Please use this identifier to cite or link to this item:
http://dspace.azjhpc.org/xmlui/handle/123456789/413
Title: | THE LOCALIZATION OF OIL LEAKS IN THE SEA USING SATELLITE AND DRONE IMAGES WITH ARTIFICIAL INTELLIGENCE MODELS |
Authors: | Abbasov, V. M. Azizov, R. E. Aghamaliyev, Z. Z. Aydinsoy, E. A. Alimadatli, N. Y. |
Keywords: | Oil;Image Localization Models;PyTorch;YOLOv8;Artificial Intelligence;Oil Leak |
Issue Date: | 25-Apr-2024 |
Publisher: | Azerbaijan State Oil and Industry University |
Abstract: | Computer Vision, Deep Learning, and Machine Learning Algorithms make it possible to detect various dynamic issues in nature. Tankers, oil fields, oil pipelines, and hydrocarbon leaks and spills create serious problems for the sea ecosystems. [1] Utilizing this type of model can help detect oil leaks promptly, guide scientists’ predictions, compile cleaning plans, make urgent decisions on time, and stop or reduce the negative impacts of those incidents. Numerous recent scientific studies have been taken on this issue [2-7]. Illegal Pollution requires continuous monitoring and remote tracking technique employing satellites is an intriguing solution for the detection of oil leaks [8]. In this article, the solution to this problem is provided with the help of a recently updated model [9]. Specifically, emphasize the automatic approach of differentiation of oil marks and other similar marks. |
URI: | http://dspace.azjhpc.org/xmlui/handle/123456789/413 |
ISSN: | 1609-1620 |
Journal Title: | PROCEEDINGS OF AZERBAIJAN HIGH TECHNICAL EDUCATIONAL INSTITUTIONS |
metadata.dc.source.booktitle: | VOLUME 26 SPECIAL ISSUE |
Volume: | 2 |
Issue: | 148 |
First page number: | 421 |
Last page number: | 431 |
Number of pages: | 11 |
Appears in Collections: | Modern Problems of Macromolecular Compound Technology 2024 |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
Modern Problems of Macromolecular Compound Technology 2024-421-431.pdf | 1.04 MB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.