Core-S1RTC

Contains a global coverage of Sentinel-1 (RTC) patches, each of size 1,068 x 1,068 pixels.

Content

¹ False colour composites are made with decibel-scale values with red green and blue defined as R:VV G:VV+VH B:VH. For each channel, a contrast-stretch is applied, transforming minimum-maximum to 0-255. This means bluer areas have relatively higher VH values, whilst brightness is a function of overall intensity. This is relative within each thumbnail because of the normalisation, and so cannot be compared across different samples.

Spatial Coverage

This is a global monotemporal dataset. Nearly every piece of Earth captured by Sentinel-1 is contained at least once in this dataset (and only once, excluding some marginal overlaps). The coverage is about 35% lower than for Core Sentinel-2 dataset due to the sensor coverage limitations.

The following figure demonstrates the spatial coverage (only black pixels are absent):

Example Use

Interface scripts are available at https://github.com/ESA-PhiLab/Major-TOM

Here's a sneak peek with a thumbnail image:

from fsspec.parquet import open_parquet_file
import pyarrow.parquet as pq
from io import BytesIO
from PIL import Image

PARQUET_FILE = 'part_03900' # parquet number
ROW_INDEX = 42 # row number (about 500 per parquet)

url = "https://huggingface.co/datasets/Major-TOM/Core-S1RTC/resolve/main/images/{}.parquet".format(PARQUET_FILE)
with open_parquet_file(url,columns = ["thumbnail"]) as f:
    with pq.ParquetFile(f) as pf:
        first_row_group = pf.read_row_group(ROW_INDEX, columns=['thumbnail'])

stream = BytesIO(first_row_group['thumbnail'][0].as_py())
image = Image.open(stream)

Cite

@inproceedings{Major_TOM,
  title={Major TOM: Expandable Datasets for Earth Observation}, 
  author={Alistair Francis and Mikolaj Czerkawski},
  year={2024},
  booktitle={IGARSS 2024 - 2024 IEEE International Geoscience and Remote Sensing Symposium}, 
  eprint={2402.12095},
  archivePrefix={arXiv},
  primaryClass={cs.CV}
}

Powered by Φ-lab, European Space Agency (ESA) 🛰️