Oil Spill Detection Research Articles

Harnessing Deep Learning for Underwater plastic Trash Identification

Abstract—Machine and deep learning (DL) offer significant opportunities for exploring and monitoring oceans and for tackling important problems ranging from litter and oil spill detection to marine biodiversity estimation. Reasonably priced hardware platforms, in the form of autonomous (AUV) and remote operated (ROV) underwater vehicles, are also becoming available, fuelling the growth of data and offering new types of ap- plication areas. This article presents a research vision for DL in the oceans, collating applications and use cases, identifying opportunities, constraints, and open research challenges. We conduct experiments on underwater marine litter detection to demon- strate the benefits DL can bring to underwater envi- ronments. Our results show that integrating DL in underwater explorations can automate and scale-up monitoring, and highlight practical challenges in enabling underwater operations. This project introduces a refined YOLOv8-based algorithm tailored for the en- hanced detection of small-scale underwater debris, to mitigate the prevalent challenges of high miss and false detection rates . The research presents the YOLOv8 algorithm, which optimizes the backbone, neck layer, and C2f module for underwater characteristics and incorporates an effective attention mech- anism. This algorithm improves the accuracy of underwater trash detection while simplifying the computational model. Empirical evidence underscores the superiority of this method over the other conventional network, manifesting in a significant uplift in detection performance. Notably, the proposed method realized a 63% mean average precision (mAP50), a 60% surge in recall (R). Transcending its foundational utility in marine conservation, this methodology harbors potential for subsequent integra- tion into remote sensing ventures. Such an adaptation could substantially enhance the precision of detection models, particularly in the realm of localized surveillance, thereby broad- ening the scope of its applicability and impact.

Oil spill detection and classification through deep learning and tailored data augmentation

Oil spill detection is an important task for protecting and minimizing the harmful effects of oil on the marine ecosystem. Currently, the application of images from unmanned aerial vehicles, along with deep learning, is widely employed. Although these methods have yielded good results, the issue of oil spill classification based on these methods has not received much attention at present. In this research, a deep learning model with a dual attention mechanism consisting of two modules was utilized. The first module focuses on capturing the spatial relationships between each pixel and the entire image, the second module identifies the characteristics between channels in the image, thereby enhancing the ability to detect and classify oil. Additionally, a data augmentation technique based on the Generative Adversarial Networks model was refined and employed to improve the model's accuracy. Experimental results, obtained through comparisons between dataset construction methods, the use of different encoders and decoders, and adjustments hyperparameters, reveal that the best model achieves a mean Intersection over Union by 72.49%. Data augmentation techniques also contribute to a 2.56% increase in mean Intersection over Union. The findings of this research provide a feasible solution not only for detecting but also for classifying oil spills, thereby assisting marine environmental managers in making timely decisions to respond to oil spill accidents.

The application of laser‑induced fluorescence in oil spill detection.

Over the past two decades, oil spills have been one of the most serious ecological disasters, causing massive damage to the aquatic and terrestrial ecosystems as well as the socio-economy. In view of this situation, several methods have been developed and utilized to analyze oil samples. Among these methods, laser-induced fluorescence (LIF) technology has been widely used in oil spill detection due to its classification method, which is based on the fluorescence characteristics of chemical material in oil. This review systematically summarized the LIF technology from the perspective of excitation wavelength selection and the application of traditional and novel machine learning algorithms to fluorescence spectrum processing, both of which are critical for qualitative and quantitative analysis of oil spills. It can be seen that an appropriate excitation wavelength is indispensable for spectral discrimination due to different kinds of polycyclic aromatic hydrocarbons' (PAHs) compounds in petroleum products. By summarizing some articles related to LIF technology, we discuss the influence of the excitation wavelength on the accuracy of the oil spill detection model and proposed several suggestions on the selection of excitation wavelength. In addition, we introduced some traditional and novel machine learning (ML) algorithms and discussed the strengths and weaknesses of these algorithms and their applicable scenarios. With an appropriate excitation wavelength and data processing algorithm, it is believed that laser-induced fluorescence technology will become an efficient technique for real-time detection and analysis of oil spills.

A near real-time automated oil spill detection and early warning system using Sentinel-1 SAR imagery for the Southeastern Mediterranean Sea

ABSTRACT The ecological and environmental impact of marine oil pollution underlines the importance and necessity of an oil spill surveillance system. This study proposes an operational automated oil spill detection and early warning system to help take quick action for oil combating operations. Oil slicks in the spaceborne Sentinel-1 synthetic aperture radar (SAR) data are detected by a trained deep learning-based oil object detector. These detected oil objects are segmented into binary masks based on the similarity and discontinuity of the backscattering coefficients, and their trajectory is simulated. The detection process was tested on one-year SAR acquisitions in 2019, covering the Southeastern Mediterranean Sea; the false discovery rate (FDR) and false negative rate (FNR) are 23.3% and 24.0%, respectively. The system takes around 1.5 h from downloading SAR images to providing slick trajectory simulation. This study highlights the capabilities of using deep learning-based techniques in an operational oil spill surveillance service.

A Scheme for Credibility of Surface Currents Derived From High Frequency Radars

Large amount of real-time ocean currents measured by HF radar have been included into the ocean observation database in many countries and regions, in support of applications for various marine activities such as marine research, Oil spill detection, tsunami warning, search and rescue. However, due to the performance of inversion algorithms and radar antennas, and other reasons, there are differences in the credibility of ocean current results at different times and locations, which brings on difficulties to applications of ocean currents. This letter proposes a method for developing a credibility model of ocean current results, mainly for all-digital multi-input and multi-output (MIMO) HF radar. By comparing the model with field experimental results, it is confirmed that the method is feasible. It will be very beneficial for radar users to select and use ocean current results, greatly reducing the work of verification of newly installed radars, and will be an innovation in the application technology of over-the-horizon marine radars.

Modelling and Mapping Oil Pollution across the Inland and Coastal Waters: A Case of Selected Segment of Gulf of Guinea

Artisanal fisheries in the Niger Delta and Nigeria’s coastal waters support the livelihood of its inhabitants, thus management of water pollution has public health significance for the regions and beyond. Oil Spills and contaminated runoff from the hinterland always find their way to the coast via the extensive network of rivers and creeks across the delta. This is a major cause of marine pollution in the region. With this understanding there is a need to map and model the dynamics of oil pollution across the coastal waters of the Niger Delta, thereby providing evidence and data support for their management. This study explored the potential of mapping oil slicks along the inland and coastal water of the Niger Delta using the SENTINEL-1 Synthetic Aperture Radar (SAR). The Interferometric Wide Swath Mode imagery Ground Range Detected Imagery for the 2nd of January 2019 was acquired for the study area, processed using Sentinel Application Platform (SNAP) and the oil spill detection algorithm was applied on the Amplitude Vertical-Vertical Band. After preprocessing, the non-polluted areas were found to have high backscatter (amplitude >18.0). For the impacted areas, lower values were recorded - amplitude decibel value of between 15.9 and 18.0. A total of 329 polygons of oil slick were identified, covering an area of 46.34 km2. The impacted surface area ranges between 0.000064 – 15 km2. The farthest impacted surface can be found around 18.25km from the Bonny coastline. Results showed that a significant proportion of the oil slicks are concentrated along the upper reaches of the Bonny River spanning Okrika and Ogu/Bolo LGA. Towards the nearshore, the oil slicks were disaggregated into several units. Contiguous units were found closer to the banks and coastline. Out of the 329 polygons, 14 were identified along the New Calabar River, emptying into the Bonny Estuary. The impacted surfaces along the New Calabar River were smaller in size compared to those found along the Bonny River. The results identified the pattern and distribution of oil slicks across a segment of the Gulf of Guinea dominated by oil and gas industry activities. This study highlighted the potential of remotely sensed data to support coastal pollution management efforts. It further highlighted the potential for spatiotemporal monitoring of oil slicks along coastal and inland waters. Challenges for deploying such a platform include the high computing power required and the lack of in-situ monitoring data for validation.

Integration of satellite surveillance and metagenomics for the monitoring and protection of water basins from oil spills

The planetary crisis regarding water resources means that new methods are needed to monitor large areas of water basins that are threatened by chemical and natural pollutants, together with climate change. With the aim to detect oil spill, we applied an algorithm, which could consistently and reliably confirm the presence of oil in four polluted lake waters analyzed. Combined algorithm application and metagenomic analysis from the spill areas that had been detected by the satellite identified drivers of the microbial response to oil pollution. Based on ortholog abundances, metabolic pathway reconstruction carried out in PICRUSt2 highlighted the degradative capacity of these microorganism. These microorganisms could be suitable candidates for treatment of crude oil, aromatic hydrocarbons and the desulfurization of persistent petroleum substances like dibenzothiophene.Environmental changes have been analysed with the combination of satellite monitoring and metagenomic in other studies. Red snow phenomenon, in Franz Josef Land's Arctic observed by satellite, is the result of a microbial succession dominated by Chlamydomonas nivalis, a unicellular, red-colored photosynthetic green algae. Similarly, satellite monitoring and metagenomic monitoring were used to assess the impact on coral reefs of a huge quantity of mud spill from iron ore mining on the Abrolhos Bank reef in Brazil.In our study, the combination of satellite sensing and metagenomics analyses offer useful tools for the real-time monitoring of water bodies threatened by oil spills, as well as for the design of recovery strategies based on the use of valuable hydrocarbonoclastic microorganisms.

Marine oil spill detection using improved polarimetric feature based on polarization SAR image

ABSTRACT Monitoring marine oil pollution holds both practical and scientific significance. Synthetic Aperture Radar (SAR) is an active microwave remote sensing technique capable of all-weather and all-day with fine spatial resolution. However, under low wind conditions, rain cells and young ice are examples of look-alikes affect the accuracy of oil spill detection. Polarimetric SAR assumes a crucial role in this context, as it can extract abundant polarimetric features by polarimetric target decomposition. Drawing inspiration from this advancement, an improved polarimetric feature named relative feature based on Cloude-Pottier target decomposition was proposed. The Jeffries-Matusita distance indicates the substantial potential of the relative feature in detecting oil spills. The improved polarimetric feature within U-Net, FCN-8s, and DeepLabv3+ResNet-18 for oil spill detection using polarization SAR images. Experiment results demonstrated that the relative feature has superior performance compared to other polarimetric features and obtained the highest accuracy and dice within U-Net compared to the other two models. These findings introduce promising concepts for achieving rapid and precise detection of oil spills in future applications.

SAR marine oil spill detection based on an encoder-decoder network

ABSTRACT Marine oil spills are currently among the most challenging problems in marine environments. Synthetic Aperture Radar (SAR), with all-day and all-weather observation, has demonstrated great potential in oil disaster monitoring. A simple encoder-decoder network model was proposed by using SAR images for effective oil spill detection. The proposed model incorporates an optimized U-Net architecture that reduces computational requirements while maintaining detection performance. This is achieved through shortened encoder and decoder stages, depthwise separable convolutions, group normalization, and bilinear interpolation-based upsampling. To improve the model’s generalization, the auto-learning rate and focal loss function are also included. Two public SAR datasets acquired by different sensors have been used to illustrate the efficiency of the proposed model. In addition, the polarimetric information has also been assessed for the detection of oil spill monitoring. The results show that the model can achieve high efficiency at a small model size. The sub-dataset with polarimetric features also achieves a high F1-score of 91.65% and an IoU of 84.59%.

Offshore Oil Spill Detection Based on CNN, DBSCAN, and Hyperspectral Imaging.

Offshore oil spills have the potential to inflict substantial ecological damage, underscoring the critical importance of timely offshore oil spill detection and remediation. At present, offshore oil spill detection typically combines hyperspectral imaging with deep learning techniques. While these methodologies have made significant advancements, they prove inadequate in scenarios requiring real-time detection due to limited model detection speeds. To address this challenge, a method for detecting oil spill areas is introduced, combining convolutional neural networks (CNNs) with the DBSCAN clustering algorithm. This method aims to enhance the efficiency of oil spill area detection in real-time scenarios, providing a potential solution to the limitations posed by the intricate structures of existing models. The proposed method includes a pre-feature selection process applied to the spectral data, followed by pixel classification using a convolutional neural network (CNN) model. Subsequently, the DBSCAN algorithm is employed to segment oil spill areas from the classification results. To validate our proposed method, we simulate an offshore oil spill environment in the laboratory, utilizing a hyperspectral sensing device to collect data and create a dataset. We then compare our method with three other models-DRSNet, CNN-Visual Transformer, and GCN-conducting a comprehensive analysis to evaluate the advantages and limitations of each model.

Oil Spill Detection and Classification from Airborne EOIR Images Using a Deep Learning Model

Oil Spill Detection and Classification from Airborne EOIR Images Using a Deep Learning Model

Conditional Generative Adversarial Networks for SAR-based Ocean Feature Monitoring

Ocean surface monitoring, detection of oil spills, and accurate maritime vehicle detection is an important set of problem statements associated with ocean pollution monitoring. This paper proposes a novel study on Pix2Pix or generative adversarial networks for the aforementioned task. A comparative analysis is presented on various convolutional network-based encoders for semantic segmentation. A publicly available Synthetic Aperture Radar (SAR) dataset from the Sentinel-1 C-band sensor is used, constituting a total of 1112 images for the task of multiclass segmentation. Each integrant image pixel is classified into five classes, Oil Spills, Ships, Look-Alikes, Land Surface, and Ocean Surface. Deep architectures like U-Nets, SegNets, LinkNets, and FCNs are thoroughly assessed to obtain a practical analysis of this approach. Through a systematic exploration of hyperparameters, we identified the best-performing encoder configuration, achieving impressive results with a weighted Mean IoU of 0.898 and an F1-score of 0.931. These findings hold great promise for the advancement of solutions in ocean pollution monitoring, contributing to more accurate and efficient detection and mitigation efforts. respectively.

Compositional Oil Spill Detection Based on Object Detector and Adapted Segment Anything Model From SAR Images

Compositional Oil Spill Detection Based on Object Detector and Adapted Segment Anything Model From SAR Images

Spectral Indices Survey for Oil Spill Detection in Coastal Areas

Spectral Indices Survey for Oil Spill Detection in Coastal Areas

Geospatial Analysis of Artisanal Refineries Distribution Pattern in the Central Niger Delta, Nigeria

The study examined the geospatial analysis of geospatial analysis of artisanal refineries distribution pattern in the Central Niger Delta, Nigeria. The study made use of locational data obtained as secondary data from National Oil Spill Monitor and National Oil Spill Detection and Response Agency. Geographic analyses such as nearest neighbour analysis, directional distribution (Standard deviation ellipse), and Mean Center were carried out. Descriptive statistics involving spatial extent, frequency and percentage were employed for the data analysis. The results showed that that a total of 1476 artisanal sites were discovered in the Central Niger Delta. Of this total 55.35% (817) were found in Bayelsa State, 5.49% (81) were found in Delta State and 39.16% (578) were found in Rivers State. The nearest neighbour analysis showed that the distribution pattern of artisanal refineries was significantly clustered in Bayelsa State (Z= 47.5252, p=0.000); Delta State (Z= -10.129, p=0.0000); Rivers State (Z= -35.13333, p=0.0000) and the entire Central Niger Delta (Z= -59.8222, p=0.0000). The mean center of the distribution of artisanal refineries in Bayelsa State was found in Southern Ijaw LGA, Delta State in Uvwie LGA, Rivers State is found in Ikwerre LGA, and the entire Central Niger Delta was found in Ogbia LGA. The directional distribution standard deviation ellipse covered 58.49% in Bayelsa State, 35.06% in Rivers State and 6.35% of dispersion in Delta State. The study can be concluded that the spatial distribution pattern of artisanal refineries in the entire Central Niger Delta was clustered in nature and Bayelsa and Rivers States were more affected. It is recommended to among others that more studies should be done on the proliferation of artisanal refineries in the study area and entire Niger Delta so that the environmental effects resulting from artisanal refinery activities are reduced to the barest minimum.

DATE on the Edge: On-Site Oil Spill Detection and Thickness Estimation using Drone-based Radar Backscattering

DATE on the Edge: On-Site Oil Spill Detection and Thickness Estimation using Drone-based Radar Backscattering

A Semi-Empirical Threshold Model for Oil Spill Detection by Analyzing Microwave Backscatter of Ocean Surface From Sentinel-1 C-Band SAR Data

A Semi-Empirical Threshold Model for Oil Spill Detection by Analyzing Microwave Backscatter of Ocean Surface From Sentinel-1 C-Band SAR Data

Mineral oil emulsion species and concentration prediction using multi-output neural network based on fluorescence spectra in the solar-blind UV band.

The accurate monitoring of oil spills is crucial for effective oil spill recovery, volume determination, and cleanup. Oil slicks become emulsified under the effects of wind and waves, which increases the consistency of the oil spills. This phenomenon makes oil spills more challenging to handle and exacerbates environmental pollution. In this study, the variation of the solar-blind ultraviolet (UV) fluorescence spectra obtained from simulated oil spills with different oil types and oil-water ratios was investigated. By designing and constructing a multi-angle excitation and detection system, an apparent fluorescence peak of the oil emulsions was observed at around 290 nm under 220 nm excitation. By utilizing competitive adaptive reweighted sampling (CARS) and multi-output neural network algorithms, both the types and concentrations of the emulsified oils were obtained simultaneously. The classification accuracy for identifying the oil type exceeds 98%, and the mean absolute percentage error (MAPE) for concentration regression is around 2%. The results indicate that active solar-blind UV fluorescence could become a supplementary method for on-site oil spill detection to achieve comprehensive monitoring of oil spills. This study provides potential applications for UV-induced fluorescence spectrometry in oil spill on-site monitoring during the daytime.

Mauritius oil spill detection using transfer learning approach for oil spill mapping and wind impact analysis using Sentinel-1 data

Mauritius oil spill detection using transfer learning approach for oil spill mapping and wind impact analysis using Sentinel-1 data

Ecological risk assessment of a coastal area using multi-source remote sensing images and in-situ sample data

Healthy coastal area have high ecological service value. However, because of the uncontrolled utilization of marine resources and the discharge of pollutants, the ecological environments of many coastal areas are faced with ecological risks, such as oil spills and Enteromorpha blooms. Therefore, it is of great significance to monitor and assess the ecological risks in coastal areas. The remote sensing technique can be used to obtain frequent and wide-coverage images of the areas of interest, and has become an effective tool to monitor the ecological risks occurring in coastal areas. In this study, taking Jiaozhou Bay in China as the study area, an innovative ecological risk assessment framework was established using multi-source remote sensing images and in-situ sample data, which fully considers the potential harm of oil spills and Enteromorpha blooms on the ecology of the bay. Specifically, the oil spill risk for Jiaozhou Bay from 2017 to 2019 was identified by a deep learning based method from synthetic aperture radar (SAR) images, and the Enteromorpha bloom risk for the corresponding years was identified by a floating algae index (FAI) threshold method from multi-source optical images. Then, using habitat information and ground observation data about the marine species in the bay, environmental vulnerability maps were generated. Finally, a disaster-causing factors and disaster-bearing factors weighted risk assessment model was developed. The risk source monitoring results show that the occurrence frequency of oil spills in Jiaozhou Bay decreased from 2017 to 2019, while the outbreak intensity of Enteromorpha blooms generally became higher. The ecological risk assessment results reveal that the high risk areas of Jiaozhou Bay are mainly around the transportation regions and the ports, with a higher risk degree in the eastern and southern parts. To support our work, the well-trained oil spill detection network model and some information about the oil spill events that have occurred in Jiaozhou Bay are provided in this paper.