Saudi Arabian Oil Company

King Abdullah University of Science and Technology

Alaa A Elyas

Ibrahim Hoteit

Ali Mothana Qasem

Khaled M Asfahani

Sabique Langodan

Systems and methods for tracking and remediating oil in a body of water include monitoring the data from the satellite for indications of an oil spill. In response to determining that the indications of oil spill are present, assessing whether criteria for triggering a forecasting process are met. In response to determining that criteria for triggering a forecasting process have been met, running a forecasting process including: (i) delineating polygons where oil is determined to be present based on the data from the satellite; (ii) receiving meteorological and hydrodynamic data for the body of water; (iii) running a trajectory model using the polygons as initial conditions for the oil spill to forecast future locations of the oil spill; and sending instructions to a vessel in the body of water.

Identifying and remediating oil spills

Systems and methods for quantifying and remediating an oil spill in a body of water can include obtaining a synthetic aperture radar image of the body of water and/or a multispectral image of the body of water. One or more features representing a physical feature of a surface of the body of water can be extracted from the image(s). The extracted features can be processed using a machine learning model trained with labeled image data representing instances of oil on the surface of the body of water to associate oil appearances code with portions of the surface of the water body based on the extracted features. Based on the processing, areas of the body of water associated with each oil appearance code as well as locations and volumes of oil in the body of water can be determined.

Systems and methods for identifying an oil spill in a body of water include obtaining an image of the body of water from a multispectral satellite for a first time period and a second time period. One or more features are extracted the from the image to form a first feature vector for the first time period and a second feature vector for the second time period with the one or more features representing a physical feature of a surface of the body of water. The feature vectors are processed using a machine learning model trained with labeled image data representing instances of oil on the surface of the body of water to determine the type and location of oil in the body of water.

Growing community of inventors

Dhahran, Saudi Arabia

Alaa A Elyas