Transforming volcanic monitoring: A dataset and benchmark for onboard volcano activity detection

2025-10-26

AI Summary

Volcanic monitoring suffers from a lack of high-quality annotated datasets and high latency due to ground-based processing. Method: This work constructs the first global multi-source remote sensing dataset specifically annotated for volcanic anomaly detection—covering thermal anomalies, eruptions, and volcanic ash emissions—and proposes a lightweight deep learning model. It further achieves, for the first time, end-to-end on-board inference on the Intel Movidius Myriad X VPU, integrated within a small-satellite constellation architecture. Contribution/Results: Full on-orbit validation demonstrates a closed-loop latency of <30 seconds—from imaging to anomaly alert—reducing delay by over 90% compared to conventional ground processing. This enables a paradigm shift from “ground-centric” to “on-board distributed intelligent sensing,” establishing a scalable technical pathway for near-real-time volcanic hazard early warning.

Technology Category

• Computer Vision: Remote Sensing / Geospatial AI
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Learning on the Edge & Model Compression

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Data management and stream processing for Web, mobile and wireless applications
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Natural disasters, such as volcanic eruptions, pose significant challenges to daily life and incur considerable global economic losses. The emergence of next-generation small-satellites, capable of constellation-based operations, offers unparalleled opportunities for near-real-time monitoring and onboard processing of such events. However, a major bottleneck remains the lack of extensive annotated datasets capturing volcanic activity, which hinders the development of robust detection systems. This paper introduces a novel dataset explicitly designed for volcanic activity and eruption detection, encompassing diverse volcanoes worldwide. The dataset provides binary annotations to identify volcanic anomalies or non-anomalies, covering phenomena such as temperature anomalies, eruptions, and volcanic ash emissions. These annotations offer a foundational resource for developing and evaluating detection models, addressing a critical gap in volcanic monitoring research. Additionally, we present comprehensive benchmarks using state-of-the-art models to establish baselines for future studies. Furthermore, we explore the potential for deploying these models onboard next-generation satellites. Using the Intel Movidius Myriad X VPU as a testbed, we demonstrate the feasibility of volcanic activity detection directly onboard. This capability significantly reduces latency and enhances response times, paving the way for advanced early warning systems. This paves the way for innovative solutions in volcanic disaster management, encouraging further exploration and refinement of onboard monitoring technologies.

Problem

• Addressing the lack of annotated datasets for volcanic activity detection
• Developing onboard detection models to reduce latency in monitoring
• Establishing benchmarks for robust volcanic anomaly identification systems

Innovation

• Introduces a novel volcanic activity detection dataset
• Establishes benchmarks using state-of-the-art detection models
• Demonstrates onboard detection feasibility using Intel VPU