A non‐invertible transformation based technique to protect a fingerprint template

Abstract

A fingerprint-based authentication system provides security to the applications of numerous fields and usually stores minutiae information in the database as a template. It has been observed from the literature that the reconstruction of an original fingerprint is possible from minutiae points information; hence the security of the stored template becomes extremely crucial. Cancellable biometric techniques based on non-invertible transformation protect the stored template. These techniques prevent the reconstruction of original fingerprint data from the compromised template and avoid unauthorized access to the system. In this paper, a technique based on the non-invertible transformation to protect a fingerprint template is proposed. In the technique, minutiae locations in a fingerprint are transformed by using the minutiae's original locations and orientation information, and a user keyset. A principal component analysis based approach to align the probe and gallery templates of fingerprint images while matching is also proposed. The evaluation of the proposed technique is carried out on seven different fingerprint databases taken from FVC2000, FVC2002, and FVC2004 databases, and its performance is compared with other existing state-of-the-art techniques in the literature. The comparative performance shows that the proposed technique is highly robust and performs exceptionally well compared to other existing techniques.

Biometric Template Attacks and Recent Protection Mechanisms: A Survey

“…This is designed with the notion of achieving a mechanism free from alignment while overcoming brute force and false accept attacks. Baghel et al [78] utilized a non-invertible transformation to specifically combat brute-force and false-accept attacks, while Li et al [76] improved on the concept of IoM hashing to derive an indexing min-max hashing technique to mitigate these attacks. Recently, Abdullahi et al [82] developed a vector permutation and shift-ordering approach to mitigate these attacks.…”

“…Furthermore, in the case of a lost key, a feature-guided index-generating method is designed to guarantee the security of the selected help vector. Other recent techniques used to counter the inversion of fingerprint templates include the works of Baghel et al [78], Agarwal and Bansal [79], Guo et al [80], Trivedi et al [81], and Abdullahi et al [82].…”

“…The EER comparison of 10 protection mechanisms is made in this category. This includes the techniques of Wang et al [74], abbreviated as PHD; Jin et al [75] as IoM; Kho et al [73] as PR-NNLS; Trivedi et al [99] as DT-MP; Badari et al [77] as DKM; Baghel et al [78] as NT; Li et al [76] as IMMH; Yang et al [100] as LC; Agarwal and Bansal [79] as AFN-TF; and Abdullahi et al [82] as VP-SO; on inversion attacks. The techniques of Wang et al [123] as ZMP, Ariona et al [125] as DF-CFM, Abdullahi et al [127] as FC-FMT, Tran and Hu [121] as M-FF, Shahzad et al [122] as WS-PDWT, and Yang et al [128] as FARP on ARM attacks.…”

“…11(c) and 11(d) depicts the performance of techniques against inversion and ARM on FVC2004 datasets, respectively. It can be seen that the most recently published techniques, including AFN-TF [79], VP-SO [82], and IMMH [76], designed to mitigate inversion attacks, achieve better recognition accuracy (performance preservation) in both datasets than the other techniques, while PR-NNLS [73] and NT [78] have the least performance, especially on DB3 of FVC2002 and DB2 of FVC2004, respectively. However, for techniques designed to mitigate ARM attacks, M-FF [121], FC-FMT [127], and DF-CFM [125] achieve better performance than other techniques in the same category on both FVC datasets.…”

Optimized Authentication System with High Security and Privacy

“…As there are numerous issues related to fingerprint biometrics [33], there is a need for a highly secure fingerprint-based authentication method. Any secure biometric system is expected to depict the essential requirements [19,34] given below:…”