Publicaties | Instituten | Personen | Datasets | Projecten | Kaarten
[ meld een fout in dit record ] Print deze pagina

RGB datasets for machine learning-based microplastic analysis - update
Citeerbaar als data publicatie
Meyers, N.; De Witte, B.; Janssen, C.; Everaert, G.; Flanders Marine Institute (VLIZ); Flanders Research Institute for Agriculture, Fisheries and Food (ILVO); Ghent University Laboratory for Environmental Toxicology (GhEnToxLab): Belgium; (2024): RGB datasets for machine learning-based microplastic analysis - update. Marine Data Archive.
Contact: Meyers, Nelle

Beschikbaarheid: Creative Commons License Deze dataset valt onder een Creative Commons Naamsvermelding 4.0 Internationaal-licentie.

This dataset features RGB statistics extracted from Nile red-stained particles, serving to train a 'Plastic Detection Model' and a 'Polymer Identification Model' using supervised machine learning techniques. By accurately discerning between plastic and natural particles, and distinguishing between different plastic polymers, the models facilitate efficient microplastic detection and identification. meer

Datasets containing RGB statistics extracted from photographed fluorescent reference particles stained with Nile red. The most abundantly produced plastic polymers worldwide as well as natural materials with high prevalence in the marine environment were considered for these datasets. The spectral data was used to construct two supervised machine learning models, i.e. a ‘Plastic Detection Model’ (PDM) and a ‘Polymer Identification Model’ (PIM), based on random forest algorithms. The PDM allows to accurately distinguish plastic from natural particles, while the PIM allows to distinguish different plastic polymers, in a cost- and time-efficient way. The datasets contain Red, Green and Blue (RGB) statistics extracted from Nile red-stained reference particles (50-1200 μm) photographed under three different microscope filters (blue, green and UV). Four different datasets can be found, two for each model (PDM vs. PIM), based on photographs acquired with two different types of microscope (Leica DM 1000 fluorescence microscope vs. Leica M205 FA fluorescence stereomicroscope). The datasets represent an updated version of earlier published RGB datasets, now containing 135 – 200 particles per polymer category (PIM), and 420-500 per particle type (plastic/organic).

Milieu-kwaliteit / vervuiling
Marien/Kust, Automated detection, Fluorescent colouration, Machine learning, Microplastics, Nile red staining, Random forest models, RGB colour data, World

Geografische spreiding

RGB (Red, Green, Blue) colour component means and percentiles Methode
RGB (Red, Green, Blue) colour component means and percentiles: Fluorescence microscopy combined with image analysis.

Bijdrage door
Vlaams Instituut voor de Zee (VLIZ), meerdata creator
Universiteit Gent; Faculteit Bio-ingenieurswetenschappen; Vakgroep Dierwetenschappen en Aquatische Ecologie; Laboratorium voor Milieutoxicologie (GhEnToxLab), meerdata creatordata creator
Vlaamse overheid; Beleidsdomein Landbouw en Visserij; Instituut voor landbouw-, visserij en voedingsonderzoek (ILVO), meerdata creator

Gerelateerde datasets
Dataset delen:
RGB-statistics derived from Nile red-stained reference plastics for the construction of the PDM (Plastics Detection Model), meer
RGB-statistics derived from Nile red-stained reference plastics for the construction of the PIM (Polymer Identification Model), meer

ANDROMEDA: Analysis techniques for quantifying nano-and microplastic particles and their degradation in the marine environment, meer

Gebaseerd op deze dataset
De Witte, B. et al. (2024). ANDROMEDA portfolio of microplastics analyses protocols. ANDROMEDA Deliverable 5.5. JPI Oceans ANDROMEDA project: [s.l.]. 88 pp., meer

Dataset status: Afgelopen
Data type: Data
Data oorsprong: Onderzoek: labo-experiment
Metadatarecord aangemaakt: 2024-04-05
Informatie laatst gewijzigd: 2024-04-05
Alle informatie in het Integrated Marine Information System (IMIS) valt onder het VLIZ Privacy beleid