CA-STIM: an interpolation model with spatio-temporal evolution characteristics and cross-attention mechanism for 2D island morphology sequences
文献类型：期刊论文
| 作者 | Zhang, Peng1,2; Wu, Wenzhou1,2; Shi, Shaochen1,3; Li, Fengyu1,2; Su, Fenzhen1,2 |
| 刊名 | INTERNATIONAL JOURNAL OF DIGITAL EARTH
 |
| 出版日期 | 2025-12-31 |
| 卷号 | 18期号:1页码:2513591 |
| 关键词 | Spatial morphology evolution of islands interpolation cross attention feature fusion deep learning |
| ISSN号 | 1753-8947 |
| DOI | 10.1080/17538947.2025.2513591 |
| 产权排序 | 1 |
| 文献子类 | Article |
| 英文摘要 | Coral islands are highly susceptible to climate change, making it crucial to understand their spatial morphological evolution for sustainable development and management. Recent advancements in high spatio-temporal resolution earth observation technologies have facilitated the analysis and prediction of island morphology over medium to short timescales. However, issues such as cloud cover and atmospheric interference often lead to poor image quality, resulting in significant missing in the 2D morphology sequences extracted from remote sensing images. To address this issue, we propose a spatio-temporal interpolation model (CA-STIM) that integrates both external environmental dynamics and the intrinsic spatio-temporal evolution characteristics of island morphology using a convolutional neural network-long short-term memory network (CNN-LSTM) framework with a cross-attention mechanism and a weighted binary cross-entropy loss function. The cross-attention mechanism incorporates external environmental factors, enhancing the interpolation accuracy, while the weighted binary cross-entropy loss function effectively addresses the challenge of directional heterogeneity. Using three coral islands in the South China Sea (Beizi, Mahuan, and Xiyue) as case studies, we perform 2D spatial morphology series interpolation. Experimental results demonstrate that our model outperforms baseline methods, achieving Dice scores of 0.9681, 0.9675, and 0.975 and Intersection-over-Union (IOU) scores of 0.9383, 0.9373, and 0.9513 on Beizi, Mahuan, and Xiyue Island, respectively. |
| URL标识 | 查看原文 |
| WOS关键词 | SPATIAL INTERPOLATION ; EARTH OBSERVATION ; FUSION |
| WOS研究方向 | Physical Geography ; Remote Sensing |
| 语种 | 英语 |
| WOS记录号 | WOS:001502972300001 |
| 出版者 | TAYLOR & FRANCIS LTD |
| 源URL | [http://ir.igsnrr.ac.cn/handle/311030/214564]  |
| 专题 | 资源与环境信息系统国家重点实验室_外文论文 |
| 通讯作者 | Wu, Wenzhou |
| 作者单位 | 1.Chinese Acad Sci, Inst Geog Sci & Nat Resources Res, State Key Lab Resources & Environm Informat Syst, Beijing 100101, Peoples R China; 2.Univ Chinese Acad Sci, Coll Resource & Environm, Beijing 100049, Peoples R China; 3.Lanzhou Jiaotong Univ, Fac Geomat, Lanzhou, Peoples R China |
| 推荐引用方式 GB/T 7714 | Zhang, Peng,Wu, Wenzhou,Shi, Shaochen,et al. CA-STIM: an interpolation model with spatio-temporal evolution characteristics and cross-attention mechanism for 2D island morphology sequences[J]. INTERNATIONAL JOURNAL OF DIGITAL EARTH,2025,18(1):2513591. |
| APA | Zhang, Peng,Wu, Wenzhou,Shi, Shaochen,Li, Fengyu,&Su, Fenzhen.(2025).CA-STIM: an interpolation model with spatio-temporal evolution characteristics and cross-attention mechanism for 2D island morphology sequences.INTERNATIONAL JOURNAL OF DIGITAL EARTH,18(1),2513591. |
| MLA | Zhang, Peng,et al."CA-STIM: an interpolation model with spatio-temporal evolution characteristics and cross-attention mechanism for 2D island morphology sequences".INTERNATIONAL JOURNAL OF DIGITAL EARTH 18.1(2025):2513591. |
入库方式： OAI收割
来源：地理科学与资源研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。