Detecting Offshore Drilling Rigs with Multitemporal NDWI: A Case Study in the Caspian Sea

Offshore drilling rigs are the foundation of oil and gas exploitation in water areas. Their spatial temporal distribution, state attributes other information directly reflect level regional resources. Therefore, it is very important to build an automatic detecting method for offshore with good performance accurately capture distribution activities. At present, there two main groups methods rigs: invariant feature-based nighttime firelight-based methods. Methods based on location more subjective terms their parameter settings require intensive computation. Nighttime light-based largely unable identify without associated waste ignition. Furthermore, multiple close proximity one another cannot be effectively distinguished low resolution imagery. To address these shortcomings, we propose a new identification Landsat-7 ETM+ images from 2018 2019, taking Caspian Sea as research area. We nominal annual cloud shadow-free Normalized Difference Water Index (NDWI) composite by designing optimal NDWI compositing influence shadow values water, bare land (island) rigs. The classification objects simultaneously done during process, following rules: body (Max_NDWI > 0.55), (Min_NDWI < −0.05) rig (0 Mean_NDWI 0.4). A threshold segmentation postprocessing were carried out further refine results. Using this method, 497 platforms automatically identified using image Google Earth Engine. Validation Sentinel-2 Multispectral Imager (MSI) demonstrated that correct rate detection 90.2%, missing judgment 5.3% wrong 4.5%, proving proposed method. This can used within large surface area relatively quickly, which great significance exploring status It also extended finer optical remote sensing images; thus small-size detected.