Home // PATTERNS 2021, The Thirteenth International Conference on Pervasive Patterns and Applications // View article

Detection of Gas Flares Using Satellite Imagery

Authors:
Alexander Troussov
Dmitry Botvich
Sergey Vinogradov

Keywords: night lights, light pollution, viirs, image processing, machine learning, deep learning.

Abstract:
During the extraction, transportation and processing of oil, associated petroleum gas is formed, which is usually disposed in flares. Monitoring these flares is an important environmental challenge. Objective instrumental methods for detecting gas flares and assessing the volumes of gas burnt on them are based on multi-spectral remote sensing of the nighttime Earth. To refine the list, a manual check is used, which includes a visual examination of the locations of the alleged gas flares on high-resolution daytime images. Automating this check reduces the cost of monitoring flares. The paper proposes a method for verifying the list of high-temperature anomalies, based on the classification of images of objects in daytime images. The classification is carried out using machine learning methods.

Pages: 45 to 49

Copyright: Copyright (c) IARIA, 2021

Publication date: April 18, 2021

Published in: conference

ISSN: 2308-3557

ISBN: 978-1-61208-850-1

Location: Porto, Portugal

Dates: from April 18, 2021 to April 22, 2021