Deep learning approaches to Earth Observation change detection

TL;DR

This paper explores two deep learning methods using convolutional neural networks for change detection in satellite imagery, aiming to improve accuracy and efficiency in applications like urban growth analysis.

Contribution

It introduces two novel CNN-based approaches—semantic segmentation and classification—for change detection in remote sensing images.

Findings

Both approaches achieve good detection accuracy.

Methods are suitable for large-scale applications.

Post-processing enhances results.

Abstract

The interest for change detection in the field of remote sensing has increased in the last few years. Searching for changes in satellite images has many useful applications, ranging from land cover and land use analysis to anomaly detection. In particular, urban change detection provides an efficient tool to study urban spread and growth through several years of observation. At the same time, change detection is often a computationally challenging and time-consuming task, which requires innovative methods to guarantee optimal results with unquestionable value and within reasonable time. In this paper we present two different approaches to change detection (semantic segmentation and classification) that both exploit convolutional neural networks to achieve good results, which can be further refined and used in a post-processing workflow for a large variety of applications.