Cross-sensor nighttime lights image calibration for DMSP/OLS and SNPP/VIIRS with Residual U-net

By Dmitry Nechaev¹, Mikhail Zhizhin^2,3,*, Alexey Poyda^{1, 4}, Tilottama Ghosh², Feng-Chi Hsu² and Christopher Elvidge²

¹ Moscow Institute of Physics and Technology; [email protected]

² Earth Observation Group, Payne Institute, Colorado School of Mines; [email protected]

³ Space Research Institute, Russian Academy of Sciences

⁴ NRC “Kurchatov Institute”; [email protected]

^* Correspondence: [email protected]

Table of Contents

0. Introduction
1. Usage
2. Citation

Introduction

This repository contains all the data, scripts, research notebooks, plots, weights, etc. for the "Cross-sensor nighttime lights image calibration for DMSP/OLS and SNPP/VIIRS with Residual U-net" paper. Master branch has the script that can be used for the generation of DMSP-like imagery from VIIRS imagery. The notebooks branch has all the research related data.

Usage

This script currently supports .npz and .envi formats!

Check the dependencies in the requirements.txt, you need to install tensorflow-addons.

Run "python generation.py --help" to check how to use it.

In case of any questions contact me on [email protected]

Citation

If you use the data or model in your research, please cite:

Nechaev, D.; Zhizhin, M.; Poyda, A.; Ghosh, T.; Hsu, F.-C.; Elvidge, C. Cross-Sensor Nighttime Lights Image Calibration for DMSP/OLS and SNPP/VIIRS with Residual U-Net. Remote Sens. 2021, 13, 5026. https://doi.org/10.3390/rs13245026