LivDet in Action - Fingerprint Liveness Detection Competition 2019

Orrù, Giulia; Casula, Roberto; Tuveri, Pierluigi; Bazzoni, Carlotta; Dessalvi, Giovanna; Micheletto, Marco; Ghiani, Luca; Marcialis, Gian Luca

doi:10.1109/icb45273.2019.8987281

Cited by 52 publications

(58 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The experimental evaluation shows the capacity of the new method to be used in high security applications. In addition, we would like to highlight that the proposed method in the Fingerprint Liveness Detection Competition 2019 achieved the best detection performance with an average accuracy of 96.17% [29]. This database was designed to evaluate the PAD generalisation performance over unknown PAIs.…”

Section: Proposed Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Fingerprint Presentation Attack Detection Based on Local Features Encoding for Unknown Attacks

et al. 2021

View full text Add to dashboard Cite

Fingerprint-based biometric systems have experienced a large development in the past. In spite of many advantages, they are still vulnerable to attack presentations (APs). Therefore, the task of determining whether a sample stems from a live subject (i.e., bona fide) or from an artificial replica is a mandatory requirement which has recently received a considerable attention. Nowadays, when the materials for the fabrication of the Presentation Attack Instruments (PAIs) have been used to train the Presentation Attack Detection (PAD) methods, the PAIs can be successfully identified in most cases. However, current PAD methods still face difficulties detecting PAIs built from unknown materials and/or unknown recepies, or acquired using different capture devices. To tackle this issue, we propose a new PAD technique based on three image representation approaches combining local and global information of the fingerprint. By transforming these representations into a common feature space, we can correctly discriminate bona fide from attack presentations in the aforementioned scenarios. The experimental evaluation of our proposal over the LivDet 2011 to 2019 databases, yielded error rates outperforming the top state-of-the-art results by up to 72% in the most challenging scenarios. In addition, the best representation achieved the best results in the LivDet 2019 competition (overall accuracy of 96.17%).INDEX TERMS Local feature encoding, presentation attack detection, fingerprint, probabilistic visual vocabulary, visual vocabulary.

show abstract

Section: Proposed Methodsmentioning

confidence: 99%

“…The experiments were conducted on the well-established benchmarks from LivDet 2011 [41], LivDet 2013 [42], LivDet 2015 [43], LivDet 2017 [31], and LivDet 2019 [44]. A summary of the PAI fabrication materials is included in Tab.…”

Section: ) Databasesmentioning

confidence: 99%

Fingerprint Presentation Attack Detection Based on Local Features Encoding for Unknown Attacks

et al. 2021

View full text Add to dashboard Cite

show abstract

“…In response to this growing threat, a series of fingerprint Liveness Detection (LivDet) competitions [14] have been held since 2009 to benchmark various spoof detection solutions. See [15] for results of the LivDet 2019. Another initiative is the IARPA ODIN Program [4] with the goal of developing robust spoof detection systems for fingerprints, face, and iris 2 https://www.bbc.com/news/world-latin-america-21756709 3 http://statenews.com/article/2016/08/how-msu-researchers-unlocked-afingerprint-secure-smartphone-to-help-police-with-homicide-case 4 https://www.medianama.com/2018/03/223-cloned-thumb-prints-used-tospoof-biometrics-and-allow-proxies-to-answer-online-rajasthan-police-exam/ 5 https://imgur.com/gallery/8aGqsSu 6 http://fortune.com/2016/04/07/guy-unlocked-iphone-play-doh/ 7 https://srlabs.de/bites/spoofing-fingerprints/ biometric modalities.…”

Section: Synthetic Livementioning

confidence: 99%

Fingerprint Spoof Detector Generalization

Chugh

Jain

2021

IEEE Trans.Inform.Forensic Secur.

View full text Add to dashboard Cite

We present a style-transfer based wrapper, called Universal Material Generator (UMG), to improve the generalization performance of any fingerprint spoof detector against spoofs made from materials not seen during training. Specifically, we transfer the style (texture) characteristics between fingerprint images of known materials with the goal of synthesizing fingerprint images corresponding to unknown materials, that may occupy the space between the known materials in the deep feature space. Synthetic live fingerprint images are also added to the training dataset to force the CNN to learn generative-noise invariant features which discriminate between lives and spoofs. The proposed approach is shown to improve the generalization performance of a state-of-the-art spoof detector, namely Fingerprint Spoof Buster, from TDR of 75.24% to 91.78% @ FDR = 0.2%. These results are based on a large-scale dataset of 5,743 live and 4,912 spoof images fabricated using 12 different materials. Additionally, the UMG wrapper is shown to improve the average cross-sensor spoof detection performance from 67.60% to 80.63% when tested on the LivDet 2017 dataset. Training the UMG wrapper requires only 100 live fingerprint images from the target sensor, alleviating the time and resources required to generate large-scale live and spoof datasets for a new sensor. We also fabricate physical spoof artifacts using a mixture of known spoof materials to explore the role of cross-material style transfer in improving generalization performance.

show abstract

“…Spoofing detection performance of different fingerprint algorithms and systems has been studied and compared during Liveness detection competition (LivDet) [ 26 , 27 , 28 , 29 ] since 2009. This competition created a benchmark for measuring liveness detection algorithms.…”

Section: Introductionmentioning

confidence: 99%

“…In 2015, the LivDet [ 28 ] competition introduced cross-material validation. In 2019, LivDet [ 26 ] focused on investigation of how the integration of a liveness detection system with a fingerprint authentication system may impact the performance. Therefore, the spoofing detection algorithms were evaluated together with sensors.…”

Section: Introductionmentioning

confidence: 99%

Towards Fingerprint Spoofing Detection in the Terahertz Range

Pałka

Kowalski

2020

Sensors

View full text Add to dashboard Cite

Spoofing attacks using imitations of fingerprints of legal users constitute a serious threat. In this study, a terahertz time domain spectroscopy (TDS) setup in a reflection configuration was used for the non-intrusive detection of fingerprint spoofing. Herein, the skin structure of the finger pad is described with a focus on the outermost stratum corneum. We identified and characterized five representative spoofing materials and prepared thin and thick finger imitations. The complex refractive index of the materials was determined in TDS in the transmission configuration. For dataset collection, we selected a group of 16 adults of various ages and genders. The reflection results were analyzed both in the time (reflected signal) and frequency (reflectivity) domains. The measured signals were positively verified with the theoretical calculations. The signals corresponding to samples differ from the finger-related signals, which facilitates spoofing detection. Thanks to deconvolution, we provide a basic explanation of the observed phenomena. We propose two spoofing detection methods, predefined time–frequency features and deep learning based. The methods achieved high true detection rates of 87.9% and 98.8%. Our results show that the terahertz technology can be successfully applied for spoofing detection with high detection probability.

show abstract

LivDet in Action - Fingerprint Liveness Detection Competition 2019

Cited by 52 publications

References 7 publications

Fingerprint Presentation Attack Detection Based on Local Features Encoding for Unknown Attacks

Fingerprint Presentation Attack Detection Based on Local Features Encoding for Unknown Attacks

Fingerprint Spoof Detector Generalization

Towards Fingerprint Spoofing Detection in the Terahertz Range

Contact Info

Product

Resources

About