Latest revision as of 20:23, 1 February 2021

Abstract

occupied by oil pipelines and storage facilities are prone to severe contamination due to leaks caused by natural forces, poor maintenance or third parties. These threats have to be detected as quickly as possible in order to prevent serious environmental damage. Periodical and emergency monitoring activities need to be carried out for successful disaster management and pollution minimization. Airborne remote sensing stands out as an appropriate choice to operate either in an emergency or periodically. Hydrocarbon Index (HI) and Hydrocarbon Detection Index (HDI) utilize the unique absorption features of hydrocarbon based materials at SWIR spectral region. These band ratio based methods require no a priori knowledge of the reference spectrum and can be calculated in real time. This work introduces a flexible airborne pipeline monitoring system based on the online quasi-operational hyperspectral remote sensing system developed at Fraunhofer IOSB, utilizing HI and HDI for oil leak detection on the data acquired by an SWIR imaging sensor. Robustness of HI and HDI compared to state of the art detection algorithms is evaluated in an experimental setup using a synthetic dataset, which was prepared in a systematic way to simulate linear mixtures of selected background and oil spectra consisting of gradually decreasing percentages of oil content. Real airborne measurements in Ettlingen, Germany are used to gather background data while the crude oil spectrum was measured with a field spectrometer. The results indicate that the system can be utilized for online and offline monitoring activities.

Original document

The different versions of the original document can be found in:

• http://dx.doi.org/10.1117/12.2194973

• http://publica.fraunhofer.de/documents/N-378927.html

• https://ui.adsabs.harvard.edu/abs/2015SPIE.9644E..0HK/abstract,

https://www.spiedigitallibrary.org/conference-proceedings-of-spie/9644/1/Concept-of-an-advanced-hyperspectral-remote-sensing-system-for-pipeline/10.1117/12.2194973.full,

https://proceedings.spiedigitallibrary.org/proceeding.aspx?articleid=2466331,

https://core.ac.uk/display/56726152,

https://academic.microsoft.com/#/detail/2172550803