It is a great pleasure to announce that our Transaction on Geoscience and Remote Sensing (TGRS) paper describing our novel ECSPLAIN framework is now available online!


The "Explainability Constrained-claSsifier for Pairing the detection and the Localization of moving Areas from SAR INterferograms" (ECSPLAIN) framework constrains the Class Activation Map (CAM) of a classifier network to force its explainability. It is applied to the detection and the segmentation of slow moving areas (landslides and rock glaciers) on the interferograms computed from Sentinel-1 images belonging to the ISSLIDE dataset [1] (available on IEEE Dataport).


You can access the TGRS paper directly on IEEE Xplore.


If our work can be of any use for use, please consider citing it as:


A. Bralet, A. M. Atto, J. Chanussot and E. Trouvé, “ECSPLAIN: Explainability Constrained-claSsifier for Pairing the detection and the Localization of moving Areas from SAR INterferograms,” in IEEE Transactions on Geoscience and Remote Sensing, doi: 10.1109/TGRS.2025.3595267.



Reference:

[1] A. Bralet, E. Trouvé, J. Chanussot and A. M. Atto, “ISSLIDE: A New InSAR Dataset for Slow SLIding Area DEtection With Machine Learning,” in IEEE Geoscience and Remote Sensing Letters, vol. 21, pp. 1-5, 2024, Art no. 3001005, doi: 10.1109/LGRS.2024.3365299.