Model Card for UNet-6depth-shuffle: venkatesh-thiru/s2l8h-UNet-6depth-shuffle

Model Description

The UNet-6depth-shuffle model harmonizes Landsat-8 and Sentinel-2 imagery by improving the spatial resolution of Landsat-8 images. This model uses Landsat-8 multispectral and pan-chromatic images to produce outputs that match the Sentinel-2's spectral and spatial characteristics.

Model Architecture

This UNet model features 6 depth levels and incorporates a shuffling mechanism to enhance image resolution and spectral accuracy. The depth and shuffling operations are tailored to achieve high-quality transformations, ensuring the output images closely resemble Sentinel-2 data.

Usage

from transformers import AutoModel

# Load the UNet-6depth-shuffle model
model = AutoModel.from_pretrained("venkatesh-thiru/s2l8h-UNet-6depth-shuffle", trust_remote_code=True)

# Harmonize Landsat-8 images
l8up = model(l8MS, l8pan)

Where:

l8MS - Landsat Multispectral images (L2 Reflectances)

l8pan - Landsat Pan-Chromatic images (L1 Reflectances)

Applications

Water quality assessment

Urban planning

Climate monitoring

Disaster response

Infrastructure oversight

Agricultural surveillance

Limitations

Minor limitations may arise in regions with different spectral properties or under extreme environmental conditions.

Reference

For more details, refer to the publication: 10.1016/j.isprsjprs.2024.04.026

Downloads last month
22
Inference Examples
Inference API (serverless) does not yet support model repos that contain custom code.