Deriving High Spatiotemporal Remote Sensing Images Using Deep Convolutional Network

Attention

Please consider using the enhanced DCSTFN (EDCSTFN) model, which shows better performance concerning the output image quality and fusion accuracy.

Code repository: https://github.com/theonegis/edcstfn

Paper online: https://www.mdpi.com/2072-4292/11/24/2898/htm

Requirements

Tested on the following environment:

Python: >=3.5

TensorFlow: 1.4.1

Keras: 2.1.2

The TensorFlow Version is not optimized for memory consumption and PyTorch Version is recommended.

Please checkout to the pytorch branch to use the PyTorch Version

If you find any bugs, pulling request is welcome!

PyTorch

The pytorch branch provides the pytorch implementation

Tested on the following environment:

• Python: >=3.6

• PyTorch: >=0.4

The PyTorch Version is recommended for produciton

Reference

http://www.mdpi.com/2072-4292/10/7/1066