# Unsupervised Change Detection of Extreme Events Using ML On-Board

arXiv: https://arxiv.org/abs/2111.02995 · HTML: https://mlsanity.com/paper/2111.02995/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 5 of its 13 references that are in the map.
- First posted on arXiv 4 Nov 2021.

## Builds on (most cited first, 5 of 5)

- [Auto-Encoding Variational Bayes](https://mlsanity.com/paper/1312.6114/) (2013)
- [CNN-generated images are surprisingly easy to spot... for now](https://mlsanity.com/paper/1912.11035/) (2019)
- [Fully Convolutional Siamese Networks for Change Detection](https://mlsanity.com/paper/1810.08462/) (2018)
- [Unsupervised Change Detection in Satellite Images Using Convolutional Neural Networks](https://mlsanity.com/paper/1812.05815/) (2018)
- [Deep Active Learning in Remote Sensing for data efficient Change Detection](https://mlsanity.com/paper/2008.11201/) (2020)

## Built on by (most cited first, 3 of 3)

- [Self-supervised Learning in Remote Sensing: A Review](https://mlsanity.com/paper/2206.13188/) (2022)
- [Unsupervised Flood Detection on SAR Time Series](https://mlsanity.com/paper/2212.03675/) (2022)
- [Fast-SEnSeI: Lightweight Sensor-Independent Cloud Masking for On-board Multispectral Sensors](https://mlsanity.com/paper/2509.20991/) (2025)

## Abstract (arXiv)

In this paper, we introduce RaVAEn, a lightweight, unsupervised approach for change detection in satellite data based on Variational Auto-Encoders (VAEs) with the specific purpose of on-board deployment. Applications such as disaster management enormously benefit from the rapid availability of satellite observations. Traditionally, data analysis is performed on the ground after all data is transferred - downlinked - to a ground station. Constraint on the downlink capabilities therefore affects any downstream application. In contrast, RaVAEn pre-processes the sampled data directly on the satellite and flags changed areas to prioritise for downlink, shortening the response time. We verified the efficacy of our system on a dataset composed of time series of catastrophic events - which we plan to release alongside this publication - demonstrating that RaVAEn outperforms pixel-wise baselines. Finally we tested our approach on resource-limited hardware for assessing computational and memory limitations.