GRSG Conference 2022: Orbit to Outcrop
Title: Evaluating image processing techniques for detecting active petroleum seepage using ASTER and Sentinel-2 data sets
Author: Mohammad H. Tayebi, Diego Fernando Ducart, Carlos Roberto de Souza Filhoa, Majid H. Tangestani
The goal of this study is to evaluate various image processing techniques for identifying Koh-e-Sokhteh active petroleum seepage in the Aghajari oil field, Zagros fold-and-thrust belt (ZFTB), SW Iran. Koh-e-Sokhteh as a unique geotourism destination provides surface manifestation for petroleum seepages in the Gachsaran Formation covering the Asmari petroleum reservoir. The spectral and geochemical characteristics of Koh-e-Sokhteh, altered and un-altreated rock units were characterized using spectroscopic investigations and X-ray diffraction (XRD) analysis.
Image end-members including altered and un-altreated rock units were extracted from the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER), and Sentinel-2 MSI (MultiSpectral Instrument) data sets using Pixel Purity Index (PPI) technique and compared to the field spectra. The Adaptive Coherence Estimator (ACE), Constrained Energy Minimization (CEM), Matched Filtering (MF), Spectral Angle Mapper (SAM), Orthogonal Subspace Projection (OSP), and Target-Constrained Interference-Minimized Filter (TCIMF) methods were evaluated.
Image classification results indicated that the ACE, SAM, OSP and TCIMF methods could not discriminate altered rocks precisely, whereas the CEM and MF results were more effective in altered rocks mapping. This study’s findings can be used to identify active petroleum seepages and improve hazard analysis in the oil fields.