IMIS

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).