Intelligent preprocessing selection for pavement crack detection based on deep reinforcement learning

Yan Lin; Guosheng Xu; Siyi Li; Guoai Xu; Jie Deng; Jiankun Cao

doi:10.18293/SEKE2021-062

Intelligent preprocessing selection for pavement crack detection based on deep reinforcement learning

Yan Lin
, Guosheng Xu
, Siyi Li
, Guoai Xu
, Jie Deng
, Jiankun Cao

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

Abstract

With the rapid increase of traffic, the pressure on road maintenance is gradually increasing. Pavement crack is a common problem in all kinds of pavement diseases. In the actual production process, pavement images have different kinds of noise influence. The proposed algorithm is to select optimal preprocessing methods for pavement images in various conditions to improve the accuracy of crack detection. The algorithm includes two parts, a crack detection network and an intelligent preprocessing decision system. The crack detection network identifies the cracks in road images. The intelligent preprocessing decision system selects the best preprocessing method for pavement images based on the deep reinforcement method. The experiment results indicate that the validity and effectiveness of our proposed method.

Original language	English
Title of host publication	Proceedings - SEKE 2021
Subtitle of host publication	33rd International Conference on Software Engineering and Knowledge Engineering
Publisher	Knowledge Systems Institute Graduate School
Pages	458-463
Number of pages	6
ISBN (Electronic)	1891706527
DOIs	https://doi.org/10.18293/SEKE2021-062
State	Published - 2021
Externally published	Yes
Event	33rd International Conference on Software Engineering and Knowledge Engineering, SEKE 2021 - Pittsburgh, United States Duration: 1 Jul 2021 → 10 Jul 2021

Publication series

Name	Proceedings of the International Conference on Software Engineering and Knowledge Engineering, SEKE
Volume	2021-July
ISSN (Print)	2325-9000
ISSN (Electronic)	2325-9086

Conference

Conference	33rd International Conference on Software Engineering and Knowledge Engineering, SEKE 2021
Country/Territory	United States
City	Pittsburgh
Period	1/07/21 → 10/07/21

Keywords

Crack detection
Deep reinforcement learning
Intelligent preprocessing decision system

Access to Document

10.18293/SEKE2021-062

Cite this

Lin, Y., Xu, G., Li, S., Xu, G., Deng, J., & Cao, J. (2021). Intelligent preprocessing selection for pavement crack detection based on deep reinforcement learning. In Proceedings - SEKE 2021: 33rd International Conference on Software Engineering and Knowledge Engineering (pp. 458-463). (Proceedings of the International Conference on Software Engineering and Knowledge Engineering, SEKE; Vol. 2021-July). Knowledge Systems Institute Graduate School. https://doi.org/10.18293/SEKE2021-062

Lin, Yan ; Xu, Guosheng ; Li, Siyi et al. / Intelligent preprocessing selection for pavement crack detection based on deep reinforcement learning. Proceedings - SEKE 2021: 33rd International Conference on Software Engineering and Knowledge Engineering. Knowledge Systems Institute Graduate School, 2021. pp. 458-463 (Proceedings of the International Conference on Software Engineering and Knowledge Engineering, SEKE).

@inproceedings{e2acb33e05754c389559ead76f252c2e,

title = "Intelligent preprocessing selection for pavement crack detection based on deep reinforcement learning",

abstract = "With the rapid increase of traffic, the pressure on road maintenance is gradually increasing. Pavement crack is a common problem in all kinds of pavement diseases. In the actual production process, pavement images have different kinds of noise influence. The proposed algorithm is to select optimal preprocessing methods for pavement images in various conditions to improve the accuracy of crack detection. The algorithm includes two parts, a crack detection network and an intelligent preprocessing decision system. The crack detection network identifies the cracks in road images. The intelligent preprocessing decision system selects the best preprocessing method for pavement images based on the deep reinforcement method. The experiment results indicate that the validity and effectiveness of our proposed method.",

keywords = "Crack detection, Deep reinforcement learning, Intelligent preprocessing decision system",

author = "Yan Lin and Guosheng Xu and Siyi Li and Guoai Xu and Jie Deng and Jiankun Cao",

note = "Publisher Copyright: {\textcopyright} 2021 Knowledge Systems Institute Graduate School. All rights reserved.; 33rd International Conference on Software Engineering and Knowledge Engineering, SEKE 2021 ; Conference date: 01-07-2021 Through 10-07-2021",

year = "2021",

doi = "10.18293/SEKE2021-062",

language = "英语",

series = "Proceedings of the International Conference on Software Engineering and Knowledge Engineering, SEKE",

publisher = "Knowledge Systems Institute Graduate School",

pages = "458--463",

booktitle = "Proceedings - SEKE 2021",

address = "美国",

}

Lin, Y, Xu, G, Li, S, Xu, G, Deng, J & Cao, J 2021, Intelligent preprocessing selection for pavement crack detection based on deep reinforcement learning. in Proceedings - SEKE 2021: 33rd International Conference on Software Engineering and Knowledge Engineering. Proceedings of the International Conference on Software Engineering and Knowledge Engineering, SEKE, vol. 2021-July, Knowledge Systems Institute Graduate School, pp. 458-463, 33rd International Conference on Software Engineering and Knowledge Engineering, SEKE 2021, Pittsburgh, United States, 1/07/21. https://doi.org/10.18293/SEKE2021-062

Intelligent preprocessing selection for pavement crack detection based on deep reinforcement learning. / Lin, Yan; Xu, Guosheng; Li, Siyi et al.
Proceedings - SEKE 2021: 33rd International Conference on Software Engineering and Knowledge Engineering. Knowledge Systems Institute Graduate School, 2021. p. 458-463 (Proceedings of the International Conference on Software Engineering and Knowledge Engineering, SEKE; Vol. 2021-July).

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

TY - GEN

T1 - Intelligent preprocessing selection for pavement crack detection based on deep reinforcement learning

AU - Lin, Yan

AU - Xu, Guosheng

AU - Li, Siyi

AU - Xu, Guoai

AU - Deng, Jie

AU - Cao, Jiankun

PY - 2021

Y1 - 2021

N2 - With the rapid increase of traffic, the pressure on road maintenance is gradually increasing. Pavement crack is a common problem in all kinds of pavement diseases. In the actual production process, pavement images have different kinds of noise influence. The proposed algorithm is to select optimal preprocessing methods for pavement images in various conditions to improve the accuracy of crack detection. The algorithm includes two parts, a crack detection network and an intelligent preprocessing decision system. The crack detection network identifies the cracks in road images. The intelligent preprocessing decision system selects the best preprocessing method for pavement images based on the deep reinforcement method. The experiment results indicate that the validity and effectiveness of our proposed method.

AB - With the rapid increase of traffic, the pressure on road maintenance is gradually increasing. Pavement crack is a common problem in all kinds of pavement diseases. In the actual production process, pavement images have different kinds of noise influence. The proposed algorithm is to select optimal preprocessing methods for pavement images in various conditions to improve the accuracy of crack detection. The algorithm includes two parts, a crack detection network and an intelligent preprocessing decision system. The crack detection network identifies the cracks in road images. The intelligent preprocessing decision system selects the best preprocessing method for pavement images based on the deep reinforcement method. The experiment results indicate that the validity and effectiveness of our proposed method.

KW - Crack detection

KW - Deep reinforcement learning

KW - Intelligent preprocessing decision system

UR - https://www.scopus.com/pages/publications/85114270020

U2 - 10.18293/SEKE2021-062

DO - 10.18293/SEKE2021-062

M3 - 会议稿件

AN - SCOPUS:85114270020

T3 - Proceedings of the International Conference on Software Engineering and Knowledge Engineering, SEKE

SP - 458

EP - 463

BT - Proceedings - SEKE 2021

PB - Knowledge Systems Institute Graduate School

T2 - 33rd International Conference on Software Engineering and Knowledge Engineering, SEKE 2021

Y2 - 1 July 2021 through 10 July 2021

ER -

Lin Y, Xu G, Li S, Xu G, Deng J, Cao J. Intelligent preprocessing selection for pavement crack detection based on deep reinforcement learning. In Proceedings - SEKE 2021: 33rd International Conference on Software Engineering and Knowledge Engineering. Knowledge Systems Institute Graduate School. 2021. p. 458-463. (Proceedings of the International Conference on Software Engineering and Knowledge Engineering, SEKE). doi: 10.18293/SEKE2021-062

Intelligent preprocessing selection for pavement crack detection based on deep reinforcement learning

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this