Oil Spill Detection Research Articles

Oil spill detection over ocean surface using deep learning: a comparative study

Oil spill detection over ocean surface using deep learning: a comparative study

Requirements for radar navigation aids for operational safety of autonomous navigation of facilities of the Arctic oil and gas complex

An analysis of the requirements for radar systems for autonomous sea vessels operating in the Arctic regions shows that to ensure the safety of transport and oil and gas facilities, it is necessary to evolution to radar systems with characteristics that significantly exceed the existing classification requirements. The application of ultra-wideband and ultra-short-pulse radars, including those based on the principles of microwave photonics, which will provide a radio-imaging mode and decimetre levels of accuracy when measuring coordinates, as well as observation of small objects, ice conditions and detection of oil spills, is proposed. The use of such radar facilities to ensure high-precision navigation will make it possible to start solving the problem of creating autonomous vessels, provided that the safety of logistics operations in the oil and gas complex is ensured.

Temporal Distribution and Characteristic Analysis of Oil Spill in Balikpapan Bay

Balikpapan Bay has an oil spill case that attracts many people because it is considered a serious environmental problem and is detrimental to the environment. One of the cases that occurred was the leak of an oil pipeline in Balikpapan Bay due to the wrong anchorage of a ship that occurred on March 31, 2018. Detection of oil spills for three months using Sentinel 1-A satellite data to determine the distribution and analysis of the same oil characteristics from the source of pipe leaks in the Gulf of Balikpapan. The multi-temporal distribution of oil spills in Balikpapan Bay in March, April, and May 2018 has a significant difference in the upstream and mouth of the bay due to a pipe leak on March 31, 2018. Characteristics of upstream oil spills represented by stations 4 and 5 have the anisotropy value is lower than at the mouth of the bay which is represented by stations 1 and 2. The characteristics of the oil spill in Balikpapan Bay have differences before and after the oil spill due to pipe leakage as indicated by the decrease in the anisotropy value.

Numerically modelling the reflectance of a rough surface covered with diesel fuel based on bidirectional reflectance distribution function

Oil spills have become a problem that negatively affects the oceanic environment and maritime transportation. Optical remote sensing technology is a potential method to monitor oil spills by analyzing the reflectance spectra of oil-polluted and clean water surface. In this paper, a numerical model for the reflectance of a rough oil surface is constructed by combining Fresnel reflection and bidirectional reflectance distribution function (BRDF). The way that visible light is reflected from the rough diesel fuel surface is quantitatively described and discussed based on the reflection theory of electromagnetic waves. The simulation result of the proposed model shows reasonable agreement with experimental measurements. With reliable prediction and a low computational complexity, the proposed model is expected to provide a theorical basis for rapid detection of oil spills on rough sea surfaces using optical remote sensing technology.

Chronic Oil Pollution from Vessels and Its Role in Background Pollution in the Southeastern Baltic Sea

The results of long-term satellite monitoring of oil pollution of the sea surface in the southeastern Baltic Sea (SEB) are discussed in this paper. From June 2004 to December 2020, in total, 2780 Synthetic Aperture Radar (SAR) images from different satellites were received and analyzed. There were 788 oil spills detected in the study area. The oil spills were concentrated along the main shipping routes in the SEB. The volume of the detected oil spills was estimated. The average size of the spill was about 2 km2 or 0.8 m3. Seasonal variability of oil pollution shows a decrease in the number of oil detections in the autumn–winter period, which is associated with the prevalence of unfavorable wind conditions that limit the use of SAR technology for oil spill detection and navigation for small ships. In situ measurements show that seasonal variation in the concentration of oil products in seawater is characterized by a maximum in April and a minimum in July. Since 2007, a decrease in oil detections has been observed for the entire Baltic Sea, including the study area. The interannual variability also shows a decrease in the concentration of oil products in the water column. In the southeastern Baltic Sea, the volume of oil products released yearly to the sea surface from ships does not exceed 0.1% of the average instantaneous presence of oil products in the water column.

Study of the efficiency of data preprocessing in multispectral devices for detecting oil spills

The paper presents the results of study of various methods for preprocessing of images obtained using a multispectral system for detecting oil spills on the sea surface in areas of oil production and drilling platforms. Based on the comprehensive analysis of a large number of heuristic methods for processing and improving the quality of images obtained in different spectral ranges, a classification of these methods is proposed in relation to the problem being solved. The method of expert evaluation was used to reveal a feasible hardware of the spill detection system and determine the most effective methods for preprocessing of multispectral images.

A new oil spill detection algorithm based on Dempster-Shafer evidence theory

Features of oil spills and look-alikes in polarimetric synthetic aperture radar (SAR) images always play an important role in oil spill detection. Many oil spill detection algorithms have been implemented based on these features. Although environmental factors such as wind speed are important to distinguish oil spills and look-alikes, some oil spill detection algorithms do not consider the environmental factors. To distinguish oil spills and look-alikes more accurately based on environmental factors and image features, a new oil spill detection algorithm based on Dempster-Shafer evidence theory was proposed. The process of oil spill detection taking account of environmental factors was modeled using the subjective Bayesian model. The Faster-region convolutional neural networks (RCNN) model was used for oil spill detection based on the convolution features. The detection results of the two models were fused at decision level using Dempster-Shafer evidence theory. The establishment and test of the proposed algorithm were completed based on our oil spill and look-alike sample database that contains 1 798 image samples and environmental information records related to the image samples. The analysis and evaluation of the proposed algorithm shows a good ability to detect oil spills at a higher detection rate, with an identification rate greater than 75% and a false alarm rate lower than 19% from experiments. A total of 12 oil spill SAR images were collected for the validation and evaluation of the proposed algorithm. The evaluation result shows that the proposed algorithm has a good performance on detecting oil spills with an overall detection rate greater than 70%.

Fluorescence hyperspectral imaging system for analysis and visualization of oil sample composition and thickness.

In this paper, a compact fluorescence hyperspectral imaging system based on a prism-grating-prism (PGP) structure is designed. Its spectrometer spectral range is 400-1000 nm with a spectral resolution of 2.5 nm, and its weight is less than 1.7 kg. The PGP imaging spectrometer combines the technical advantages of prism and grating, by not only using six lenses for imaging and collimation to realize the dual telecentres of object and image but also having a "straight cylinder" structure, which makes the installation and adjustment simple, compact, and stable. By the push-broom method, we obtained the three-dimensional cubic data of different oil products. By normalization processing, minimum noise separation transformation processing, visualization processing, and support vector machine classification processing of different oil fluorescence hyperspectral data, we demonstrate that the fluorescence hyperspectral imaging system can identify different kinds of oil and recognize the oil film thickness. The fluorescence hyperspectral imaging system can be used in oil spill detection, resource exploration, natural disaster monitoring, environmental pollution assessment, and many other fields.

Influence of Dispersed Oil on the Remote Sensing Reflectance-Field Experiment in the Baltic Sea.

Remote sensing techniques currently used to detect oil spills have not yet demonstrated their applicability to dispersed forms of oil. However, oil droplets dispersed in seawater are known to modify the local optical properties and, consequently, the upwelling light flux. Theoretically possible, passive remote detection of oil droplets was never tested in the offshore conditions. This study presents a field experiment which demonstrates the capability of commercially available sensors to detect significant changes in the remote sensing reflectance Rrs of seawater polluted by six types of dispersed oils (two crude oils, cylinder lubricant, biodiesel, and two marine gear lubricants). The experiment was based on the comparison of the upwelling radiance Lu measured in a transparent tank floating in full immersion in seawater in the Southern Baltic Sea. The tank was first filled with natural seawater and then polluted by dispersed oils in five consecutive concentrations of 1–15 ppm. After addition of dispersed oils, spectra of Rrs noticeably increased and the maximal increase varied from 40% to over three-fold at the highest oil droplet concentration. Moreover, the most affected Rrs band ratios and band differences were analyzed and are discussed in the context of future construction of algorithms for dispersed oil detection.

Oil Spill Detection Using Satellite Imagery

<span>Since oil exploration began, oil spills have become a serious problem. When drilling for oil, there is always a risk of an oil spill. With the new development of technology over the years, oil spill detection has become much easier making the clean-up of a spill to happen much faster reducing the risk of a large spread. In this study, remote sensing techniques were used to detect the Deep-water Horizon oil spill through a change detection method. The change detection method allows the viewer to determine the difference of an area before and after an oil spill as well as detect the irregular difference on a surface. To confirm the effectiveness of change detection method, two approaches were used each showing the differences in the images before and after the spill allowing the size and shape to be identified. The swipe tool in the ArcGIS software was used to visually show the changes. The difference tool was also used to both visually and statistically to investigate the difference before and after the Deep-water Horizon oil spill event.</span>

Feature Merged Network for Oil Spill Detection Using SAR Images

The frequency of marine oil spills has increased in recent years. The growing exploitation of marine oil and continuous increase in marine crude oil transportation has caused tremendous damage to the marine ecological environment. Using synthetic aperture radar (SAR) images to monitor marine oil spills can help control the spread of oil spill pollution over time and reduce the economic losses and environmental pollution caused by such spills. However, it is a significant challenge to distinguish between oil-spilled areas and oil-spill-like in SAR images. Semantic segmentation models based on deep learning have been used in this field to address this issue. In addition, this study is dedicated to improving the accuracy of the U-Shape Network (UNet) model in identifying oil spill areas and oil-spill-like areas and alleviating the overfitting problem of the model; a feature merge network (FMNet) is proposed for image segmentation. The global features of SAR image, which are high-frequency component in the frequency domain and represents the boundary between categories, are obtained by a threshold segmentation method. This can weaken the impact of spot noise in SAR image. Then high-dimensional features are extracted from the threshold segmentation results using convolution operation. These features are superimposed with to the down sampling and combined with the high-dimensional features of original image. The proposed model obtains more features, which allows the model to make more accurate decisions. The overall accuracy of the proposed method increased by 1.82% and reached 61.90% compared with the UNet. The recognition accuracy of oil spill areas and oil-spill-like areas increased by approximately 3% and reached 56.33%. The method proposed in this paper not only improves the recognition accuracy of the original model, but also alleviates the overfitting problem of the original model and provides a more effective monitoring method for marine oil spill monitoring. More importantly, the proposed method provides a design principle that opens up new development ideas for the optimization of other deep learning network models.

Design of android based Unmanned Surface Vehicles (USV) for oil spill monitoring

In this research, a USV will be developed that can detect oil spills on the waters, so in the future it can overcome the problem of water pollution. To be able to determine the thickness of the oil on the water surface, TEMT6000 and LED sensors are used. The light intensity will be converted into ADC value by the TEMT6000 sensor. Furthermore, on this system USV will be controlled by Android and can monitor water surface conditions. Android applications will be connected to Wemos via wifi. Then Wemos is connected serially with a RF915 and another RF915 will be connected to the microcontroller of the USV. From the results of the USV test, the USV can be controlled via Android by moving forward, backward, turning left and right. The maximum range of RF915 is 493.3 meters. Then the oil spill detection average error is 5.45% and the the average of the success rate is 94.55%.

Monitoring offshore oil pollution using multi-class convolutional neural networks

Oil and gas production operations are a major source of environmental pollution that expose people and habitats in many coastal communities around the world to adverse health effects. Detecting oil spills in a timely and precise manner can help improve the oil spill response process and channel required resources more effectively to affected regions. In this research, convolutional neural networks, a branch of artificial intelligence (AI), are trained on a visual dataset of oil spills containing images from different altitudes and geographical locations. In particular, a VGG16 model is adopted through transfer learning for oil spill classification (i.e., detecting if there is oil spill in an image) with an accuracy of 92%. Next, Mask R–CNN and PSPNet models are used for oil spill segmentation (i.e., pixel-level detection of oil spill boundaries) with a mean intersection over union (IoU) of 49% and 68%, respectively. Lastly, to determine if there is an oil rig or vessel in the vicinity of a detected oil spill and provide a holistic view of the oil spill surroundings, a YOLOv3 model is trained and used, yielding a maximum mean average precision (mAP) of ~71%. Findings of this research can improve the current practices of oil pollution cleanup and predictive maintenance, ultimately leading to more resilient and healthy coastal communities.

Oil Spill Detection in SAR Images Using Online Extended Variational Learning of Dirichlet Process Mixtures of Gamma Distributions

In this paper, we propose a Dirichlet process (DP) mixture model of Gamma distributions, which is an extension of the finite Gamma mixture model to the infinite case. In particular, we propose a novel online nonparametric Bayesian analysis method based on the infinite Gamma mixture model where the determination of the number of clusters is bypassed via an infinite number of mixture components. The proposed model is learned via an online extended variational Bayesian inference approach in a flexible way where the priors of model’s parameters are selected appropriately and the posteriors are approximated effectively in a closed form. The online setting has the advantage to allow data instances to be treated in a sequential manner, which is more attractive than batch learning especially when dealing with massive and streaming data. We demonstrated the performance and merits of the proposed statistical framework with a challenging real-world application namely oil spill detection in synthetic aperture radar (SAR) images.

Oil Spill Detection with Multiscale Conditional Adversarial Networks with Small-Data Training

We investigate the problem of training an oil spill detection model with small data. Most existing machine-learning-based oil spill detection models rely heavily on big training data. However, big amounts of oil spill observation data are difficult to access in practice. To address this limitation, we developed a multiscale conditional adversarial network (MCAN) consisting of a series of adversarial networks at multiple scales. The adversarial network at each scale consists of a generator and a discriminator. The generator aims at producing an oil spill detection map as authentically as possible. The discriminator tries its best to distinguish the generated detection map from the reference data. The training procedure of MCAN commences at the coarsest scale and operates in a coarse-to-fine fashion. The multiscale architecture comprehensively captures both global and local oil spill characteristics, and the adversarial training enhances the model’s representational power via the generated data. These properties empower the MCAN with the capability of learning with small oil spill observation data. Empirical evaluations validate that our MCAN trained with four oil spill observation images accurately detects oil spills in new images.

Tracing illegal oil discharges from vessels using SAR and AIS in Bohai Sea of China

In China's 14th Five-Year Plan, it is necessary to coordinate the promotion of marine ecological protection and prevent environmental risks such as marine oil spills. The illegal discharge of oily waste from ships is an important source of oil spill pollution in the Bohai Sea, and has become an important factor hindering the purification of the marine environment. Because SAR remote sensing is not restricted by light conditions, it can work at night, in heavy fog, overcast and rainy days, and has become one of the important techniques for monitoring oil spills on the sea. AIS can provide real-time and historical position and track information of ships. The combination of SAR and AIS can provide a great help for maritime regulatory authorities to find the source of oil spill pollution. However, China has not yet established a mature and effective oil spill supervision system based on SAR and AIS. This paper proposes a framework for oil spill detection and accountability using the combination of SAR, AIS and boarding inspection, and takes an example of the monitoring and tracking process of illegal oil discharge from a ship in the Bohai Sea. As far as we know, this is the first time that China has used SAR and AIS to identify ships that illegally discharge oily sewage. This work framework can provide a reference for China's maritime agencies to improve the oil spill supervision system during the 14th Five-Year Plan.

A scattering similarity based calculation model for oil film identification

To identify oil spill at sea, a method of characterizing the scattering of the detected oil spill target is proposed, based on its polarization characteristics. Due to the unique feature that Mueller matrix is able to comprehensively describe polarization characteristics and has a close relation to the target itself, we propose to use a Mueller matrix polarimeter to calculate Mueller matrix. Combined with the conditions for physical realizability, the Mueller matrix is filtered to achieve a higher level of accuracy. Based on Pauli basis decomposition, the scattering similarity parameters (SSP) between the target and the typical scattering are studied. The formulas of SSP expressed by amplitude ratio and phase retardation in Fresnel reflection are derived, which help us discover the relationship among SSPs. A new concept of scattering similarity parameter difference (SSPD) is proposed and the relationship between SSPD and degree and polarization (DOP) is derived. The difference of SSPD curves between water and oil film is used to distinguish them, which can effectively detect oil spill.

Modelling of the optical signature of oil slicks at sea for the analysis of multi- and hyperspectral VNIR-SWIR images.

This work focuses on the contribution of modelling for the interpretation of multi- or hyperspectral optical images for the detection, characterisation and quantification of oil spills. Many parameters contribute to the spectral signature of an oil layer on the sea surface: the optical properties of the water column and of the oil, the film thickness, the surface roughness, the atmospheric radiance reaching the surface (direct and diffuse components), the geometry of observation and illumination. The number of these contributors and their combinations make the analysis of the spectral variability of oil signatures at the sea surface complex. Modelling approaches allow us to consider all those parameters and can then provide useful information to improve the interpretation of optical images. The model presented in this paper simulates the radiance of an oil layer from visible to short wave infrared spectral domains, taking into account all the above-mentioned parameters. The damping influence of the oil layer on sea surface waves is also considered. Comparisons of the simulations with in situ measurements shows a good overall agreement despite the lack of knowledge of some input parameters of the model. In combination with laboratory and in-the-field measurements, the model is then used to assess the expected contrast between water and oil and to estimate oil slick volume.

SAR Oil Spill Detection System through Random Forest Classifiers

A set of open-source routines capable of identifying possible oil-like spills based on two random forest classifiers were developed and tested with a Sentinel-1 SAR image dataset. The first random forest model is an ocean SAR image classifier where the labeling inputs were oil spills, biological films, rain cells, low wind regions, clean sea surface, ships, and terrain. The second one was a SAR image oil detector named “Radar Image Oil Spill Seeker (RIOSS)”, which classified oil-like targets. An optimized feature space to serve as input to such classification models, both in terms of variance and computational efficiency, was developed. It involved an extensive search from 42 image attribute definitions based on their correlations and classifier-based importance estimative. This number included statistics, shape, fractal geometry, texture, and gradient-based attributes. Mixed adaptive thresholding was performed to calculate some of the features studied, returning consistent dark spot segmentation results. The selected attributes were also related to the imaged phenomena’s physical aspects. This process helped us apply the attributes to a random forest, increasing our algorithm’s accuracy up to 90% and its ability to generate even more reliable results.

MULTIDIMENSIONAL STATISTICAL MODEL FOR DETECTING OIL POLLUTION SITES BASED ON SATELLITE IMAGERY

The article considers an algorithm for determining the statistical model from several inhomogeneous images of the Earth's surface obtained by different sensors (optoelectronic scanning device, synthetic aperture radar (SAR)) over the sea areas. The object of the study are the methods of remote sensing of the Earth used for detection and mapping of oil spills. The aim of the research was to perform testing for a possible variation of the statistical model inside a non-uniform sliding window based on a semi-automatic approach. The proposed algorithm makes it possible to determine the spatial extent of oil production sites and oil pollution in offshore waters using multi-time RSA data and a multi-zone combined image with a spatial resolution of 10 m. First, homogeneous regions are analyzed in the image, and then the model of the analysis zone is expanded to the more general case of inhomogeneous regions that are observed in the analysis windows.