UAV-based thermal anomaly detection for distributed heating networks

verfasst von
A. Sledz, J. Unger, C. Heipke
Abstract

District Heating Systems (DHS) distribute heat in terms of hot water or steam. Loss of media (water or steam), and thus energy, is expensive and has a negative impact on the environment. It is therefore of great interest to develop techniques to detect and localize potential leakages fast and cost-effectively. To avoid interference with the operating process of a DHS, airborne thermography comes into place. The use of Unmanned Aerial Vehicles (UAV) as a flexible and low-cost platform equipped with a Thermal InfraRed (TIR) camera is a promising alternative to a conventional manned flight.

This paper describes a method for automatic thermal anomaly detection of a DHS. Thermal data acquisition using a UAV is followed by photogrammetric processing of the TIR images. In this way, a thermal orthophoto is produced. The next step is an identification of anomalies by means of image analysis. We apply the Laplacian of Gaussian (LoG) blob detector to find high temperature regions in areas of interest of a thermal orthomosaic. This area of interest is defined around the DHS position in the images as defined in a Geographic Information System. Finally, segmentation and classification are employed to reduce false alarms and localize thermal anomalies. An experimental evaluation using real-world data is presented, showing that the developed method deliverers promising results.

Organisationseinheit(en)
Institut für Photogrammetrie und Geoinformation
Typ
Konferenzaufsatz in Fachzeitschrift
Journal
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives
Band
43
Seiten
499-505
Anzahl der Seiten
7
ISSN
1682-1750
Publikationsdatum
06.08.2020
Publikationsstatus
Veröffentlicht
Peer-reviewed
Ja
ASJC Scopus Sachgebiete
Information systems, Geografie, Planung und Entwicklung
Ziele für nachhaltige Entwicklung
SDG 7 – Erschwingliche und saubere Energie
Elektronische Version(en)
https://doi.org/10.5194/isprs-archives-XLIII-B1-2020-499-2020 (Zugang: Offen)