Figure 1 - available via license: Creative Commons Attribution 4.0 International
Content may be subject to copyright.
Source publication
Fingerprints have been used for decades to verify the identity of an individual for various security reasons. Attackers have developed many approaches to deceive a fingerprint verification system, ranging from the sensor level, where gummy fingers are created, to gaining access to the decision-maker level, where the decision is made based on low ma...
Contexts in source publication
Context 1
... technologies inherit the general known vulnerabilities to attacks on any biometric system, as shown in Figure 1. The first vulnerability is attacking and fooling the sensor, where an attacker provides a fake fingerprint. ...
Context 2
... typical solution to attacks 3-8 is applying the challenge-response method to ensure that the data being received are from a trusted unaltered component of the biometric system. All oval shapes in Figure 1 represent the possible attack, and all rectangular shapes represent a typical biometric component to authenticate a user. All inherited vulnerabilities create the potential for successful attacks on any fingerprint-based biometric system to fool the authentication system and provide an impostor access. ...
Context 3
... is hypothesized to be unique when compared to an attacker and serves as a behavioral feature to differentiate when a user is coerced or willing to provide their fingerprint. Figure 10 depicts the detected features at first touch, mid-touch, and final complete fingertip touch on the glass surface, and it clearly shows the gradual appearance of fingerprint features over time. However, instead of capturing only three instances, all 960 images per second for 5 s were analyzed to provide a detailed appearance of the sequence of fingerprint features and the location and orientation of each feature, bifurcation, and termination of ridges, where each shape, circle triangle or square, depicts the an instance of the fingerprint where new fingerprint features were detected. ...
Context 4
... provides the detail of the fingerprint surface shape as well as the detailed microbehavior of the user. Figure 11 depicts the fingerprint contour detection over time. Finally, the system analyzes the final 1920 frames after a fingerprint is completely placed on the glass surface and reports the micro-movements' directional changes. ...
Context 5
... MFAS then matches the newly acquired micro-behavioral features with those already stored and returns the results. Figure 12 depicts the overall MFAS data analysis as a fully automated fingerprint-based biometric system from the start of fingerprint placement until a decision of access is granted or denied. ...
Context 6
... the results show promise in terms of the capability of the MFAS in supporting the main hypothesis, which states that "Micro-behavior measurement of the fingertip as it is placed on the touch-based sensor surface over time until a fingerprint is fully formed is a valid mechanism to verify if the fingerprint is fake or real and if a user is coerced". Figure 13 depicts the overall results, with and without the MFAS, in the four scenarios. It clearly shows a decrease in the matching score between an attacker using a fake fingerprint and a legitimate user and between a legitimate user when willing to provide their fingerprint and when being forced to do so. ...
Context 7
... clearly shows a decrease in the matching score between an attacker using a fake fingerprint and a legitimate user and between a legitimate user when willing to provide their fingerprint and when being forced to do so. Figure 13. The overall results, with and without the MFAS, in the four scenarios, where 1 is a legitimate enrollment, 2 is for a fake fingerprint used by the legitimate user, 3 is for a fake fingerprint used by an attacker, and 4 is for an unwilling submission of a legitimate user's fingerprint. ...
Context 8
... this improves the functionality in terms of the time of the MFAS, it impacts the accuracy of the detection. Table 5 shows the MFAS results after reducing the number of samples used when analyzing the micro-movement behavior of a fingerprint, and Figure 14 depicts the overall results after the reduction. Table 5. Results of a traditional fingerprint-based biometric system using MFAS OCSVM at a reduced sample size for improving the acceptability of the system. ...
Similar publications
Finger Texture (FT) is currently attracting significant attentions in the area of human recognition. Finger texture covers the area between the lower knuckle of the finger and the upper phalanx before the fingerprint. It involves rich features which can be efficiently used as a biometric characteristic. In this paper, we contribute to this growing...
Citations
The robustness of automatic fingerprint identification systems (AFIS) against presentation attacks, a pressing concern in biometric security, is critically dependent on the efficacy of fingerprint liveness detection (FLD) methodologies. This article presents a systematic review of the latest advancements in FLD, focusing on publications from 2022 to 2023. A meticulous analysis of 27 studies sourced from the Web of Science (WoS) and ScienceDirect databases reveals significant strides in FLD techniques, aimed at enhancing the resilience of AFIS against increasingly sophisticated spoofs made from everyday materials like wood glue, playdoh, and latex. These innovative approaches, encompassing advanced machine learning algorithms, IoT-based multimodal detection, and novel material-based detection methods, reflect a concerted effort to counteract the evolving tactics of fraudulent entities. Despite these technological advancements, the study identifies ongoing challenges that impede the full-proof security of AFIS. These include data privacy concerns in hardware-based systems, the emergence of thin-layered and subtle spoofing methods, the complexities of puppet attacks, and difficulties in cross-material detection, all of which hinder the generalization of live fingerprint identification. The study also delves into the sector-specific implications of these developments in critical domains such as law enforcement, banking, and personal security, underscoring the balance between enhanced security and privacy concerns, especially in systems employing multiple biometric modalities. This study not only highlights the current achievements in FLD but also underscores the necessity for continued research and development. The objective is to address these persisting challenges and to ensure the robustness of biometric security systems in the face of rapidly evolving threats, thereby bolstering their integrity and reliability.
