Loading [MathJax]/extensions/MathMenu.js
Hyperspectral Remote Sensing Benchmark Database for Oil Spill Detection With an Isolation Forest-Guided Unsupervised Detector | IEEE Journals & Magazine | IEEE Xplore

Hyperspectral Remote Sensing Benchmark Database for Oil Spill Detection With an Isolation Forest-Guided Unsupervised Detector


Abstract:

Oil spill detection has attracted increasing attention in recent years, since marine oil spill accidents severely affect environments, natural resources, and the lives of...Show More

Abstract:

Oil spill detection has attracted increasing attention in recent years, since marine oil spill accidents severely affect environments, natural resources, and the lives of coastal inhabitants. Hyperspectral remote sensing images provide rich spectral information which is beneficial for the monitoring of oil spills in complex ocean scenarios. However, most of the existing approaches are based on supervised and semi-supervised frameworks to detect oil spills from hyperspectral images (HSIs), which require a massive amount of effort to annotate a certain number of high-quality training sets. In this study, we make the first attempt to develop an unsupervised oil spill detection method based on isolation forest (iForest) for HSIs. First, a Gaussian statistical model is designed to remove the bands corrupted by severe noise. Then, kernel principal component analysis (KPCA) is employed to reduce the high dimensionality of the HSIs. Next, the probability of each pixel belonging to one of the classes of seawater and oil spills is estimated with the iForest, and a set of pseudolabeled training samples is automatically produced using the clustering algorithm on the detected probability. Finally, an initial detection map can be obtained by performing the support vector machine (SVM) on the dimension-reduced data, and the initial detection result is further optimized with the extended random walker (ERW) model so as to improve the detection accuracy of oil spills. Experiments on hyperspectral oil spill database (HOSD) created by ourselves demonstrate that the proposed method obtains superior detection performance with respect to other state-of-the-art detection approaches. We will make HOSD and our developed library for oil spill detection publicly available at https://github.com/PuhongDuan/HOSD to further promote this research topic.
Article Sequence Number: 5509711
Date of Publication: 20 April 2023

ISSN Information:

Funding Agency:


I. Introduction

In the wake of marine oil exploration and transportation, the accidents of oil spills have occurred frequently around the world, which leads to the severe pollution of the marine environment and the huge damage of coastal species [1], [2], [3]. On April 20, 2010, the explosion of Deepwater Horizon oil drilling platform led to a severe leakage. Millions of barrels of oil polluted the Gulf of Mexico (GM) with the area of about 10000 square kilometers [4], [5]. Due to this accident, the marine ecosystems, such as fish and seabirds, have been seriously destroyed. On June 4, 2011, the Penglai 19–3 oilfield in Bohai Bay, Northeast China, witnessed a serious oil spill incident, which caused the leak of more than 7000 tons of oil into the sea [6], [7], [8]. The polluted area was almost 6200 square kilometers. If the oil would not be timely monitored after the leakage, the oil slick would be washed onto the coast by the sea waves. This situation would pose a huge threat to coastal aquaculture fishery resources and human health. Therefore, it is of great importance to effectively detect oil spills on the sea surface to monitor the distribution, impact, and volume of oil spills [9].

Contact IEEE to Subscribe

References

References is not available for this document.