0
Article ? AI-assigned paper type based on the abstract. Classification may not be perfect — flag errors using the feedback button. Tier 2 ? Original research — experimental, observational, or case-control study. Direct primary evidence. Detection Methods Sign in to save

Microplastics_SEM

Open MIND 2026
Bin Shi, Bin Shi, Jane Howe, Jane Howe

Summary

This is a research dataset and code repository — not a standalone paper — associated with a study that trained deep learning models to automatically detect and classify microplastics in scanning electron microscope images, supporting more efficient and reproducible analysis.

Here are the datasets for the paper (Automatic quantification and classification of microplastics in scanning electron micrographs via deep learning) published the Science of the Total Environment. The complete training dataset of semantic segmentation and shape classification is in dataset1.zip. The dataset of instance segmentation is present in multiple forms in dataset2.zip. The experiment setting for each micrograph was recorded together with the original SEM micrographs and all ground-truth labels in the file dataset3.zip. Main results were saved in result.zip. Paper link: https://doi.org/10.1016/j.scitotenv.2022.153903 Code link: https://github.com/benjamin3344/Microplastics_Unet Updates on 2026-02-22: Pre-trained weights for models of different sizes (4*5 fold) are included in result_weights.zip. Any further cooperation can be addressed to Professor Jane Howe.

Share this paper