VPInt

This repository contains the code required to run VPint (value propagation-based spatial interpolation), associated with our DAMI publication VPint: value propagation-based spatial interpolation (https://doi.org/10.1007/s10618-022-00843-2), and VPint2, associated with our ISPRS Journal of Photogrammetry and Remote Sensing paper (https://doi.org/10.1016/j.isprsjprs.2024.07.030).

Documentation

The code contains docstrings with detailed explanations. We hope to move this documentation to ReadTheDocs when we can.

Installation

To install VPint (choose between requirements_slim.txt and requirements.txt based on desired level of optional functionalities):

conda create --name VPint python=3.12.3
conda activate VPint
git clone git@github.com:LaurensArp/VPint.git
cd VPint
pip install -r requirements.txt
python setup.py install

Running VPint2 for cloud removal in Earth Observation data

A minimal example to run VPint2 cloud removal, assuming appropriate data is already loaded:

from VPint.VPint2 import VPint2_interpolator

VPint2 = VPint2_interpolator(target, features, mask=mask)
target_clean = VPint2.run()

For more details, please see the cloud_removal_example.ipynb notebook in the examples folder, and the accompanying blog post/tutorial here: https://adaresearch.wordpress.com/2024/08/07/removing-clouds-from-optical-earth-observation-imagery-using-vpint2-in-2-lines-of-python-code/.

Further questions

If you have further questions, please feel free to reach out to me at my work email address specified in the paper.

Citing this work

To read our 2024 ISPRS Journal of Photogrammetry and Remote Sensing paper on VPint2 for cloud removal, please see: https://doi.org/10.1016/j.isprsjprs.2024.07.030 When publishing work using VPint2 cloud removal, please cite:

@article{ArpEtAl24,
    author = {Laurens Arp and Holger Hoos and Peter {van Bodegom} and Alistair Francis and James Wheeler and Dean {van Laar} and Mitra Baratchi},
    title = {Training-free thick cloud removal for Sentinel-2 imagery using value propagation interpolation},
    journal = {ISPRS Journal of Photogrammetry and Remote Sensing},
    volume = {216},
    pages = {168-184},
    year = {2024},
    issn = {0924-2716},
    doi = "https://doi.org/10.1016/j.isprsjprs.2024.07.030",
    url = "https://www.sciencedirect.com/science/article/pii/S0924271624002995",
    tags = {Remote sensing, spatial interpolation, cloud removal},
}

To read our 2022 DAMI paper on VPint, please see: https://doi.org/10.1007/s10618-022-00843-2 When publishing work using VPint, please cite:

@article{ArpEtAl22,
    author = "Arp, Laurens and Baratchi, Mitra and Hoos, Holger",
    title = "VPint: value propagation-based spatial interpolation",
    journal = "Data Mining and Knowledge Discovery",
    volume = "36",
    pages = "",
    publisher = "Springer",
    year = "2022",
    issn = "1573-756X",
    doi = "https://doi.org/10.1007/s10618-022-00843-2",
    url = "https://link.springer.com/article/10.1007/s10618-022-00843-2",
}