Comet Time Series (CometTS) Visualizer

• Installation Instructions
• Dependencies
• License

---

Base Functionality

Comet Time Series (CometTS) is an open source tool coded in python including jupyter notebooks and command line utility that enables users to visualize or extract relevant statistics from almost any format time series of overhead imagery within a specific region of interest (ROI). To use CometTS, you must define your ROI, provide a CSV file documenting how your imagery is organized, and then run one of the CometTS analysis tools. This usually takes the following steps

1. Outline and download your ROI with a service like geojson.io
2. Organize your imagery and document it with the CometTS.CSV_It tool
3. Analyze your data using:
   • CometTS for trend analysis (optionally, mask unwanted clouds and other features with --maskit option)
   • or CometTS.ARIMA for averaging and anomaly detection
4. Plot the results using the plotting notebook

A full walkthrough of this functionality with example data is included in two notebooks: CSV_Creator and CometTS_Visualizer

File Formats:

Supported Raster Formats

Supported Vector Formats

Installation

Python 2.7 or 3.6 are the base requirements plus several packages. CometTS can be installed in multiple ways including conda, pip, docker, and cloning this repository.

Clone it

We recommend cloning to add all sample data and easier access to the jupyter notebooks that leverage our plotting functions.

git clone https://github.com/CosmiQ/CometTS.git

If you would like the full functionality of a python package we have several options.

pip

pip install CometTS

pip installs may fail on macs with python3 as GDAL is finicky. Use some of the alternative approaches below.

Docker

docker pull jss5102/cometts
docker run -it -v /nfs:/nfs --name cometts jss5102/cometts /bin/bash 

Conda

Create a conda environment!

git clone https://github.com/CosmiQ/CometTS.git
cd CometTS
conda env create -f environment.yml
source activate CometTS
pip install CometTS

Dependencies

All dependencies can be found in the docker file Dockerfile or environment.yml or requirements.txt.

Examples

Agadez, Niger

Seasonal variation in brightness that likely indicates seasonal migrations and population fluctuations in central Niger, Africa.

Suruc Refugee Camp, Turkey

Increase in brightness coinciding with the establishment of a refugee camp in southern Turkey, north of Syria.

Allepo, Syria

Brightness declines (i.e., putative population decline) as a result of Syrian Civil War and military actions in Aleppo.

NDVI Visualization north of Houston, Texas

A visualization of the Normalized Difference Vegetation Index (NDVI) in a field north of Houston using a time-series of Landsat imagery.

Landsat Multispectral Visualization

A visualization of three Landsat bands in New Orleans, Louisiana. Note the effects of Katrina in 2005.

Contribute or debug?

Interested in proposing a change, fixing a bug, or asking for help? Check out the contributions guidance.

License

See LICENSE.

Traffic