Measuring Airport Activity from Sentinel-2 Imagery to Support Decision-Making during COVID-19 Pandemic
This repository contains the implementation of our strategy to detect structural breaks caused by the COVID-19 pandemic, in time series of detected flying airplanes through remote sensing imagery, and the recovery rate of 30 busiest airports in countries with some integration to the European Union.
- Mauricio Pamplona Segundo (USF)
- Rodrigo Minetto (UTFPR)
- Allan Pinto (Unicamp)
- Cole Hill (USF)
- Ricardo da Silva Torres (NTNU)
- Sudeep Sarkar (USF)
- Clone this repository:
git clone --recurse-submodules https://github.com/allansp84/covid19-airports-activity-analysis.git
cd covid19-airports-activity-analysis
-
(Optional) Install the Anaconda or Miniconda environment by following the instructions available in the Anaconda's website or in the Miniconda's website.
-
(Optional) Create a virtual environment using Anaconda or Miniconda by running the commands below:
conda create -n airport-analysis-env python=3.7 -y
conda activate airport-analysis-env
- Install all dependencies required to run this software using the following command:
bash install_dependencies.sh
- To see the options available to run the software to analyze the Airports' activity, please run the following command:
python airports_activity_analysis.py -h
- To reproduce the results achieved by the SMA crossover method to detect the breaking point (caused by the COVID-19 pandemic) in the time series, please run the following command:
python airports_activity_analysis.py --time_series_path timeseries/ --breakout_algo sma-algo --output_path working
- To reproduce the results achieved by using the Twitter's algorithm to detect the breaking point (caused by the COVID-19 pandemic) in the time series, please run the following command:
python airports_activity_analysis.py --time_series_path timeseries/ --breakout_algo twitter-algo --output_path working
- The winner of the RACE Upscaling Competition launched by the European Space Agency (ESA) in coordination with the European Commission and managed by Euro Data Cube group (Final result).
If you find the code in this repository useful in your research, please consider citing:
TODO
TODO