← Back to all news
arXiv Machine Learning By V\'it R\r{u}\v{z}i\v{c}ka, Adam Chlus, Andrew Thorpe, David R. Thompson

On-board ML for Trace Gas detection in Imaging Spectroscopy data

Read the original on arXiv Machine Learning →

The paper reports the first on‑board detection of methane point source emissions using imaging spectroscopy data from the AVIRIS‑5 sensor during the Tokyo Field Campaign in March 2026. It describes how a compact machine‑learning model was deployed on the aircraft to predict potential trace‑gas events in real time, circumventing communication bottlenecks that prevent full data downlink during flight. The approach enables immediate identification of transient gas releases without waiting for ground‑based processing.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

Related stories

arXiv Machine Learning
Jul 30

Global monitoring of methane point sources using deep learning on hyperspectral radiance measurements from EMIT

arXiv:2604. 10094v2 Announce Type: replace-cross Abstract: Anthropogenic methane (CH4) point sources are critical drivers of near-term climate forcing, safety hazards, and system-inefficiencies.

By Vishal V. Batchu, Michelangelo Conserva, Alex Wilson, Anna M. Michalak, Varun Gulshan, Philip G. Brodrick, Andrew K. Thorpe, Christopher V. Arsdale
llmscomputer-visionefficiencybenchmarkssafety
More like this →
arXiv AI
Sep 7

Methane Detection On Board Satellites from Unorthorectified Imagery

The paper introduces UnorthoDOS, a dataset and machine‑learning approach that enables methane plume detection directly on unorthorectified hyperspectral satellite imagery. Using U‑Net models trained on this data, the authors achieve performance close to models trained on orthorectified images (IoU 16.91% vs. 18.47%) and far surpass the traditional matched‑filter baseline (IoU 4.76%). They also demonstrate that FP16 compression can reduce model size by half with negligible loss in output accuracy, making onboard deployment feasible.

By Luca Marini, Maggie Chen, Hala Lamdouar, Laura Mart\'inez-Ferrer, Dr C. P. Bridges, Giacomo Acciarini
More like this →
arXiv Machine Learning
Aug 11

Real-time physics inversion for retrieval of sub-pixel wildfire temperatures from VSWIR imaging spectroscopy

arXiv:2608. 07580v1 Announce Type: cross Abstract: In this work, we present a wildfire temperature retrieval framework for VSWIR imaging spectroscopy data, employed on data from NASA's Airborne Visible Infrared Imaging Spectrometer (AVIRIS-3).

By William R. Keely, Philip G. Brodrick, Katherine Mistick, Adam Chlus, Robert O. Green, Philip E. Dennison
benchmarks
More like this →
arXiv AI
Oct 9

AI-Based On-Board Maritime Object Detection for Earth Observation Payload Data Reduction on Versal Embedded Hardware

The paper presents an AI-based system for onboard maritime object detection on Earth‑observation satellites, aiming to reduce data volume by selectively downlinking relevant vessel information. A YOLOX‑S detector is trained on a dataset of 68 Maxar scenes with 43 vessel classes, then quantized and deployed on a Versal VC1902 DPU with minimal loss in accuracy and a few‑second processing time per scene. Various downlink modes—metadata only, image crops, tiles, or full scenes—are evaluated, showing that tiles retain 98% of vessels while reducing scene volume to 29%, and crops retain 83% with only 3% of the volume.

By Thomas Goudemant, Aur\'elien Bobey, Omar Hlimi, Marjorie Bellizzi
computer-visionefficiency
More like this →