SAR Oil Spill Detection System through Random Forest Classifiers

Conceição, Marcos Reinan A; Mendonça, Luís Felipe Ferreira de; Lentini, Carlos A. D.; Lima, A. T. da Cunha; Lopes, José Marques; Vasconcelos, Rodrigo Nogueira de; Gouveia, Mainara Biazati; Porsani, Milton J.

doi:10.3390/rs13112044

Cited by 26 publications

(6 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, the secondary statistics are obtained based on GLCM. Haralick et al (Conceio et al, 2021) defined 14 characteristic parameters of GLCM for texture analysis: second-order moment, contrast, correlation, difference moment, inverse difference moment, sum average, sum variance, sum entropy, difference variance, difference entropy, correlation measure 1, correlation measure 2, and maximum correlation coefficient. Researchers found that among the 14 texture features based on GLCM, four feature parameters are relevant.…”

Section: Feature Extraction (Glcm)mentioning

confidence: 99%

Detection of oil spills based on gray level co-occurrence matrix and support vector machine

Yu²,

Xu³

et al. 2022

Front. Environ. Sci.

View full text Add to dashboard Cite

Accurate reconnaissance of Marine oil spill is very important for emergency management of Marine oil spill accidents. Unmanned aerial vehicles (UAV) is a suitable carrier for offshore oil spill reconnaissance because of its fast deployment speed and low cost. Aiming at the identification accuracy of small oil spill accident in offshore port area and the problem of day and night reconnaissance, this study takes thermal infrared remote sensing images of oil leakage captured by UAV as the research object and proposes an oil spill detection method based on a Gray Level Co-occurrence Matrix (GLCM) and Support Vector Machine (SVM) method. Firstly, the extraction steps of image GLCM feature and the basic principle of SVM classification are studied. Then, the thermal infrared image data collected by UAV is preprocessed, including image filtering, clipping and rotation, and the sample database is generated. Subsequently, GLCM features of the samples were extracted, and the energy and correlation in GLCM were selected as classification features and sent to the SVM classifier to complete the oil spill detection of real-time thermal infrared images. The experimental results show that, compared with Classification and Regression Tree algorithm (CART) and Random Forests of Decision Trees (RF) algorithm, the detection accuracy of the method proposed in this paper reaches 95%, which were 10 and 2 percentage points higher than them respectively. The proposed method in this paper has fast recognition speed and high accuracy, and can provide all-weather recognition of oil spills for the detection of small oil spills in the offshore port area.

show abstract

Section: Feature Extraction (Glcm)mentioning

confidence: 99%

Detection of oil spills based on gray level co-occurrence matrix and support vector machine

Yu²,

Xu³

et al. 2022

Front. Environ. Sci.

View full text Add to dashboard Cite

show abstract

“…We also intend to develop a multiclass study based on q-EFE to differentiate between look-alikes and oil spills, investigating their respective patterns as in (Moura et al 2022;Huang et al 2022;Conceição et al 2021). Also, as a topic of ongoing research, we will evaluate its application in other contexts, e.g., health (radiological imagery to support diagnosis) and structural reliability (crack detection).…”

Section: Description Feature Vector Sizementioning

confidence: 99%

Automated detection of oil spills in images: combining a novel feature extraction technique based on the q- Exponential distribution with machine learning models

Negreiros

Lins

Maior

et al. 2022

Preprint

View full text Add to dashboard Cite

Oil spills are harmful, with negative environmental, social, and economic consequences. Generally, a risk-based framework involves preventing, detecting, and mitigating these undesirable events. Regarding detection, rapid oil spill identification is essential for mitigation, which fosters the use of automated procedures. Usually, automated oil spill detection involves radar images, computer vision, and machine learning techniques for classification. In this work, we propose a novel feature extraction method based on the q-Exponential probability distribution, named q-EFE. Such a model is suitable to account for atypical extreme pixel values, as it can have the power-law behavior. The q-EFE is combined with machine learning (ML) models, comprising a computer vision methodology to automatically classify images as “with oil spill” or “without oil spill”. We used a public dataset with 1112 Synthetic Aperture Radar (SAR) images to validate our methodology. Considering the proposed q-Exponential-based feature extraction, the SVM and XGB models outperformed deep learning models, including a ResNet50 one, and LBP and GLCM techniques for the biggest dataset size. The obtained results suggest that the proposed q-EFE can extract complex features from SAR images. Combined with ML models, it can perform image classification with satisfactory balanced accuracy.

show abstract

“…The labeling inputs were oil spills, biological films, rain cells, low wind regions, clean sea surface, ships, and terrain. The second was a SAR image oil detector called the Radar Image Oil Spill Seeker (RIOSS) (Conceição et al, 2021), which classified oillike targets between Latitude 0 • -23 • S and Longitude 49 • W -24 • W, which corresponds to the affected area in 2019. We developed an optimized feature space to serve as input to such classification models, both in terms of variance and computational efficiency.…”

Section: Detectionmentioning

confidence: 99%

“…It is a process of modifying dataset images into a set of variations (Shorten and Khoshgoftaar, 2019) before training or testing an algorithm so that learning is maximized in each sample. After the training step, the application of the classification method followed the RIOSS algorithm (Conceição et al, 2021). An input SAR image was divided into blocks by employing a clustering routine.…”

Section: Detectionmentioning

confidence: 99%

Coastal Ocean Observing and Modeling Systems in Brazil: Initiatives and Future Perspectives

et al. 2021

View full text Add to dashboard Cite

Coastal ocean observing and modeling systems (coastal observatories), connected with regional and global ocean systems, improve the quality of information and forecasts for effective management of safe and sustainable maritime activities. The public availability of systematic and long-term information of the ocean is an engine for the Blue Economy, boosting economic growth, employment, and innovation. An overview of some Brazilian initiatives is presented in this paper, involving universities, private companies, federal and state agencies, covering institutions from south to north of Brazil. Although these initiatives focus mainly on ocean physics, integrated efforts can extend the scope to include biogeochemistry and marine biodiversity, helping to address interdisciplinary problems. Existing initiatives can be connected, and new ones fostered, to fill in the gaps of temporal and spatial coverage of ocean monitoring in the vast oceanic area under Brazilian jurisdiction (nationally referred to as the Blue Amazon, in reference to the similar richness of the Amazon Forest). The alignment among national and regional initiatives, as well as with international programs, can be promoted if coordinated by a national-level organization, maximizing the return of public investment and socioeconomic benefits. In light of international examples, possible future institutional arrangements are discussed, leveraging from existing national public policies and international cooperation that Brazil is taking part. The United Nations Decade of Ocean Science for Sustainable Development is a timely opportunity to encourage an institutional arrangement to support and articulate an integrated network of coastal observatories in Brazil.

show abstract

SAR Oil Spill Detection System through Random Forest Classifiers

Cited by 26 publications

References 48 publications

Detection of oil spills based on gray level co-occurrence matrix and support vector machine

Detection of oil spills based on gray level co-occurrence matrix and support vector machine

Automated detection of oil spills in images: combining a novel feature extraction technique based on the q- Exponential distribution with machine learning models

Coastal Ocean Observing and Modeling Systems in Brazil: Initiatives and Future Perspectives

Contact Info

Product

Resources

About