Paper Info
Title
Enhanced Delaunay Triangulation Sea Ice Tracking Algorithm with Combining Feature Tracking and Pattern Matching
Abstract
Sea ice drift detection has the key role of global climate analysis and waterway planning. The ability to detect sea ice drift in real-time also contributes to the safe navigation of ships and the prevention of offshore oil platform accidents. In this paper, an Enhanced Delaunay Triangulation (EDT) algorithm for sea ice tracking was proposed for dual-polarization sequential Synthetic Aperture Radar (SAR) images, which was implemented by combining feature tracking with pattern matching based on integrating HH and HV polarization feature information. A sea ice retrieval algorithm for feature detection, matching, fusion, and outlier detection was specifically developed to increase the system's accuracy and robustness. In comparison with several state-of-the-art sea ice drift retrieval algorithms, including Speeded Up Robust Features (SURF) and the Oriented FAST and Rotated BRIEF (ORB) method, the results of the experiment provided compelling evidence that our algorithm had a higher accuracy than the SURF and ORB method. Furthermore, the results of our method were compared with the drift vector and direction of buoys data. The drift direction is consistent with buoys, and the velocity deviation was about 10 m. It was proved that this method can be applied effectively to the retrieval of sea ice drift.
Year
DOI
Venue
2020
10.3390/rs12030581
REMOTE SENSING
Keywords
DocType
Volume
Delaunay Triangulation,dual-polarization,feature tracking,pattern matching,sea ice tracking,Sentinel-1,Synthetic Aperture Radar (SAR)
Journal
12
Issue
Citations 
PageRank 
3
0
0.34
References 
Authors
0
6
Name
Order
Citations
PageRank
Ming Zhang100.68
Jubai An28510.07
Jie Zhang312.78
Dahua Yu400.34
Junkai Wang500.34
Xiaoqi Lv600.34