{"title":"IoT-based multimodal liveness detection using the fusion of ECG and fingerprint","authors":"A. Gona, Subramoniam M.","doi":"10.1108/ijpcc-10-2021-0248","DOIUrl":null,"url":null,"abstract":"\nPurpose\nBiometric scans using fingerprints are widely used for security purposes. Eventually, for authentication purposes, fingerprint scans are not very reliable because they can be faked by obtaining a sample of the fingerprint of the person. There are a few spoof detection techniques available to reduce the incidence of spoofing of the biometric system. Among them, the most commonly used is the binary classification technique that detects real or fake fingerprints based on the fingerprint samples provided during training. However, this technique fails when it is provided with samples formed using other spoofing techniques that are different from the spoofing techniques covered in the training samples. This paper aims to improve the liveness detection accuracy by fusing electrocardiogram (ECG) and fingerprint.\n\n\nDesign/methodology/approach\nIn this paper, to avoid this limitation, an efficient liveness detection algorithm is developed using the fusion of ECG signals captured from the fingertips and fingerprint data in Internet of Things (IoT) environment. The ECG signal will ensure the detection of real fingerprint samples from fake ones.\n\n\nFindings\nSingle model fingerprint methods have some disadvantages, such as noisy data and position of the fingerprint. To overcome this, fusion of both ECG and fingerprint is done so that the combined data improves the detection accuracy.\n\n\nOriginality/value\nSystem security is improved in this approach, and the fingerprint recognition rate is also improved. IoT-based approach is used in this work to reduce the computation burden of data processing systems.\n","PeriodicalId":43952,"journal":{"name":"International Journal of Pervasive Computing and Communications","volume":null,"pages":null},"PeriodicalIF":0.6000,"publicationDate":"2022-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Journal of Pervasive Computing and Communications","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1108/ijpcc-10-2021-0248","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS","Score":null,"Total":0}
引用次数: 0
Abstract
Purpose
Biometric scans using fingerprints are widely used for security purposes. Eventually, for authentication purposes, fingerprint scans are not very reliable because they can be faked by obtaining a sample of the fingerprint of the person. There are a few spoof detection techniques available to reduce the incidence of spoofing of the biometric system. Among them, the most commonly used is the binary classification technique that detects real or fake fingerprints based on the fingerprint samples provided during training. However, this technique fails when it is provided with samples formed using other spoofing techniques that are different from the spoofing techniques covered in the training samples. This paper aims to improve the liveness detection accuracy by fusing electrocardiogram (ECG) and fingerprint.
Design/methodology/approach
In this paper, to avoid this limitation, an efficient liveness detection algorithm is developed using the fusion of ECG signals captured from the fingertips and fingerprint data in Internet of Things (IoT) environment. The ECG signal will ensure the detection of real fingerprint samples from fake ones.
Findings
Single model fingerprint methods have some disadvantages, such as noisy data and position of the fingerprint. To overcome this, fusion of both ECG and fingerprint is done so that the combined data improves the detection accuracy.
Originality/value
System security is improved in this approach, and the fingerprint recognition rate is also improved. IoT-based approach is used in this work to reduce the computation burden of data processing systems.