Rodrigo Caye Daudt

About Me

I received my Electrical Engineering degree from Universidade de Brasília in 2015, and my MSc in Computer Vision and Robotics from the VIBOT Erasmus Mundus joint masters degree (Université de Bourgogne, University of Girona, Heriot-Watt University). My main areas of expertise are computer vision, machine learning and image processing, and I am currently working on the problem of applying deep learning for the analysis of multi-temporal series of Earth observation images for my PhD thesis.

Link to personal website.

Onera Satellite Change Detection Dataset

The Dataset

The Onera Satellite Change Detection dataset addresses the issue of detecting changes between satellite images at different dates.

It comprises 24 pairs of multispectral images taken from the Sentinel-2 satellites between 2015 and 2018. Locations have been chosen all over the world: in Brazil, USA, Europe, Middle-East and Asia. For each location, registered pairs of 13-band multispectral satellite images obtained by the Sentinel-2 satellites are provided. Images vary in spatial resolution between 10m, 20m and 60m.

Pixel-level change ground truth is provided for 14 of the image pairs. The annotated changes focus on urban changes, such as new buildings or new roads. These data can be used for training and setting parameters of change detection algorithms.

Example: “beirut” image pair and associated change map.
Example: “beirut” image pair and associated change map.

 

The Benchmark

The algorithms can be tested in a benchmark for change detection.

The ground truth for the 10 remaining images remain undisclosed. Change prediction maps can be uploaded for evaluation on the IEEE GRSS DASE website. Various metrics such as per-class accuracy and confusion matrices are automatically computed on the website, and are available for participants. Comparison to the best performing methods is provided in the leader board associated with this benchmark.

 

Links

You can find the dataset at IEEE GRSS DASE webpages:

 

References

If you use this work for your projects, please take the time to cite our paper:

Urban Change Detection for Multispectral Earth Observation Using Convolutional Neural Networks R. Caye Daudt, B. Le Saux, A. Boulch, Y. Gousseau IEEE International Geoscience and Remote Sensing Symposium (IGARSS’2018) Valencia, Spain, July 2018

@inproceedings{daudt-igarss18,
  author = {{Caye Daudt}, R. and {Le Saux}, B. and Boulch, A. and Gousseau, Y.},
  title = {Urban Change Detection for Multispectral Earth Observation Using Convolutional Neural Networks},
  booktitle = {IEEE International Geoscience and Remote Sensing Symposium (IGARSS'2018)},
  venue =  {Valencia, Spain},
  month = {July},
  year = {2018},
}

The Team

Rodrigo Caye Daudt, rodrigo.daudt@onera.fr
Bertrand Le Saux, bertrand.le_saux@onera.fr
Alexandre Boulch, alexandre.boulch@onera.fr
Yann Gousseau, yann.gousseau@telecom-paristech.fr