An efficient real time segmentation of oil spill in SAR image based on U-Net++

View at publisher Summarise with VIRAT
Authors: Nagarani Nagarajan, Sukha Dev Alagarsamy, Gokul Priyan Gujuluva Venkatesh, Sivanesan Rajendra Kumar, Pradeep Kumar Govinda Raj
Cited by: 0
References: 7

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

The threat of oil spills poses a significant risk to coastal and marine environments, emphasizing the necessity for efficient surveillance and prompt identification. Synthetic Aperture Radar (SAR) sensors are essential for detecting oil slicks in all weather conditions. This study proposes employing the U-Net++ architecture to enhance oil leakage identification in SAR pictures, with the goal of enhancing precision and accuracy. U-Net++ addresses the inherent semantic gap in the U-Net model by introducing nested skip pathways, allowing the decoder to access and incorporate features from the encoder. The research focuses on implementing U-Net++ on a labeled SAR image dataset to improve detection accuracy for oil spills. Through analysis, this study demonstrates the effectiveness of U-Net++ in enhancing segmentation outcomes and refining detection capabilities. In conclusion, the application of U-Net++ for oil leakage identification and segmentation in SAR images shows promise in advancing environmental monitoring and conservation efforts. This research contributes to the progression of the Identification of oil leakages and processing of SAR pictures, with the aim of establishing benchmarks for future studies.

Fields of Study & Subjects

Oil Spill Detection and Mitigation Remote-Sensing Image Classification Advanced SAR Imaging Techniques

Publication details

Journal / source
AIP conference proceedings
Volume
3341
Pages
020074–020074
Publication date
1 January 2026
Language
EN