Offshore Petroleum Leaking Source Detection Method From Remote Sensing Data via Deep Reinforcement Learning With Knowledge Transfer

Abstract

A marine oil spill is an environmental pollution incident that generally has the attributes of a high speed, widespread, and long duration. It seriously threatens the marine ecological environment and related industries. It is vital to determine the source of the oil leakage so that it may be stopped and related hazards can be reduced. Oil spill accidents in the sea are generally located in offshore and navigation channels. With the rapid development of remote-sensing techniques, oil leak extraction using remote-sensing data has played an essential role in oil spill research. This paper proposes a Monte Carlobased Deep Q-Transfer-learning Network (DQTN) offshore oil leak detection method that uses remote-sensing data. Remotesensing data are utilized to continuously monitor a marine oil spill on the surface. The Estuarine and Coastal Ocean Model (ECOM) is utilized to simulate a marine oil spill event. The Deep Q Network (DQN) method with offline transferred knowledge is then utilized to determine the marine oil spill source location. In an experiment, based on the Bohai oil spill incident on June 2, 2011, the effectiveness of the remote-sensing-based DQTN marine oil spill search algorithm is verified. The accuracy of the targeted oil spill point is up to 98.97%.

Improving Geological Remote Sensing Interpretation Via a Contextually Enhanced Multiscale Feature Fusion Network

“…DL methodologies leverage their robust nonlinear adaptability and feature learning transformation capabilities, achieving remarkable feats across various remote sensing tasks in recent years [34]. These methodologies can automatically learn spectral and spatial information from remote sensing and geographical data, translating them into advanced feature representations [35], [36]. As a result, they dominate image processing fields like image classification, semantic segmentation, and object detection.…”

Inversion of Reservoir Parameters for Oil Extraction Based on Deformation Monitoring With InSAR

“…In order to lessen this occurrence, we employ the two-sample t-test to find statistically homogeneous pixels (SHP) and the Eigendecomposition-based Maximumlikelihood-estimator of Interferometric phase (EMI) to optimise phase [36][37][38]. Furthermore, the temporal coherence of distributed targets is calculated by (1), after which pixels with coherence greater than 0.5 are identified as Distributed Scatterers Candidates (DSCs). Then, adding them to the time series InSAR method for the study of deformation monitoring.…”

“…This strategic commodity is closely related with the progress of human civilization. Nevertheless, the continual extraction will deplete the reservoir pressure, which could enhance the viscosity of crude oil and pose a significant threat to recovery [1]. To alleviate this phenomenon, the frequent method is to inject produced water into the reservoir [2].…”