Fully Convolutional Siamese Networks for Change Detection

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...

Full description

Saved in:
Bibliographic Details
Published inProceedings - International Conference on Image Processing pp. 4063 - 4067
Main Authors Caye Daudt, Rodrigo, Le Saux, Bertr, Boulch, Alexandre
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.10.2018
Subjects
Cats
Change detection
Change detection algorithms
Computer architecture
Earth
Earth observation
fully convolutional networks
Image analysis
Machine learning
supervised machine learning
Training
Online AccessGet full text

Cover

More Information
Summary: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.
ISSN:2381-8549
DOI:10.1109/ICIP.2018.8451652