Comparison between oil spill images and look-alikes: an evaluation of SAR-derived observations of the 2019 oil spill incident along Brazilian waters

Abstract

Three SAR-derived observations of dark surface patches along the Northeastern Brazilian coastline by the end of 2019 were misreported in the Brazilian media as oil spill-related. Unfortunately, these observations were misled by false positives or look-alikes. Therefore, this paper aims to technically evaluate these look-alike classes by analyzing image attributes found to be helpful to the identification of ocean targets, including oil spills, rain cells, biofilms, and low wind conditions. We use image augmentation to extend our dataset size and create the probability density function curves. The processing includes image segmentation, optimal attribute extraction, and classification with random forest classifiers. Our results contrast with the open-source oil spill detection system and patch classifier methodology called “RIOSS.” Analysis of the feature probability density functions based on optimal attributes is promising since we could capture most of the false positive targets in the three SAR-reported images in 2019. The only exception was the biofilm slick observed on October 28th, where the RIOSS mistakenly classified this organic patch as a low wind region with oil spots. This pitfall is acceptable at this project stage since we had only five biogenic film samples to train the algorithm.

Key words
decision-tree; false-positive; random-forest; RIOSS; Sentinel-1; SAR