Multi-sensor infrared imaging for floating waste monitoring

Jukka Sassi; Sami Siikanen; Thomasine Kamerling; Anssi Mikola; Saurabh Gupta

doi:10.1117/12.2616564

Multi-sensor infrared imaging for floating waste monitoring

Jukka Sassi^*
, Sami Siikanen
, Thomasine Kamerling
, Anssi Mikola
, Saurabh Gupta

^*Corresponding author for this work

BA6301 Optical measurements

Huhtamäki Oyj
Riverrecycle Oy
Earth5R

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

Abstract

Multi-sensor imaging option consisting of infrared (TIR), near infrared hyper-spectral (NIR HS) and red-green-blue (RGB) sensors was demonstrated to confirm the ability of the sensors to detect the floating debris and differentiate plastic from organic material in a realistic river environment. A drone and a fixed installation were utilised. The target objects were typical plastic products used in the households and the organic material consisted of pieces of wood and branches. The results suggest that multi-imaging with NIR HS, TIR and RGB sensors is a promising method for separating floating plastic waste from organic material. Further efforts will be targeted in possibility to distinguish different plastic types from each other and how this process could by applied by utilising machine learning methods.

Original language	English
Title of host publication	Thermosense: Thermal Infrared Applications XLIV conference
Editors	Arantza Mendioroz, Nicolas P. Avdelidis
Publisher	International Society for Optics and Photonics SPIE
Number of pages	7
ISBN (Electronic)	9781510650947
DOIs	https://doi.org/10.1117/12.2616564
Publication status	Published - 27 May 2022
MoE publication type	Not Eligible
Event	SPIE Defense + Commercial Sensing 2022 - Orlando, United States Duration: 27 Jun 2022 → 27 Jun 2022

Publication series

Series	Proceedings of SPIE
Volume	12109
ISSN	0277-786X

Conference

Conference	SPIE Defense + Commercial Sensing 2022
Country/Territory	United States
City	Orlando
Period	27/06/22 → 27/06/22

Funding

The authors would like to express their gratitude to Huhtamäki Oyj, sustainable food packaging company, who sponsored the project. Also, the project partners Riverrecycle Oy and Earth5R provided insights to the challenging task of floating waste monitoring. This work is also part of the Academy of Finland Flagship Programme, Photonics Research and Innovation (PREIN), decision 320168.

UN SDGs

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

SDG 11 Sustainable Cities and Communities

Keywords

Drone
hyperspectral imaging
multi-sensor
NIR
plastic waste
RGB
TIR

Access to Document

10.1117/12.2616564

Cite this

@inproceedings{f84f9e4250a043159810e613b57933a6,

title = "Multi-sensor infrared imaging for floating waste monitoring",

abstract = "Multi-sensor imaging option consisting of infrared (TIR), near infrared hyper-spectral (NIR HS) and red-green-blue (RGB) sensors was demonstrated to confirm the ability of the sensors to detect the floating debris and differentiate plastic from organic material in a realistic river environment. A drone and a fixed installation were utilised. The target objects were typical plastic products used in the households and the organic material consisted of pieces of wood and branches. The results suggest that multi-imaging with NIR HS, TIR and RGB sensors is a promising method for separating floating plastic waste from organic material. Further efforts will be targeted in possibility to distinguish different plastic types from each other and how this process could by applied by utilising machine learning methods.",

keywords = "Drone, hyperspectral imaging, multi-sensor, NIR, plastic waste, RGB, TIR",

author = "Jukka Sassi and Sami Siikanen and Thomasine Kamerling and Anssi Mikola and Saurabh Gupta",

note = "Publisher Copyright: {\textcopyright} COPYRIGHT SPIE. Downloading of the abstract is permitted for personal use only.; SPIE Defense + Commercial Sensing 2022 ; Conference date: 27-06-2022 Through 27-06-2022",

year = "2022",

month = may,

day = "27",

doi = "10.1117/12.2616564",

language = "English",

series = "Proceedings of SPIE",

publisher = "International Society for Optics and Photonics SPIE",

editor = "Arantza Mendioroz and Avdelidis, \{Nicolas P.\}",

booktitle = "Thermosense: Thermal Infrared Applications XLIV conference",

address = "United States",

}

Sassi, J , Siikanen, S, Kamerling, T, Mikola, A & Gupta, S 2022, Multi-sensor infrared imaging for floating waste monitoring. in A Mendioroz & NP Avdelidis (eds), Thermosense: Thermal Infrared Applications XLIV conference., 1210909, International Society for Optics and Photonics SPIE, Proceedings of SPIE, vol. 12109, SPIE Defense + Commercial Sensing 2022, Orlando, Florida, United States, 27/06/22. https://doi.org/10.1117/12.2616564

Multi-sensor infrared imaging for floating waste monitoring. / Sassi, Jukka ; Siikanen, Sami; Kamerling, Thomasine et al.
Thermosense: Thermal Infrared Applications XLIV conference. ed. / Arantza Mendioroz; Nicolas P. Avdelidis. International Society for Optics and Photonics SPIE, 2022. 1210909 (Proceedings of SPIE, Vol. 12109).

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

TY - GEN

T1 - Multi-sensor infrared imaging for floating waste monitoring

AU - Sassi, Jukka

AU - Siikanen, Sami

AU - Kamerling, Thomasine

AU - Mikola, Anssi

AU - Gupta, Saurabh

PY - 2022/5/27

Y1 - 2022/5/27

N2 - Multi-sensor imaging option consisting of infrared (TIR), near infrared hyper-spectral (NIR HS) and red-green-blue (RGB) sensors was demonstrated to confirm the ability of the sensors to detect the floating debris and differentiate plastic from organic material in a realistic river environment. A drone and a fixed installation were utilised. The target objects were typical plastic products used in the households and the organic material consisted of pieces of wood and branches. The results suggest that multi-imaging with NIR HS, TIR and RGB sensors is a promising method for separating floating plastic waste from organic material. Further efforts will be targeted in possibility to distinguish different plastic types from each other and how this process could by applied by utilising machine learning methods.

AB - Multi-sensor imaging option consisting of infrared (TIR), near infrared hyper-spectral (NIR HS) and red-green-blue (RGB) sensors was demonstrated to confirm the ability of the sensors to detect the floating debris and differentiate plastic from organic material in a realistic river environment. A drone and a fixed installation were utilised. The target objects were typical plastic products used in the households and the organic material consisted of pieces of wood and branches. The results suggest that multi-imaging with NIR HS, TIR and RGB sensors is a promising method for separating floating plastic waste from organic material. Further efforts will be targeted in possibility to distinguish different plastic types from each other and how this process could by applied by utilising machine learning methods.

KW - Drone

KW - hyperspectral imaging

KW - multi-sensor

KW - NIR

KW - plastic waste

KW - RGB

KW - TIR

UR - https://www.scopus.com/pages/publications/85135860882

U2 - 10.1117/12.2616564

DO - 10.1117/12.2616564

M3 - Conference article in proceedings

T3 - Proceedings of SPIE

BT - Thermosense: Thermal Infrared Applications XLIV conference

A2 - Mendioroz, Arantza

A2 - Avdelidis, Nicolas P.

PB - International Society for Optics and Photonics SPIE

T2 - SPIE Defense + Commercial Sensing 2022

Y2 - 27 June 2022 through 27 June 2022

ER -

Multi-sensor infrared imaging for floating waste monitoring

Abstract

Publication series

Conference

Funding

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this