Researchers at the Universitat de Barcelona, BIOST, and GRBIO in Spain have published a paper showing how a deep learning web application, written in R, could address the problem of large-scale marine litter. Meet MARLIT.
"Automatic aerial photography techniques combined with analytical algorithms are more efficient protocols for the control and study of [macro-litter in marine environments]", says first author Odei García-Garin of the study's focus. "However, automated remote sensing of these materials is at an early stage."
"There are several factors in the ocean (waves, wind, clouds, etc.) that harden the detection of floating litter automatically with the aerial images of the marine surface. This is why there are only a few studies that made the effort to work on algorithms to apply to this new research context."
To address this the researchers used existing imagery from pollution monitoring campaigns, gathered from planes and drones, and worked on the development of an algorithm which could recognise and quantify macro-scale litter floating on the sea. The result: MARLIT.
MARLIT is a web app, written as an R Shiny package and provided as open access for "all managers and professionals in the study of the detection and quantification of floating marine macro-litter with aerial images", which allows for automated analysis of both whole images and sub-divided segments for the presence of floating litter and, if detected, to provide an estimation of its density.
"The great amount of images of the marine surface," García-Garin notes, "obtained by drones and planes in monitoring campaigns on marine litter — also in experimental studies with known floating objects — enabled us to develop and test a new algorithm that reaches a 80% of precision in the remote sensing of floating marine macro-litter."
The researchers have indicated that MARLIT could assist with adherence to the European Union's Marine Strategy Framework Directive, which requires continuous assessment of the environmental state of the marine environment.
The paper has been published under closed-access terms in the journal Environmental Pollution.