Scholarlink

<p>Sea ice drift (SID) provides favorable conditions to form openings (leads), developing the convergent and divergent zones and providing a safer navigation route in the polar sea ice regions. The present study demonstrates a unified approach using Sentinel-1 Synthetic Aperture Radar (SAR) data to estimate the SID. A comprehensive python-based module has been defined as a collective Sea Ice Drift Detection Algorithm (SIDDA), which integrates Scale-Invariant Feature Transform (SIFT), FLANN based matcher, Lowe’s ratio based thresholding, RANdom SAmple Consensus (RANSAC) and Inverse Distance Weighted (IDW) interpolation technique. We have observed a greater number of keypoints generated using SIFT in comparison with the&#xa0;Oriented FAST and Rotated BRIEF (ORB) algorithms. Moreover, SID estimation using HH polarization outperforms HV polarization. Exclusivity of individual algorithms makes SIDDA robust in terms of entirely automatizing SID vectors. SIDDA based derived products have been validated with International Arctic Buoy Program (IABP) data and found that the difference in drift magnitude and the deviation of drift direction are ~ 1%, which endorses high confidence. The proposed algorithm optimally performs with a&#xa0;maximum temporal resolution of ~ 1–2&#xa0;day/s corresponding to SAR image pair.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Development of a Novel Sea Ice Drift Detection Algorithm (SIDDA) Using SAR Data

  • 期刊论文

  • 发表日期：

    2025年1月20日

查看全文
  • Purvee Joshi,
  • Jay Das,
  • Madhukar Srigyan,
  • Shailendra Saini,
  • Sushil Kumar Singh,
  • Shital Shukla,
  • Sandip Oza

摘要

Sea ice drift (SID) provides favorable conditions to form openings (leads), developing the convergent and divergent zones and providing a safer navigation route in the polar sea ice regions. The present study demonstrates a unified approach using Sentinel-1 Synthetic Aperture Radar (SAR) data to estimate the SID. A comprehensive python-based module has been defined as a collective Sea Ice Drift Detection Algorithm (SIDDA), which integrates Scale-Invariant Feature Transform (SIFT), FLANN based matcher, Lowe’s ratio based thresholding, RANdom SAmple Consensus (RANSAC) and Inverse Distance Weighted (IDW) interpolation technique. We have observed a greater number of keypoints generated using SIFT in comparison with the Oriented FAST and Rotated BRIEF (ORB) algorithms. Moreover, SID estimation using HH polarization outperforms HV polarization. Exclusivity of individual algorithms makes SIDDA robust in terms of entirely automatizing SID vectors. SIDDA based derived products have been validated with International Arctic Buoy Program (IABP) data and found that the difference in drift magnitude and the deviation of drift direction are ~ 1%, which endorses high confidence. The proposed algorithm optimally performs with a maximum temporal resolution of ~ 1–2 day/s corresponding to SAR image pair.