/Graph_Learning_Based_on_Signal_Smoothness_Representation_for_Change_Detection

Graph signal processing for change detection

Primary LanguageMATLAB

Graph Learning Based on Signal Smoothness Representation for Homogeneous and Heterogeneous Change Detection

This repository is a proposed approach based on Graph Signal processing for change detection under the folllowing name:

Datasets

------------Available datasets and Parameters------------

Dataset Vx = Vy K Alpha
Mulargia_dataset 11 1464 0.1
Omodeo_dataset 20 174 0.109
Alaska_dataset 11 1479 0.013
Madeirinha_dataset 10 410 0.2
Canada_dataset 44 2400 0.1
Gloucester_1_dataset 144 171 0.1
Katios_dataset 23 102 0.738
Atlantico_dataset 17 2029 0.103
SF_dataset 16 260 0.215
Wenchuan_dataset 8 1740 0.5
Toulouse_dataset 86 192 0.002
California_dataset 73 291 0.1
Bastrop_dataset 23 393 0.1
Gloucester_2_dataset 150 444 0.042

These datasets are part of the following works:

Please if you use the datasets and the codes cite our works as:

@article{   JimenezSierra2022graph,
   author={Jimenez-Sierra, David Alejandro and Quintero-Olaya, David Alfredo and Alvear-Mu{\~n}oz, Juan Carlos and Ben{\'i}tez-Restrepo, Hern{\'a}n Dar{\'i}o and Florez-Ospina, Juan Felipe and Chanussot, Jocelyn},
   journal={IEEE Transactions on Geoscience and Remote Sensing},
   title={Graph Learning Based on Signal Smoothness Representation for Homogeneous and Heterogeneous Change Detection},
   year={2022},
   volume={60},
   number={},
   pages={1-16},
   doi={10.1109/TGRS.2022.3168126}
}

@article{   JimenezSierra2020graph,
   title={Graph-Based Data Fusion Applied to: Change Detection and Biomass Estimation in Rice Crops},
   author={Jimenez-Sierra, David Alejandro and Ben{\'i}tez-Restrepo, Hern{\'a}n Dar{\'i}o and Vargas-Cardona, Hern{\'a}n Dar{\'i}o and Chanussot, Jocelyn},
   journal={Remote Sensing},
   volume={12},
   number={17},
   pages={2683},
   year={2020},
   publisher={Multidisciplinary Digital Publishing Institute}
}

Requirements

In order to run the code, you will need to download and add to the path the following toolboxes:

  • GMMSP toolbox for superpixel segmentation
  • Graph signal processing toolbox for graph smoothness prior.
  • Unlocbox toolbox for graph learning optimization.

Canada and California datasets

Please read the data readme file to download these datasets.

PYTHON VERSION AVAILABLE

Now you can have access to the Python version of this code in the Juan Florez-Ospina Github