English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Ocean surface monitoring, especially oil slick detection, has becomemandatory due to its importance for oil exploration and risk prevention onecosystems. For years, the detection task has been performed manually byphoto-interpreters using Synthetic Aperture Radar (SAR) images with the help ofcontextual data such as wind. This tedious manual work cannot handle theincreasing amount of data collected by the available sensors and thus requiresautomation. Literature reports conventional and semi-automated detectionmethods that generally focus either on oil slicks originating fromanthropogenic (spills) or natural (seeps) sources on limited data collections.As an extension, this paper presents the automation of offshore oil slicks onan extensive database with both kinds of slicks. It builds upon the slickannotations of specialized photo-interpreters on Sentinel-1 SAR data for 4years over 3 exploration and monitoring areas worldwide. All the considered SARimages and related annotation relate to real oil slick monitoring scenarios.Further, wind estimation is systematically computed to enrich the datacollection. Paper contributions are the following : (i) a performancecomparison of two deep learning approaches: semantic segmentation usingFC-DenseNet and instance segmentation using Mask-RCNN. (ii) the introduction ofmeteorological information (wind speed) is deemed valuable for oil slickdetection in the performance evaluation. The main results of this study showthe effectiveness of slick detection by deep learning approaches, in particularFC-DenseNet, which captures more than 92% of oil instances in our test set.Furthermore, a strong correlation between model performances and contextualinformation such as slick size and wind speed is demonstrated in theperformance evaluation. This work opens perspectives to design models that canfuse SAR and wind information to reduce the false alarm rate.

Deep learning based automatic detection of offshore oil slicks using SAR data and contextual information