29 Apr 2024
 | 29 Apr 2024
Status: this preprint is open for discussion.

Combining the U-Net model and a Multi-textRG algorithm for fine SAR ice-water classification

Yan Sun, Shaoyin Wang, Xiao Cheng, Teng Li, Chong Liu, Yufang Ye, and Xi Zhao

Abstract. Sea ice classification faces challenges due to the similarity among surfaces such as wind-driven open water (OW), smooth thin ice, level first-year ice (FYI), and melted ice surfaces. Previous algorithms combine unsupervised region segmentation and supervised neural networks, yet struggle due to limited manual labels and inaccurate region segmentation. We propose to adopt a supervised neural network followed by a region segmentation algorithm with experiential knowledge involved to solve the ambiguous recognition question and sample number limitation. Provided by the AI4Arctic competition, the preprocessed GCOM-W1 AMSR2 36.5GHz H polarization and Sentinel-1 SAR EW dual-polarization data, the CIS/DMI ice chart labels, and the pre-trained U-Net CNN model are employed to perform semantic segmentation of ice and water with near-100 % accuracy. Subsequently, within the U-Net semantically segmented ice mask, a multistage pixel-based ice detection algorithm developed on GLCM textures of SAR images and region growing approach, the Multi-textRG algorithm, refines the ice edge details. We validate the results on Landsat-8 and Sentinel-2 optical data yielding an overall accuracy of 83.11 %, low false negative (FN) of 4.03 % indicating underestimated low backscatter ice surfaces and higher false positive (FP) of 12.86 % reflecting their resolution difference along edges. More importantly, we fused the SAR-based ice detection with CIS/DMI ice charts and AMSR2 ASI SIC product obtaining SAR-Chart and SAR-AMSR2 labels, which enhance ice edge depictions and SIC variation contours. Repeating the two-step procedure with the high-precision SIC labels demonstrates the U-Net model's capability to extract detailed ice edges information and stability of the Multi-textRG algorithm. The U-Net model trained on SAR-AMSR2 label achieves the highest R2-score of 91.993 %, the largest OWrecall (recall of OW) of 99.268 %, and large ov40recall (recall of ice with over-40 % SIC) of 99.207 %. Our algorithm framework solves the accurate ice-water classification at all seasons and facilitates the sample labelling for improving SIC estimation accuracy based on CNN models.

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this preprint. The responsibility to include appropriate place names lies with the authors.
Yan Sun, Shaoyin Wang, Xiao Cheng, Teng Li, Chong Liu, Yufang Ye, and Xi Zhao

Status: open (until 12 Jun 2024)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
Yan Sun, Shaoyin Wang, Xiao Cheng, Teng Li, Chong Liu, Yufang Ye, and Xi Zhao
Yan Sun, Shaoyin Wang, Xiao Cheng, Teng Li, Chong Liu, Yufang Ye, and Xi Zhao


Total article views: 170 (including HTML, PDF, and XML)
HTML PDF XML Total Supplement BibTeX EndNote
127 38 5 170 75 3 3
  • HTML: 127
  • PDF: 38
  • XML: 5
  • Total: 170
  • Supplement: 75
  • BibTeX: 3
  • EndNote: 3
Views and downloads (calculated since 29 Apr 2024)
Cumulative views and downloads (calculated since 29 Apr 2024)

Viewed (geographical distribution)

Total article views: 178 (including HTML, PDF, and XML) Thereof 178 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
Latest update: 25 May 2024
Short summary
Arctic sea ice has rapidly declined due to global warming, leading to extreme weather events. Accurate ice monitoring is vital for understanding and forecasting these impacts. Combining SAR and AMSR2 data with machine learning is efficient but requires sufficient labels. We propose a framework integrating the U-Net model with the Multi-textRG algorithm to achieve ice-water classification at SAR-level resolution and to generate accurate labels for improved U-Net model training.