Stripe segmentation of oceanic internal waves in SAR images based on SegNet

The development of ocean remote sensing makes it possible to obtain valuable information from a large amount of data. Deep learning is a powerful tool that is beneficial for obtaining ocean information from remote sensing data. Oceanic internal waves play an essential role in ocean activities. To ob...

Full description

Saved in:
Bibliographic Details
Published inGeocarto international Vol. 37; no. 25; pp. 8567 - 8578
Main Authors Zheng, Ying-gang, Zhang, Hong-sheng, Qi, Kai-tuo, Ding, Long-yu
Format Journal Article
LanguageEnglish
Published Taylor & Francis 13.12.2022
Subjects
Deep learning
oceanic internal waves
SAR
SegNet
stripe segmentation
Online AccessGet full text

Cover

More Information
Summary:The development of ocean remote sensing makes it possible to obtain valuable information from a large amount of data. Deep learning is a powerful tool that is beneficial for obtaining ocean information from remote sensing data. Oceanic internal waves play an essential role in ocean activities. To obtain information on irregular stripes from Synthetic Aperture Radar (SAR) images, a stripe segmentation algorithm for oceanic internal waves is proposed based on SegNet. The research results show that the proposed method can identify whether the SAR images contain oceanic internal waves and obtain the respective locations of the light and dark stripes of the oceanic internal waves from SAR images. Furthermore, because this method can accurately determine the relative locations of the light and dark stripes, it can distinguish the moment when oceanic internal waves undergo polarity conversion.
ISSN:1010-6049
1752-0762
DOI:10.1080/10106049.2021.2002430