Helmet Detection System for Motorcycle Riders with Explainable Artificial Intelligence Using Convolutional Neural Network and Grad-CAM

Suradej Intagorn; Suriya Pinitkan; Mathuros Panmuang; Chonnikarn Rodmorn

doi:10.1007/978-3-031-20992-5_4

Helmet Detection System for Motorcycle Riders with Explainable Artificial Intelligence Using Convolutional Neural Network and Grad-CAM

Suradej Intagorn
, Suriya Pinitkan
, Mathuros Panmuang
, Chonnikarn Rodmorn

Kasetsart University, Kamphaeng Saen Campus
Rajamangala Institute of Technology
King Mongkut's University of Technology North Bangkok

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

3 Citations (Scopus)

Abstract

Motorcycle accidents are one of the most common causes of injury and death in road users. This research has applied convolutional neural network (CNN) and explainable AI to detect motorcyclist without helmet and explain why CNN made that decision. The concept is based on deep learning and CNN principles applied to automatically detect motorcycle riders without helmet from images using three baseline classifiers: support vector machine, random forest and logistic regression to compare with deep convolutional neural network and measure accuracy with accuracy and F1-Score. The results revealed that CNN's F1-Score was 0.8326, which was the highest among all predictive models.

Original language	English
Title of host publication	Multi-disciplinary Trends in Artificial Intelligence - 15th International Conference, MIWAI 2022, Proceedings
Editors	Olarik Surinta, Kevin Kam Fung Yuen
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	40-51
Number of pages	12
ISBN (Print)	9783031209918
DOIs	https://doi.org/10.1007/978-3-031-20992-5_4
Publication status	Published - 2022
Externally published	Yes
Event	15th Multi-disciplinary International Conference on Artificial Intelligence, MIWAI 2022 - Virtual, Online Duration: 17 Nov 2022 → 19 Nov 2022

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	13651 LNAI
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	15th Multi-disciplinary International Conference on Artificial Intelligence, MIWAI 2022
City	Virtual, Online
Period	17/11/22 → 19/11/22

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being

Keywords

Convolutional neural network
Explainable AI
Helmet

Access to Document

10.1007/978-3-031-20992-5_4

Cite this

Intagorn, S., Pinitkan, S., Panmuang, M., & Rodmorn, C. (2022). Helmet Detection System for Motorcycle Riders with Explainable Artificial Intelligence Using Convolutional Neural Network and Grad-CAM. In O. Surinta, & K. Kam Fung Yuen (Eds.), Multi-disciplinary Trends in Artificial Intelligence - 15th International Conference, MIWAI 2022, Proceedings (pp. 40-51). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 13651 LNAI). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-031-20992-5_4

Intagorn, Suradej ; Pinitkan, Suriya ; Panmuang, Mathuros et al. / Helmet Detection System for Motorcycle Riders with Explainable Artificial Intelligence Using Convolutional Neural Network and Grad-CAM. Multi-disciplinary Trends in Artificial Intelligence - 15th International Conference, MIWAI 2022, Proceedings. editor / Olarik Surinta ; Kevin Kam Fung Yuen. Springer Science and Business Media Deutschland GmbH, 2022. pp. 40-51 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{4f06e24b3414406980b7a889c5b6c97c,

title = "Helmet Detection System for Motorcycle Riders with Explainable Artificial Intelligence Using Convolutional Neural Network and Grad-CAM",

abstract = "Motorcycle accidents are one of the most common causes of injury and death in road users. This research has applied convolutional neural network (CNN) and explainable AI to detect motorcyclist without helmet and explain why CNN made that decision. The concept is based on deep learning and CNN principles applied to automatically detect motorcycle riders without helmet from images using three baseline classifiers: support vector machine, random forest and logistic regression to compare with deep convolutional neural network and measure accuracy with accuracy and F1-Score. The results revealed that CNN's F1-Score was 0.8326, which was the highest among all predictive models.",

keywords = "Convolutional neural network, Explainable AI, Helmet",

author = "Suradej Intagorn and Suriya Pinitkan and Mathuros Panmuang and Chonnikarn Rodmorn",

note = "Publisher Copyright: {\textcopyright} 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.; 15th Multi-disciplinary International Conference on Artificial Intelligence, MIWAI 2022 ; Conference date: 17-11-2022 Through 19-11-2022",

year = "2022",

doi = "10.1007/978-3-031-20992-5\_4",

language = "English",

isbn = "9783031209918",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "40--51",

editor = "Olarik Surinta and \{Kam Fung Yuen\}, Kevin",

booktitle = "Multi-disciplinary Trends in Artificial Intelligence - 15th International Conference, MIWAI 2022, Proceedings",

}

Intagorn, S, Pinitkan, S, Panmuang, M & Rodmorn, C 2022, Helmet Detection System for Motorcycle Riders with Explainable Artificial Intelligence Using Convolutional Neural Network and Grad-CAM. in O Surinta & K Kam Fung Yuen (eds), Multi-disciplinary Trends in Artificial Intelligence - 15th International Conference, MIWAI 2022, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 13651 LNAI, Springer Science and Business Media Deutschland GmbH, pp. 40-51, 15th Multi-disciplinary International Conference on Artificial Intelligence, MIWAI 2022, Virtual, Online, 17/11/22. https://doi.org/10.1007/978-3-031-20992-5_4

Helmet Detection System for Motorcycle Riders with Explainable Artificial Intelligence Using Convolutional Neural Network and Grad-CAM. / Intagorn, Suradej; Pinitkan, Suriya; Panmuang, Mathuros et al.
Multi-disciplinary Trends in Artificial Intelligence - 15th International Conference, MIWAI 2022, Proceedings. ed. / Olarik Surinta; Kevin Kam Fung Yuen. Springer Science and Business Media Deutschland GmbH, 2022. p. 40-51 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 13651 LNAI).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Helmet Detection System for Motorcycle Riders with Explainable Artificial Intelligence Using Convolutional Neural Network and Grad-CAM

AU - Intagorn, Suradej

AU - Pinitkan, Suriya

AU - Panmuang, Mathuros

AU - Rodmorn, Chonnikarn

PY - 2022

Y1 - 2022

N2 - Motorcycle accidents are one of the most common causes of injury and death in road users. This research has applied convolutional neural network (CNN) and explainable AI to detect motorcyclist without helmet and explain why CNN made that decision. The concept is based on deep learning and CNN principles applied to automatically detect motorcycle riders without helmet from images using three baseline classifiers: support vector machine, random forest and logistic regression to compare with deep convolutional neural network and measure accuracy with accuracy and F1-Score. The results revealed that CNN's F1-Score was 0.8326, which was the highest among all predictive models.

AB - Motorcycle accidents are one of the most common causes of injury and death in road users. This research has applied convolutional neural network (CNN) and explainable AI to detect motorcyclist without helmet and explain why CNN made that decision. The concept is based on deep learning and CNN principles applied to automatically detect motorcycle riders without helmet from images using three baseline classifiers: support vector machine, random forest and logistic regression to compare with deep convolutional neural network and measure accuracy with accuracy and F1-Score. The results revealed that CNN's F1-Score was 0.8326, which was the highest among all predictive models.

KW - Convolutional neural network

KW - Explainable AI

KW - Helmet

UR - https://www.scopus.com/pages/publications/85142673827

U2 - 10.1007/978-3-031-20992-5_4

DO - 10.1007/978-3-031-20992-5_4

M3 - Conference contribution

AN - SCOPUS:85142673827

SN - 9783031209918

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 40

EP - 51

BT - Multi-disciplinary Trends in Artificial Intelligence - 15th International Conference, MIWAI 2022, Proceedings

A2 - Surinta, Olarik

A2 - Kam Fung Yuen, Kevin

PB - Springer Science and Business Media Deutschland GmbH

T2 - 15th Multi-disciplinary International Conference on Artificial Intelligence, MIWAI 2022

Y2 - 17 November 2022 through 19 November 2022

ER -

Intagorn S, Pinitkan S, Panmuang M, Rodmorn C. Helmet Detection System for Motorcycle Riders with Explainable Artificial Intelligence Using Convolutional Neural Network and Grad-CAM. In Surinta O, Kam Fung Yuen K, editors, Multi-disciplinary Trends in Artificial Intelligence - 15th International Conference, MIWAI 2022, Proceedings. Springer Science and Business Media Deutschland GmbH. 2022. p. 40-51. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-031-20992-5_4

Helmet Detection System for Motorcycle Riders with Explainable Artificial Intelligence Using Convolutional Neural Network and Grad-CAM

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this