Low-Cost Hyperspectral Imaging in Macroalgae Monitoring

Marc Allentoft-Larsen, Joaquim Santos, Christian Pedersen, Paul Michael Petersen, Hans Jakobsen

Research output: Chapter in Book/Report/Conference proceedingConference abstract in proceedingsResearchpeer-review

Abstract

In this study, we present an approach to macroalgae monitoring using an innovative, cost-effective hyperspectral camera system. Kelp beds, acknowledged for their ecological significance, provide essential fish habitats and contribute to nutrient cycling. With the increasing importance of responding to environmental changes, continuous monitoring has become essential, driven by European Union (EU)legislation. Hyperspectral imaging (HSI) is a powerful tool in this context due to its ability to detect pigment-characteristic fingerprints, but its high cost is a barrier to in situ monitoring. Our study showcases the development of an inexpensive HSI setup combining a GoPro camera with a rotating continuous variable spectral band pass filter, with cost-effective design and application. Experimental tests comprised a selection of two macro-algae species with overlapping spectral features and a controlled aquatic environment. Using a support vector machine (SVM) model for species discrimination, we were able to demonstrate promising discriminatory power of HSI over conventional RGB imaging. This work represents a leaping step towards achieving large-scale, automated ecological monitoring
Original languageEnglish
Title of host publication22. Danske Havforskermøde Abstract book
PublisherTechnical University of Denmark
Publication date2024
Pages8-8
Publication statusPublished - 2024
Event22. Danske Havforskermøde - DTU Aqua, Lyngby, Denmark
Duration: 23 Jan 202425 Jan 2024
Conference number: 22

Conference

Conference22. Danske Havforskermøde
Number22
LocationDTU Aqua
Country/TerritoryDenmark
CityLyngby
Period23/01/202425/01/2024

Fingerprint

Dive into the research topics of 'Low-Cost Hyperspectral Imaging in Macroalgae Monitoring'. Together they form a unique fingerprint.

Cite this