Through thick and thin : imaging through obscurant using SPAD array

Joyce Mau; Vladimyros Devrelis; Geoffrey Day; Graeme Nash; Jochen Trumpf; Dennis Delic

doi:10.1109/SENSORS47125.2020.9278706

Through thick and thin : imaging through obscurant using SPAD array

Joyce Mau, Vladimyros Devrelis, Geoffrey Day, Graeme Nash, Jochen Trumpf, Dennis Delic

Western Sydney University

Research output: Chapter in Book / Conference Paper › Conference Paper › peer-review

5 Citations (Scopus)

Abstract

![CDATA[Preliminary work on 3D image collection and classification of targets in the presence of obscurant using a Flash LiDAR system is discussed in this paper. The system is based around a DST designed 32 x 32 Single Photon Avalanche Diode (SPAD) array to image either targets or silhouettes of targets. The collected data included military targets that were obscured either by camouflage nets or fog. For camouflage net, the target was detected using an algorithm implemented on the Nvidia Jetson TX2. Targets obscured by fog are detected and classified where the classification accuracy is 100% for fog visibility down to 17.3m and 89.5% for 14.1m. This algorithm was not implemented on the TX2 but its simplicity shows potential for it in the future. This initial approach opens the road to eventually operate SPAD based systems for real-time classification through dust or smoke.]]

Original language	English
Title of host publication	Proceedings of 2020 IEEE Sensors Conference (SENSORS 2020), Rotterdam, Netherlands, 25-28 October 2020
Publisher	IEEE
Number of pages	4
ISBN (Print)	9781728168012
DOIs	https://doi.org/10.1109/SENSORS47125.2020.9278706
Publication status	Published - 2020
Event	IEEE Sensors Conference - Duration: 25 Oct 2020 → …

Publication series

Name
ISSN (Print)	2168-9229

Conference

Conference	IEEE Sensors Conference
Period	25/10/20 → …

Access to Document

10.1109/SENSORS47125.2020.9278706

Cite this

@inproceedings{a263d4f719b14325a0e82e903dba9935,

title = "Through thick and thin : imaging through obscurant using SPAD array",

abstract = "![CDATA[Preliminary work on 3D image collection and classification of targets in the presence of obscurant using a Flash LiDAR system is discussed in this paper. The system is based around a DST designed 32 x 32 Single Photon Avalanche Diode (SPAD) array to image either targets or silhouettes of targets. The collected data included military targets that were obscured either by camouflage nets or fog. For camouflage net, the target was detected using an algorithm implemented on the Nvidia Jetson TX2. Targets obscured by fog are detected and classified where the classification accuracy is 100% for fog visibility down to 17.3m and 89.5% for 14.1m. This algorithm was not implemented on the TX2 but its simplicity shows potential for it in the future. This initial approach opens the road to eventually operate SPAD based systems for real-time classification through dust or smoke.]]",

author = "Joyce Mau and Vladimyros Devrelis and Geoffrey Day and Graeme Nash and Jochen Trumpf and Dennis Delic",

year = "2020",

doi = "10.1109/SENSORS47125.2020.9278706",

language = "English",

isbn = "9781728168012",

publisher = "IEEE",

booktitle = "Proceedings of 2020 IEEE Sensors Conference (SENSORS 2020), Rotterdam, Netherlands, 25-28 October 2020",

note = "IEEE Sensors Conference ; Conference date: 25-10-2020",

}

TY - GEN

T1 - Through thick and thin : imaging through obscurant using SPAD array

AU - Mau, Joyce

AU - Devrelis, Vladimyros

AU - Day, Geoffrey

AU - Nash, Graeme

AU - Trumpf, Jochen

AU - Delic, Dennis

PY - 2020

Y1 - 2020

N2 - ![CDATA[Preliminary work on 3D image collection and classification of targets in the presence of obscurant using a Flash LiDAR system is discussed in this paper. The system is based around a DST designed 32 x 32 Single Photon Avalanche Diode (SPAD) array to image either targets or silhouettes of targets. The collected data included military targets that were obscured either by camouflage nets or fog. For camouflage net, the target was detected using an algorithm implemented on the Nvidia Jetson TX2. Targets obscured by fog are detected and classified where the classification accuracy is 100% for fog visibility down to 17.3m and 89.5% for 14.1m. This algorithm was not implemented on the TX2 but its simplicity shows potential for it in the future. This initial approach opens the road to eventually operate SPAD based systems for real-time classification through dust or smoke.]]

AB - ![CDATA[Preliminary work on 3D image collection and classification of targets in the presence of obscurant using a Flash LiDAR system is discussed in this paper. The system is based around a DST designed 32 x 32 Single Photon Avalanche Diode (SPAD) array to image either targets or silhouettes of targets. The collected data included military targets that were obscured either by camouflage nets or fog. For camouflage net, the target was detected using an algorithm implemented on the Nvidia Jetson TX2. Targets obscured by fog are detected and classified where the classification accuracy is 100% for fog visibility down to 17.3m and 89.5% for 14.1m. This algorithm was not implemented on the TX2 but its simplicity shows potential for it in the future. This initial approach opens the road to eventually operate SPAD based systems for real-time classification through dust or smoke.]]

UR - https://hdl.handle.net/1959.7/uws:67441

U2 - 10.1109/SENSORS47125.2020.9278706

DO - 10.1109/SENSORS47125.2020.9278706

M3 - Conference Paper

SN - 9781728168012

BT - Proceedings of 2020 IEEE Sensors Conference (SENSORS 2020), Rotterdam, Netherlands, 25-28 October 2020

PB - IEEE

T2 - IEEE Sensors Conference

Y2 - 25 October 2020

ER -

Through thick and thin : imaging through obscurant using SPAD array

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this