torchgeo/yolo11s_marine_vessel_detection

Rehosted from https://hf.co/mayrajeo/marine-vessel-yolo using the following code:

import os
import hashlib

import ultralytics

url = "https://hf.co/mayrajeo/marine-vessel-yolo/resolve/main/yolo11s_tci.pt"
model = ultralytics.YOLO(url)
print(model.args)
"""
{
    'task': 'detect',
    'data': '/run/nvme/job_26457590/data/tci/fold_1.yaml',
    'imgsz': 640,
    'single_cls': False,
    'model': 'saved_model.pt',
}
"""

# save and compute hash
filename = "yolo11s_sentinel2_rgb_marine_vessel_detection.pt"
model.ckpt = model.state_dict()
model.save(filename)
md5 = hashlib.md5(open(filename, "rb").read()).hexdigest()[:8]
os.rename(filename, filename.replace(".pt", f"-{md5}.pt"))

If you use this model in your research please cite the following paper:

@article{mayraMappingRecreational2025,
title = {Mapping recreational marine traffic from Sentinel-2 imagery using YOLO object detection models},
journal = {Remote Sensing of Environment},
volume = {326},
pages = {114791},
year = {2025},
issn = {0034-4257},
doi = {https://doi.org/10.1016/j.rse.2025.114791},
url = {https://www.sciencedirect.com/science/article/pii/S0034425725001956},
author = {Janne Mäyrä and Elina A. Virtanen and Ari-Pekka Jokinen and Joni Koskikala and Sakari Väkevä and Jenni Attila},
keywords = {Marine vessel detection, Object detection, Satellite imagery, Deep learning, Human pressures},
}