Built independently by an author, for readers. Read the story and support ChapterPal
ChapterPal Logo
Log inSign up

keyword

Onera Satellite Change Detection

Onera Satellite Change Detection is an open-access benchmark dataset designed for training and evaluating machine learning algorithms that identify surface modifications in Earth observation imagery over time. Created by researchers at the French aerospace laboratory ONERA, the dataset contains coregistered pairs of multispectral images acquired by Sentinel-2 satellites between 2015 and 2018 across diverse geographic locations worldwide. It provides manually annotated, pixel-level binary change masks that primarily highlight urban and artificial developments, such as new buildings and infrastructure, while excluding natural variations resulting from seasonality or weather. Due to its standardized multi-band data and precise ground truth annotations, it serves as a standard resource for developing and assessing deep learning models, such as convolutional and Siamese neural networks, in remote sensing and urban monitoring applications.

1 item

Fully Convolutional Siamese Networks for Change Detection

Fully Convolutional Siamese Networks for Change Detection

Rodrigo Caye Daudt, Bertrand Le Saux, Alexandre Boulch

OrganizationsDTISInstitut polytechnique de ParisONERATélécom ParisUniversité Paris-Saclay

Why you should read this

Proposes fully convolutional Siamese architectures for remote sensing change detection that train from scratch on coregistered image pairs to deliver superior accuracy at over 500 times the speed of previous methods.

This paper presents three fully convolutional neural network architectures which perform change detection using a pair of coregistered images. Most notably, we propose two Siamese extensions of fully convolutional networks which use heuristics about the current problem to achieve the best results in our tests on two open change detection datasets, using both RGB and multispectral images. We show that our system is able to learn from scratch using annotated change detection images. Our architectures achieve better performance than previously proposed methods, while being at least 500 times faster than related systems. This work is a step towards efficient processing of data from large scale Earth observation systems such as Copernicus or Landsat.

Added

2026-09-24