Data and code for: Automated Shadowgraph Imaging and Deep Learning for Real-Time Detection of Planktonic Salmon Lice

This repository contains the complete training dataset, model weights, processing pipeline, and representative field data supporting the manuscript: Automated Shadowgraph Imaging and Deep Learning for Rare-Target Detection and High-Throughput Plankton Monitoring by Mats Huserbråten (Institute of Marine Research, Bergen, Norway) For further details see README.md and metadata.csv for file-level metadata for the entire deposit ### Licenses- All image data, ROIs, model weights and metadata: **CC-BY 4.0** - All source code and scripts: **MIT License** ### CitationPlease cite this record when reusing any part of the data or code: Huserbråten, M. (2026). Data and code for: Automated Shadowgraph Imaging and Deep Learning for Rare-Target Detection and High-Throughput Plankton Monitoring [Data set]. Zenodo. https://doi.org/10.5281/zenodo.19466046 Keywords: salmon lice, Lepeophtheirus salmonis, copepodid, shadowgraph imaging, deep learning, ConvNeXt, plankton monitoring, aquaculture, in-situ imaging, AI