Satellite Image Time Series Analysis under Time Warping

Here is a simple toolbox that allows you to run a Kmeans clustering on a series of images. This is an unoptimized Java version of the algorithm. I hope there is no bug in it

How to

To run the jar file, please execute: java -jar -Xmx2g DTW-SITS.jar

You will then be asked to select your set of images. They constitute your series of images to analyse. They have to be named so that a lexicographical sort on their names gives them in the right order in the series. For example:

You will then be asked in how many clusters you want to group your sequences, as well as the number of iterations for the Kmeans clustering.

This version is the represents a simple case:

  • Only png images
  • No cloud or global masks
  • No constraints on the maximum time shift in the warping

For more functionalities, please contact me.

Please cite as:

  • F. Petitjean, J. Inglada & P. Gançarski
    Satellite Image Time Series Analysis under Time Warping
    IEEE Transactions on Geoscience and Remote Sensing, 2012, Vol. 50, Num. 8, pp. 3081-3095.
    doi: 10.1109/TGRS.2011.2179050

The package is licensed under a GPLv3 (license in the Jar file). For any commercial license, feel free to contact me.