Densely Connected Convolutional Network Optimized by Genetic Algorithm for Fingerprint Liveness Detection

Wen, Jian; Zhou, Yujie; Li, Hongming

doi:10.1109/access.2020.3047723

Cited by 30 publications

(10 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Zhang et al [126] also tested their proposed method against attacks from new materials and achieved an ACE of 3.31% on the LivDet 2015.…”

Section: Generalization Efficient/wrapper Methodsmentioning

confidence: 99%

“…Zhang et al [126] proposed a lightweight CNN known as FLD to achieve improved PAD on new materials and to minimize complexity. To address the issue of global average pooling, an attention pooling layer was used.…”

Section: Neural Networkmentioning

confidence: 99%

“…[126] 2020 LivDet 2013, 2015 CNN ACE of 1.76% over all sensors ACE of 0.25% on the LivDet 2013Jian et al…”

mentioning

confidence: 99%

See 2 more Smart Citations

A comprehensive survey of fingerprint presentation attack detection

Κarampidis¹,

Rousouliotis²,

Linardos³

et al. 2021

JSSS

View full text Add to dashboard Cite

“…Zhang et al [126] also tested their proposed method against attacks from new materials and achieved an ACE of 3.31% on the LivDet 2015.…”

Section: Generalization Efficient/wrapper Methodsmentioning

confidence: 99%

Section: Neural Networkmentioning

confidence: 99%

See 1 more Smart Citation

A comprehensive survey of fingerprint presentation attack detection

Κarampidis¹,

Rousouliotis²,

Linardos³

et al. 2021

JSSS

View full text Add to dashboard Cite

“…Another type of PDMS phantoms cast from patterns etched in silicon has been applied for the development of ultrasound scanners . Conductive PDMS inconjunction with 3D printing technique has been recently adapted to create fingerprint phantoms as well. , Further advancement in the accuracy and complexity of fingerprint phantoms (such as including simulated subdermal blood vessels) will essentially accelerate the development of next-generation scanners and matching algorithms with improved reliability and security. − …”

Section: Introductionmentioning

confidence: 99%

Three-Dimensional Conductive Fingerprint Phantoms Made of Ethylene-Vinyl Acetate/Graphene Nanocomposite for Evaluating Smartphone Scanners

Schultz

Fawzy

Nasirpouri

et al. 2021

ACS Appl. Electron. Mater.

View full text Add to dashboard Cite

Fingerprints consist of unique patterns of skin ridges and valleys, and are commonly described by three levels of feature details. Most fingerprint identification systems rely on matching only first- and second-level details; their calibration using unrealistic targets combined with human sampling are cumbersome and time-consuming. We have developed a true-to-life fingerprint phantom with optimized conductive properties and third-level fingerprint details for developing more reliable matching algorithms for popular capacitive fingerprint scanners. An impression of a live finger is made into a solvent-softened polycarbonate template, which is subsequently adapted to thermally mold ethylene-vinyl acetategraphene (EVA-G) nanocomposite of satisfactory conductivity (1.2 ± 0.1 × 10–3 S/m) and mechanical flexibility. It was confirmed with SEM, optical, and profilometry imaging that the phantom made of EVA-G nanocomposite replicates the three-dimensional morphology of fingerprint with high fidelity, including well-defined third-level details. The EVA-G phantoms can operate capacitive scanners in popular brands of smartphones and tablets with either iOS or Android operating systems (>80% success rate). As these permanent phantoms include defined third-level details based on real fingerprints, their potential application as calibration standards for developing the next generation of scanners with improved security is warranted.

show abstract

“…Neuroevolution could be used by researchers from other fields without requiring advanced expertise in DL. For instance, neuroevolution has been applied successfully to medical images [32]- [36], speech recognition [37]- [39], emotion recognition [40], [41], scene classification [42], among others [43]- [51].…”

Section: Introductionmentioning

confidence: 99%

Two-Level Genetic Algorithm for Evolving Convolutional Neural Networks for Pattern Recognition

Montecino

Perez²,

Bowyer

2021

IEEE Access

View full text Add to dashboard Cite

The aim of Neuroevolution is to find neural networks and convolutional neural network (CNN) architectures automatically through evolutionary algorithms. A crucial problem in neuroevolution is search time, since multiple CNNs must be trained during evolution. This problem has led to fitness acceleration approaches, generating a trade-off between time and fitness fidelity. Also, since search spaces for this problem usually include only a few parameters, this increases the human bias in the search. In this work, we propose a novel two-level genetic algorithm (GA) for addressing the fidelity-time trade-off problem for the fitness computation in CNNs. The first level evaluates many individuals quickly, and the second evaluates only those with the best results more finely. We also propose a search space with few restrictions, and an encoding with unexpressed genes to facilitate the crossover operation. This search space allows CNN architectures to have any sizes, shapes, and skip-connections among nodes. The two-level GA was applied to the pattern recognition problem on seven datasets, five MNIST-Variants, Fashion-MNIST, and CIFAR-10, achieving significantly better results than all those previously published. Our results show an improvement of 39.89% (4.2% error reduction) on the most complex dataset of MNIST (MRDBI), and on average 30.52% (1.35% error reduction) on all the five datasets. Furthermore, we show that our algorithm performed as well as precise-training GA, but took only the time of a fast-training GA. These results can be relevant and useful not only for image classification problems but also for GA-related problems.

show abstract

Densely Connected Convolutional Network Optimized by Genetic Algorithm for Fingerprint Liveness Detection

Cited by 30 publications

References 34 publications

A comprehensive survey of fingerprint presentation attack detection

A comprehensive survey of fingerprint presentation attack detection

Three-Dimensional Conductive Fingerprint Phantoms Made of Ethylene-Vinyl Acetate/Graphene Nanocomposite for Evaluating Smartphone Scanners

Two-Level Genetic Algorithm for Evolving Convolutional Neural Networks for Pattern Recognition

Contact Info

Product

Resources

About