A Convolutional Neural Network Based Early Warning System to Prevent Elephant-Train Collisions

Gunasekara, S; Jayasuriya, M; Harischandra, N; Samaranayake, L; Dissanayake, G

A Convolutional Neural Network Based Early Warning System to Prevent Elephant-Train Collisions

Gunasekara, S Jayasuriya, M Harischandra, N Samaranayake, L Dissanayake, G

Permalink

Publisher:: IEEE
Publication Type:: Conference Proceeding
Citation:: 2021 IEEE 16th International Conference on Industrial and Information Systems (ICIIS), 2022, 00, pp. 271-276
Issue Date:: 2022-01-03

Closed Access

	Filename	Description	Size
	A_Convolutional_Neural_Network_Based_Early_Warning_System_to_Prevent_Elephant-Train_Collisions.pdf		3.71 MB	Adobe PDF	View/Open

Copyright Clearance Process

Recently Added
In Progress
Closed Access

This item is closed access and not available.

Full metadata record

Field	Value	Language
dc.contributor.author	Gunasekara, S
dc.contributor.author	Jayasuriya, M
dc.contributor.author	Harischandra, N
dc.contributor.author	Samaranayake, L
dc.contributor.author	Dissanayake, G https://orcid.org/0000-0002-7992-0680
dc.date	2021-12-09
dc.date.accessioned	2023-03-10T04:13:33Z
dc.date.available	2023-03-10T04:13:33Z
dc.date.issued	2022-01-03
dc.identifier.citation	2021 IEEE 16th International Conference on Industrial and Information Systems (ICIIS), 2022, 00, pp. 271-276
dc.identifier.isbn	9781665426374
dc.identifier.uri	http://hdl.handle.net/10453/166964
dc.description.abstract	One serious facet of the worsening Human-Elephant Conflict (HEC) in nations such as Sri Lanka involves elephant-train collisions. Endangered Asian elephants are maimed or killed during such accidents, which also often results in orphaned or disabled elephants. Furthermore, railway services incur significant financial losses and disruptions to services annually due to such accidents. Most elephant-train collisions occur due to a lack of adequate reaction time due to poor driver visibility at sharp turns, night-time operation, and poor weather conditions. Initial investigations also indicate that most collisions occur in localised “hotspots” where elephant pathways/corridors intersect with railway tracks. Taking these factors into consideration, this work proposes the leveraging of recent developments in Convolutional Neural Network (CNN) technology to detect elephants using an RGB/infrared capable camera, around known hotspots along the railway track. The CNN was trained using a curated dataset of elephants collected on field visits to elephant sanctuaries and wildlife parks in Sri Lanka. With this vision-based detection system at its core, a prototype unit of an early warning system was designed and tested. Initial results indicate that detection accuracy is sufficient under varying lighting situations, provided that comprehensive training datasets that represent a wide range of challenging conditions are available. The overall hardware prototype was shown to be robust and reliable. We envision a network of such units may help contribute to reducing the problem of elephanttrain collisions and has the potential to act as an important surveillance mechanism in dealing with the broader issue of the human-elephant conflict.
dc.language	en
dc.publisher	IEEE
dc.relation.ispartof	2021 IEEE 16th International Conference on Industrial and Information Systems (ICIIS)
dc.relation.ispartof	2021 IEEE 16th International Conference on Industrial and Information Systems
dc.relation.isbasedon	10.1109/iciis53135.2021.9660651
dc.rights	info:eu-repo/semantics/closedAccess
dc.title	A Convolutional Neural Network Based Early Warning System to Prevent Elephant-Train Collisions
dc.type	Conference Proceeding
utslib.citation.volume	00
utslib.location.activity	Kandy, Sri Lanka
pubs.organisational-group	/University of Technology Sydney
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology/School of Mechanical and Mechatronic Engineering
utslib.copyright.status	closed_access	*
dc.date.updated	2023-03-10T04:13:31Z
pubs.finish-date	2021-12-11
pubs.publication-status	Published
pubs.start-date	2021-12-09
pubs.volume	00

Abstract:

One serious facet of the worsening Human-Elephant Conflict (HEC) in nations such as Sri Lanka involves elephant-train collisions. Endangered Asian elephants are maimed or killed during such accidents, which also often results in orphaned or disabled elephants. Furthermore, railway services incur significant financial losses and disruptions to services annually due to such accidents. Most elephant-train collisions occur due to a lack of adequate reaction time due to poor driver visibility at sharp turns, night-time operation, and poor weather conditions. Initial investigations also indicate that most collisions occur in localised “hotspots” where elephant pathways/corridors intersect with railway tracks. Taking these factors into consideration, this work proposes the leveraging of recent developments in Convolutional Neural Network (CNN) technology to detect elephants using an RGB/infrared capable camera, around known hotspots along the railway track. The CNN was trained using a curated dataset of elephants collected on field visits to elephant sanctuaries and wildlife parks in Sri Lanka. With this vision-based detection system at its core, a prototype unit of an early warning system was designed and tested. Initial results indicate that detection accuracy is sufficient under varying lighting situations, provided that comprehensive training datasets that represent a wide range of challenging conditions are available. The overall hardware prototype was shown to be robust and reliable. We envision a network of such units may help contribute to reducing the problem of elephanttrain collisions and has the potential to act as an important surveillance mechanism in dealing with the broader issue of the human-elephant conflict.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/166964