A Novel Approach for Oil Spill Detection from SAR Images

The main aim of this study is to identify instances of oil spills through the use of images taken from satellite. Synthetic Aperture Radar (SAR) imaging is a prevalent method for oil spills detection. Oil spills into the ocean are becoming commonplace, and they constitute a substantial portion of marine pollution, which can have negative impacts on the biological and economic systems. Object detection is a crucial and commonly studied area in image processing and earth observation, with important applications for society. The prompt detection of a marine oil spill is essential to the success of any cleanup operation, and this can be facilitated by efficient object detection methods. Currently, the system utilizes SENTINEL software that employs machine learning and deep learning techniques. However, the software has limitations in differentiating between mineral spills and biogenic surface films. MATLAB is a powerful technical computing language that enables users to perform complex computations, create visualizations, and write programs in an intuitive environment. It uses scientific notation to express problems and solutions in a straightforward manner. Image processing is utilized as a method for detecting oil spills, and this approach is implemented and tested using MAT LAB software. The accuracy of the results is verified through testing.