Monica Magdy, Omar Saleh, Monica Hany, Madonna Said

Supervised by : Dr. Ayman Nabil

Assistant Supervised by: Eng. Youmna Ibrahim, Eng. Maha sayed

Publishing Date



The high demand of oil caused a huge loss in form of oil spills during the process of exporting the product, which lead to an increase in the pollution specially the marine environment. Our project maintain this problem using modern technology by detecting Oil spills using satellite imagery. This detection is achieved with the assistance of various machine learning technologies and image processing methods using satellite hyperspectral images. The geologist team had defined hyperspectral images as the principle feature which focus on detecting the oil spills during all the process during the exporting process. The system uses this spectrum and special feature that will be extracted from images using some filters and classifiers. we perform preprocessing on the image, going through the process then using various classifiers to accomplish the best accuracy. The study area was decided based on the huge amount of oil detected in such as Gulf Of Mexico which is obtained using AVIRIS dataset.

1.1 Background

With in the last century, Crude oil became the most demanded mineral in the global industry as it supports more than 40% of the global energy needs. It has been proven by theories and methodologies that Oil prices has impact on the world’s economy. Crude oil has multiple usability in various fields of economical development made it essential for human life to grow. The main reason of that is limitation more oil resources are found the more Oil became our fuel supplier for our vehicles, an energy to create power for our homes, run machine lubricants and raw materials for a number of manufacturing industries is utilized in a wide scope of things we use in day by day life. The oil industry includes the environmental procedure of investigation, extraction, refining, transferring and promoting oil production. The Oil industry depend on some main processes to be able to product Oil. One of those processes is detecting the oil fields which detecting team to know about the oil properties. The Oil is mainly formed under the earth’s surface which was such a mystery in the beginning of it’s discovery due to the significant properties oil contain. Over years geologists and after tons of research they discovered that by the decomposition of plants and marine animals that has been crushed under the weight of sedimentation have produced crude oil. Over a million of years that process happened and due to the low density of crude oil it have arisen over water and filled the earth’s crust pores and reservoirs is formed. It’s known that the oil that are found in the wells are mixed with natural gases and hydrogen carbon molecules. Throughout the years, geologists have found that those hydrocarbon components emits some sort of emissions that made them conclude that their might be an oil reservoir.

1.2 Motivation

The main idea that motivated us to work on that system is combining between the machine learning utilities and image processing as it became a very interesting field to most computer science researchers. Also, using hyperspectral images not in detecting the oil spill but detecting the oil wells that will be an effective help to the geologists teams during their researches. Hyperspectral imagery is a collection of satellite sensors which is able to identify and distinguish spectral unique materials. this process id done by collecting and processing hundreds of contiguous narrow wavebands. The main advantage of the hyperspectral image provides a higher accuracy for potential results more than any remotely sensed data [7]. That will down size the cost and the time taken for the extraction of oil and will prevent the oil spill from happening.

1.3 Problem Statement

The huge amount of oil spills whether near the reservoir or while the exporting process, such spills cause damage to the environment specially the marine one. Our goal is to develop a system able to identify oil spills locations.