |first=Michael |year=2012 |publisher=Perseus Books/Vanguard Press |page=99 |url=https://books.google.com/books?id=gJqxM8sAJmIC|isbn=9780306822988 }}</ref><ref>{{Cite news | title = This is the 'biometric war' Michael Saylor was talking about | url = http://www.bizjournals.com/washington/blog/techflash/2013/10/this-is-the-biometric-war-michael.html | date = 3 October 2013 | work = Washington Business Journal | author = Bill Flook | url-status = live | archive-url = https://web.archive.org/web/20131007031912/http://www.bizjournals.com/washington/blog/techflash/2013/10/this-is-the-biometric-war-michael.html | archive-date = 7 October 2013}}</ref>

Multimodal biometric systems can fuse these unimodal systems sequentially, simultaneously, a combination thereof, or in series, which refer to sequential, parallel, hierarchical and serial integration modes, respectively.
Fusion of the biometrics information can occur at different stages of a recognition system. In case of feature level fusion, the data itself or the features extracted from multiple biometrics are fused. Matching-score level fusion consolidates the scores generated by multiple [[classification rule|classifiers]] pertaining to different modalities. Finally, in case of decision level fusion the final results of multiple classifiers are combined via techniques such as [[majority voting]]. Feature level fusion is believed to be more effective than the other levels of fusion because the feature set contains richer information about the input biometric data than the matching score or the output decision of a classifier. Therefore, fusion at the feature level is expected to provide better recognition results.<ref name=dca />

Furthermore, the evolving biometric market trends underscore the importance of technological integration, showcasing a shift towards combining multiple biometric modalities for enhanced security and identity verification, aligning with the advancements in multimodal biometric systems.<ref>{{cite web |title=What is Biometrics? Definition, Data Types, Trends (2024) |url=https://www.aratek.co/news/what-is-biometrics-definition-data-types-trends |website=Aratek Biometrics |access-date=11 April 2024}}</ref>

[[Spoof attack]]s consist in submitting fake biometric traits to biometric systems, and are a major threat that can curtail their security. Multi-modal biometric systems are commonly believed to be intrinsically more robust to spoof attacks, but recent studies<ref>Zahid Akhtar, [http://pralab.diee.unica.it/sites/default/files/Akhtar_PhD2012.pdf "Security of Multimodal Biometric Systems against Spoof Attacks"] (PDF). {{webarchive|url=https://web.archive.org/web/20150402104529/http://pralab.diee.unica.it/sites/default/files/Akhtar_PhD2012.pdf |date=2 April 2015 }}. Department of Electrical and Electronic Engineering, University of Cagliari. Cagliari, Italy, 6 March 2012.</ref> have shown that they can be evaded by spoofing even a single biometric trait.

One such proposed system of Multimodal Biometric Cryptosystem Involving the Face, Fingerprint, and Palm Vein by Prasanalakshmi<ref>Prasanalakshmi,[https://citeseerx.ist.psu.edu/document?repid=rep1&type=pdf&doi=803002976e55927583ef8a57b9586c50b4398a11"Multimodal Biometric Cryptosystem Involving Face, Fingerprint, and Palm Vein"], July 2011 </ref> The [[Cryptosystem]] Integration combines biometrics with [[cryptography]], where the palm vein acts as a cryptographic key, offering a high level of security since palm veins are unique and difficult to forge. The [[Fingerprint]] Involves minutiae extraction (terminations and bifurcations) and matching techniques. Steps include image enhancement, binarization, [[Region_of_interest|ROI]] extraction, and minutiae thinning. The Face system uses class-based scatter matrices to calculate features for recognition, and the [[Vein_matching#Palm_vein_biometrics_key_generation|Palm Vein acts as an unbreakable cryptographic key]], ensuring only the correct user can access the system. The cancelable Biometrics concept allows biometric traits to be altered slightly to ensure privacy and avoid theft. If compromised, new variations of biometric data can be issued.
The Encryption fingerprint template is encrypted using the palm vein key via [[XOR_gate|XOR]] operations. This encrypted Fingerprint is hidden within the face image using steganographic techniques. Enrollment and Verification for the Biometric data (Fingerprint, palm vein, face) are captured, encrypted, and embedded into a face image. The system extracts the biometric data and compares it with stored values for Verification. The system was tested with fingerprint databases, achieving 75% verification accuracy at an equal error rate of 25% and processing time approximately 50 seconds for enrollment and 22 seconds for Verification. High security due to palm vein encryption, effective against biometric spoofing, and the multimodal approach ensures reliability if one biometric fails. Potential for integration with [[Smart_card|smart cards]] or on-card systems, enhancing security in [[Identity_document|personal identification]] systems.