Abstract. Sea Ice Drift (SID) is an important parameter in understanding the Arctic climate dynamics and in maintaining navigation safety for the Arctic waters. SID is typically derived from keypoints extracted and matched from Synthetic Aperture Radar (SAR) imagery and is represented as a grid-based field. However, in feature tracking, the widely used Oriented FAST and Rotated BRIEF (ORB) is inherently vulnerable to low contrast, speckle noise, and complex sea ice deformation in feature tracking. To address these limitations, we propose an SID estimation framework that replaces the conventional ORB with deep learning-based methods such as SuperGlue and Local Feature TRansformer (LoFTR). In addition, multi-polarization is applied to exploit complementary information across both the feature tracking and pattern matching stages. Under polarization integration, SuperGlue reduced speed and directional RMSE by 50.8% and 37.3%, respectively, compared to ORB, while LoFTR achieved the best performance with reductions of 70.0% and 41.0%. These results demonstrate that deep learning-based methods can effectively replace the conventional ORB approach for SID estimation in the Arctic environments.
Ki-Yeong Mun, Junhwa Chi· The International Archives o...· 0 citations
Abstract. The rapid decline of Arctic sea ice has increased spatiotemporal variability in the core marginal seas along the Northern Sea Route (NSR), thereby hindering reliable short-term forecasting. While deep learning models offer computationally efficient alternatives to physics-based numerical models, previous studies have often relied on global average errors and have provided limited assessment of boundary-preservation performance. In addition, few studies have systematically compared spatial feature extraction strategies within non-recurrent spatiotemporal architectures for sea ice forecasting. To address this gap, this study benchmarks CNN backbone and Transformer backbone families for 10-day sea ice forecasting. The evaluation focuses on five dynamic marginal seas along the NSR. Metrics included Integrated Ice Edge Error (IIEE), Mean Boundary Error (MBE), Intersection over Union (IoU), and Anomaly Correlation Coefficient (ACC). Results indicate that CNN-based models, including PoolFormer, generally show more favorable performance in boundary preservation and prediction stability than Transformer-based models. Regionally, the Barents and Kara Seas were more difficult to predict, whereas the Laptev and East Siberian Seas were relatively more predictable. The Chukchi Sea exhibited particularly high uncertainty during rapid summer ice retreat. Across all models, boundary-based performance degraded significantly as lead time increased, with MBE exceeding 30 km from T+5 onward. These results suggest that, under the current univariate setting, boundary-preservation performance remains relatively stable at short lead times, but degrades thereafter. Overall, CNN-based non-recurrent architectures show relative strengths over the NSR test regions, although future work incorporating multivariate inputs is needed to extend reliable lead times.
Bo-Ram Kim, Junhwa Chi· The International Archives o...· 0 citations
Abstract. Satellite remote sensing is inherently constrained by a trade-off between spatial and temporal resolution. As a result, high-temporal-frequency sensors such as Geostationary Ocean Color Imager-II provide operationally valuable observations but at coarse spatial resolution. Reference-Based Super-Resolution (Ref-SR) can address this limitation by transferring high-resolution textures from an external reference image, but temporal mismatch between the target and reference images often leads to unreliable texture transfer and severe artifacts. This problem becomes more critical in extreme low-resolution (LR) settings, where structural information is already severely degraded. To address this issue, we propose the Dynamic Ref-SR Framework, which computes a pixel-wise weight map from intensity differences between the LR and reference images to selectively control reference transfer. The resulting weights promote reference use in stable regions while suppressing it in temporally inconsistent regions. The framework was validated on three backbone architectures—CNN (EDSR), Swin Transformer, and GAN—using a Sentinel-2 dataset for four-band reconstruction (RGB and NIR). Across all metrics and architectures, the proposed Ref-SR framework consistently outperformed the SISR baseline in both structural and spectral evaluations. Among the tested backbones, the GAN-based model achieved the best overall performance, with a PSNR of 35.60 dB, an SSIM of 0.92, a SAM of 2.20°, and an ERGAS of 74.71. These results demonstrate that the proposed framework can improve LR satellite imagery while reducing the risk of reference misuse under temporal mismatch.
Chae-Eun Kim, Junhwa Chi· The International Archives o...· 0 citations
Abstract. We propose a systematic stepwise optimization pipeline for building change detection in dense urban environments using high-resolution CAS500-1 satellite imagery. To support robust model development, we constructed a dataset comprising 3,816 bi-temporal patch pairs across 28 urban regions. The framework employs a Mamba-based architecture as the baseline, leveraging its efficient global context modeling capability for binary change detection. The pipeline integrates three sequential optimization stages to enhance detection accuracy and stability. First, we evaluated normalization techniques tailored for 12-bit radiometric resolution, comparing percentile-based scaling, gamma correction, and log transformations. Second, we implemented an augmentation strategy that extends standard geometric transformations with optical and temporal methods to improve generalization in structurally complex urban settings. Third, we explored various ensemble configurations, including confidence-weighted and hierarchical aggregation to mitigate individual model scale limitations. Performance was validated through multi-faceted evaluation metrics covering pixel-level, contour-based, and object-based metrics. Experimental results demonstrate that gamma-based normalization, comprehensive augmentation, and hierarchical ensemble consistently outperform baseline configurations across multiple evaluation metrics. The final optimized pipeline achieved an F1-Score of 0.8070, making a significant improvement over the 0.7629 baseline. This work provides an extensible framework for operational satellite-based change detection and establishes a practical foundation for future ensemble-based architectures.
DongHyuk Jin, Junhwa Chi· The International Archives o...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
