An Emulated Platform for Detecting and Size Measuring of Oil Spills Using An Unmanned Aerial Vehicle System

Abstract

This paper presents the machine vision techniques implemented in an emulated environment that mimics the detecting process of oil spills in the ocean. An image processing algorithm is developed to achieve accurate detection and size measurement for studying oil spill cases. This study demonstrates the required setups and adjustments performed to mimic this process on a smaller scale, in a lab-based experiment. The proposed emulated system generates the required path for the quadrotor, used in this experiment, to maneuver around the arena and detect the oil spill. The drone successfully detected and accurately provided the location of the oil spot for several trails. For these attempts, the areas of the detected oil were calculated and compared. Some discussions were stated regarding some findings and exceptional cases. In general, the results attest to the efficacy of the proposed system.