Dr A Zhang, Dr H Zhou
No more applications being accepted
Competition Funded PhD Project (Students Worldwide)
About the Project
Oil spills or leaks are the release of a liquid petroleum hydrocarbon into the environment, especially the marine ecosystem. It has been estimated that approximately 706 million gallons of oil enter the ocean every year (See-The-Sea.org). Of this amount of waste oil, over half comes from land drainage and waste disposal, offshore drilling operations contribute 2.1%, and transportation accidents account for 5.2%.
Evidence shows that a small scale (a few thousand gallons) of oil spills or leaks may still present enormous harm to deep sea and coastal fishing and fisheries. Short-term effects of toxic and smothering oil waste include mass mortality and contamination of sea species, and long-term ecological consequences may result in marine organic substrate, interruption of the food chain and collapse of fishing enterprises.
To effectively prevent disastrous consequences, early detection of oil spills or leaks must be achieved before cleaning up and remediating of the environment can be efficiently launched. For this purpose, in this project, our aim is to develop novel machine learning and artificial intelligence (AI) algorithms to identify and locate oil spill regions in synthetic aperture radar (SAR) images.
This is a multi-disciplinary research project, pulling together expertise in Mathematics, Informatics, Environmental Science and Geography at Leicester. The following specific objectives form the core activities of the proposed research programme:
1. To develop a new image segmentation scheme for properly outlining potential oil spill areas in the image.
2. To develop a new image classification strategy for separating real oil spill areas from the cluttered background.
3. To experimentally demonstrate that the proposed software tool at a system-level can produce better performance (efficiency and accuracy) than current state-of-the-art techniques.
The expected outcome of the work is a novel automated system of identifying oil spills or leaks on the sea. In regard to the dissemination of our work, in the short term, the student to be recruited will aim to publish at least three papers in top journals, such as IEEE Trans. The student will present the research findings at relevant leading conferences such as IEEE Computer Vision, GISRUK annual conferences, IFoA’s relevant AI conferences and Pattern Recognition, and publicise them via press releases and social media (including LinkedIn) In the longer term, the student will pursue possible commercialisation of the developed system.
Funding Notes
• A full UK/EU fee waiver for 3 years
• An annual tax free stipend of £14,777 (2018/19)
• A Research Training Support Grant to support project costs, fieldwork and conferences where applicable.
Studentships are open to UK Home / EU applicants and partial funding is available for international applicants
References
1. G.-S. Xia, G. Liu, W. Yang, and L. Zhang, “Meaningful object segmentation from SAR images via a multiscale nonlocal active contour model,” IEEE Transactions on Geoscience and Remote Sensing, vol. 54, no. 3, pp. 1860–1873, 2016.
2. L. W. Mdakane and W. Kleynhans, “An image-segmentation-based framework to detect oil slicks from moving vessels in the southern african oceans using SAR imagery,” IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 10, no. 36, pp. 2810–2818, 2017.
3. S. Ren, K. He, R. Girshick and J. Sun, “Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 39, no. 6, pp. 1137-1149, 2016
Continue with Facebook
