Skin identification using deep convolutional neural network

Mahdi Maktab Dar Oghaz; Vasileios Argyriou; Dorothy Monekosso; Paolo Remagnino

doi:10.1007/978-3-030-33720-9_14

Skin identification using deep convolutional neural network

Mahdi Maktab Dar Oghaz
, Vasileios Argyriou
, Dorothy Monekosso
, Paolo Remagnino

Research output: Contribution to conference › Paper › peer-review

Abstract

Skin identification can be used in several security applications such as border‘s security checkpoints and facial recognition in bio-metric systems. Traditional skin identification techniques were unable to deal with the high complexity and uncertainty of human skin in uncontrolled environments. To address this gap, this research proposes a new skin identification technique using deep convolutional neural network. The proposed sequential deep model consists of three blocks of convolutional layers, followed by a series of fully connected layers, optimized to maximize skin texture classification accuracy. The proposed model performance has been compared with some of the well-known texture-based skin identification techniques and delivered superior results in terms of overall accuracy. The experiments were carried out over two datasets including FSD Benchmark dataset as well as an in-house skin texture patch dataset. Results show that the proposed deep skin identification model with highest reported accuracy of 0.932 and minimum loss of 0.224 delivers reliable and robust skin identification.

Original language	English
DOIs	https://doi.org/10.1007/978-3-030-33720-9_14
Publication status	Published - Oct 2019
Event	14th International Symposium on Visual Computing, ISVC 2019 - Lake Tahoe, NV, U.S. Duration: 7 Oct 2019 → 9 Oct 2019

Conference

Conference	14th International Symposium on Visual Computing, ISVC 2019
Period	7/10/19 → 9/10/19

Bibliographical note

Note: Published in: Bebis, George, Boyle, Richard, Parvin, Bahram, Koracin, Darko, Ushizima, Daniela, Chai, Sek, Sueda, Shinjiro, Lin, Xin, Lu, Aidong, Thalmann, Daniel, Wang, Chaoli, Xu, Panpan (eds.) (2019) 14th International Symposium on Visual Computing, ISVC 2019, Lake Tahoe, NV, USA, October 7-9, 2019, Proceedings, Part I. Cham, Switzerland : Springer International Publishing. pp. 181-193. (Lecture Notes in Computer Science, no. 11844, part of the Image Processing, Computer Vision, Pattern Recognition, and Graphics book sub-series, LNIP, volume 11844) ISSN (print) 0302-9743 ISBN 9783030337193.

Organising Body: International Symposium on Visual Computing (ISVC)

Keywords

Computer science and informatics
Convolutional Neural Networks
Deep learning Segmentation
Skin texture analysis

Access to Document

10.1007/978-3-030-33720-9_14

https://doi.org/10.1007/978-3-030-33720-9_14

1 Conference contribution

Skin identification using deep convolutional neural network
Argyriou, V., Monekosso, D., Remagnino, P. & Maktab Dar Oghaz, M., Oct 2019, Published in: Bebis, George, Boyle, Richard, Parvin, Bahram, Koracin, Darko, Ushizima, Daniela, Chai, Sek, Sueda, Shinjiro, Lin, Xin, Lu, Aidong, Thalmann, Daniel, Wang, Chaoli, Xu, Panpan (eds.) (2019) 14th International Symposium on Visual Computing, ISVC 2019, Lake Tahoe, NV, USA, October 7-9, 2019, Proceedings, Part I. Cham, Switzerland : Springer International Publishing. pp. 181-193. (Lecture Notes in Computer Science, no. 11844, part of the Image Processing, Computer Vision, Pattern Recognition, and Graphics book sub-series, LNIP, volume 11844) ISSN (print) 0302-9743 ISBN 9783030337193. Organising Body: International Symposium on Visual Computing (ISVC) Organising Body: International Symposium on Visual Computing (ISVC).
Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Cite this

@conference{70d1b248c7494c65b2bc085b13f74053,

title = "Skin identification using deep convolutional neural network",

abstract = "Skin identification can be used in several security applications such as border{\textquoteleft}s security checkpoints and facial recognition in bio-metric systems. Traditional skin identification techniques were unable to deal with the high complexity and uncertainty of human skin in uncontrolled environments. To address this gap, this research proposes a new skin identification technique using deep convolutional neural network. The proposed sequential deep model consists of three blocks of convolutional layers, followed by a series of fully connected layers, optimized to maximize skin texture classification accuracy. The proposed model performance has been compared with some of the well-known texture-based skin identification techniques and delivered superior results in terms of overall accuracy. The experiments were carried out over two datasets including FSD Benchmark dataset as well as an in-house skin texture patch dataset. Results show that the proposed deep skin identification model with highest reported accuracy of 0.932 and minimum loss of 0.224 delivers reliable and robust skin identification.",

keywords = "Computer science and informatics, Convolutional Neural Networks, Deep learning Segmentation, Skin texture analysis",

author = "Oghaz, \{Mahdi Maktab Dar\} and Vasileios Argyriou and Dorothy Monekosso and Paolo Remagnino",

note = "Note: Published in: Bebis, George, Boyle, Richard, Parvin, Bahram, Koracin, Darko, Ushizima, Daniela, Chai, Sek, Sueda, Shinjiro, Lin, Xin, Lu, Aidong, Thalmann, Daniel, Wang, Chaoli, Xu, Panpan (eds.) (2019) 14th International Symposium on Visual Computing, ISVC 2019, Lake Tahoe, NV, USA, October 7-9, 2019, Proceedings, Part I. Cham, Switzerland : Springer International Publishing. pp. 181-193. (Lecture Notes in Computer Science, no. 11844, part of the Image Processing, Computer Vision, Pattern Recognition, and Graphics book sub-series, LNIP, volume 11844) ISSN (print) 0302-9743 ISBN 9783030337193. Organising Body: International Symposium on Visual Computing (ISVC); 14th International Symposium on Visual Computing, ISVC 2019 ; Conference date: 07-10-2019 Through 09-10-2019",

year = "2019",

month = oct,

doi = "10.1007/978-3-030-33720-9\_14",

language = "English",

}

TY - CONF

T1 - Skin identification using deep convolutional neural network

AU - Oghaz, Mahdi Maktab Dar

AU - Argyriou, Vasileios

AU - Monekosso, Dorothy

AU - Remagnino, Paolo

N1 - Note: Published in: Bebis, George, Boyle, Richard, Parvin, Bahram, Koracin, Darko, Ushizima, Daniela, Chai, Sek, Sueda, Shinjiro, Lin, Xin, Lu, Aidong, Thalmann, Daniel, Wang, Chaoli, Xu, Panpan (eds.) (2019) 14th International Symposium on Visual Computing, ISVC 2019, Lake Tahoe, NV, USA, October 7-9, 2019, Proceedings, Part I. Cham, Switzerland : Springer International Publishing. pp. 181-193. (Lecture Notes in Computer Science, no. 11844, part of the Image Processing, Computer Vision, Pattern Recognition, and Graphics book sub-series, LNIP, volume 11844) ISSN (print) 0302-9743 ISBN 9783030337193. Organising Body: International Symposium on Visual Computing (ISVC)

PY - 2019/10

Y1 - 2019/10

N2 - Skin identification can be used in several security applications such as border‘s security checkpoints and facial recognition in bio-metric systems. Traditional skin identification techniques were unable to deal with the high complexity and uncertainty of human skin in uncontrolled environments. To address this gap, this research proposes a new skin identification technique using deep convolutional neural network. The proposed sequential deep model consists of three blocks of convolutional layers, followed by a series of fully connected layers, optimized to maximize skin texture classification accuracy. The proposed model performance has been compared with some of the well-known texture-based skin identification techniques and delivered superior results in terms of overall accuracy. The experiments were carried out over two datasets including FSD Benchmark dataset as well as an in-house skin texture patch dataset. Results show that the proposed deep skin identification model with highest reported accuracy of 0.932 and minimum loss of 0.224 delivers reliable and robust skin identification.

AB - Skin identification can be used in several security applications such as border‘s security checkpoints and facial recognition in bio-metric systems. Traditional skin identification techniques were unable to deal with the high complexity and uncertainty of human skin in uncontrolled environments. To address this gap, this research proposes a new skin identification technique using deep convolutional neural network. The proposed sequential deep model consists of three blocks of convolutional layers, followed by a series of fully connected layers, optimized to maximize skin texture classification accuracy. The proposed model performance has been compared with some of the well-known texture-based skin identification techniques and delivered superior results in terms of overall accuracy. The experiments were carried out over two datasets including FSD Benchmark dataset as well as an in-house skin texture patch dataset. Results show that the proposed deep skin identification model with highest reported accuracy of 0.932 and minimum loss of 0.224 delivers reliable and robust skin identification.

KW - Computer science and informatics

KW - Convolutional Neural Networks

KW - Deep learning Segmentation

KW - Skin texture analysis

U2 - 10.1007/978-3-030-33720-9_14

DO - 10.1007/978-3-030-33720-9_14

M3 - Paper

T2 - 14th International Symposium on Visual Computing, ISVC 2019

Y2 - 7 October 2019 through 9 October 2019

ER -

Skin identification using deep convolutional neural network

Abstract

Conference

Bibliographical note

Keywords

Access to Document

Fingerprint

Research output

Skin identification using deep convolutional neural network

Cite this