2017
DOI: 10.1007/978-981-10-4154-9_39
|View full text |Cite
|
Sign up to set email alerts
|

Fingerprint Spoof Detection Using Contrast Enhancement and Convolutional Neural Networks

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
19
0
1

Year Published

2017
2017
2022
2022

Publication Types

Select...
5
3

Relationship

0
8

Authors

Journals

citations
Cited by 28 publications
(20 citation statements)
references
References 9 publications
0
19
0
1
Order By: Relevance
“…In 2017, Jang et al proposed contrast enhancement and block-wise processing of the fingerprint to improve the state of the art results achieved with DL [25]. The blocks were then combined with a majority voting rule.…”
Section: B Deep Learning For Conventional Sensorsmentioning
confidence: 99%
“…In 2017, Jang et al proposed contrast enhancement and block-wise processing of the fingerprint to improve the state of the art results achieved with DL [25]. The blocks were then combined with a majority voting rule.…”
Section: B Deep Learning For Conventional Sensorsmentioning
confidence: 99%
“…Histogram equalization is performed on each RGB channel in order to increase the contrast and improve the network performance. 41 Then the images are normalized to the [0, 1] range. Normalization is typically performed in neural networks because the nonlinear functions that are employed work better in this range.…”
Section: Data Preprocessingmentioning
confidence: 99%
“…The new era of the image classification problem has started with the birth of Deep Convolutional Neural Networks (DCNN). DCNN have rapidly shown their effectiveness in fingerprint liveness detection [ 24 , 25 , 26 , 27 , 28 , 29 ]. There are two approaches to use convolutional neural networks in fake fingerprint detection.…”
Section: Literature Review On Presentation Attack Detectionmentioning
confidence: 99%
“…The final decision is made by the voting strategy, i.e., if the number of fake patches is greater than or equal to that of live patches, the fingerprint is fake. Wang et al [ 27 ], Jang et al [ 28 ] and Park et al [ 29 ] experimented with different network models and different patch sizes. Wang et al [ 27 ] divided a fingerprint image into non-overlapped patches of a size of pixels and then used a four-layer CNN to classify each patch.…”
Section: Literature Review On Presentation Attack Detectionmentioning
confidence: 99%
See 1 more Smart Citation