TDSCCNet: twin-depthwise separable convolution connect network for change detection with heterogeneous images
文献类型：期刊论文
| 作者 | Wang, Min1; Huang, Liang1,2; Tang, Bo-Hui1,2,3; Le, Weipeng1; Tian, Qiuyuan1 |
| 刊名 | GEOCARTO INTERNATIONAL
 |
| 出版日期 | 2024 |
| 卷号 | 39期号:1页码:21 |
| 关键词 | Heterogeneous images change detection image domain transformation TDSCCNet deep learning |
| ISSN号 | 1010-6049 |
| DOI | 10.1080/10106049.2024.2329673 |
| 通讯作者 | Huang, Liang(kmhuangliang@kust.edu.cn) |
| 英文摘要 | The task of change detection (CD) in optical and SAR images is an ever-evolving and demanding subject within the realm of remote sensing (RS). It holds great significance to identify the target areas by using complementary information between the two. Due to the distinct imaging mechanisms employed by optical and SAR sensors, effectively and accurately identifying changing regions can be challenging. To this end, a novel heterogeneous RS images CD network (Twin-Depthwise Separable Convolution Connect Network, TDSCCNet) is proposed in this paper. Image domain transformation is a front-end task, while the back-end employs a single-branch bilayer depthwise separable convolution-connected encoder-decoder to accomplish CD work. Specifically, first, the cycle-consistent adversarial network (CycleGAN) serves to integrate the optical and SAR visual domains, and a consistent feature expression is obtained. Second, the single-branch encoder structure of bilayer depthwise separable convolution is employed to realize change feature extraction. Finally, the multiscale connected decoder reconstructed by change map is utilized to reconstruct the original images and solve the local discontinuities and holes in the binary change map. Multiscale loss is designated for optimizing the global and local effects to alleviate the class imbalance problem. It was tested on four representative datasets from Gloucester, Shuguang Village, Italy and WV-3 datasets with an overall accuracy of 97.36%, 97.01%, 97.62%, and 98.01% respectively. By comparing the existing methods, experimental results confirmed the effectiveness of the proposed method. |
| WOS关键词 | CLASSIFICATION |
| 资助项目 | National Natural Science Foundation of China |
| WOS研究方向 | Environmental Sciences & Ecology ; Geology ; Remote Sensing ; Imaging Science & Photographic Technology |
| 语种 | 英语 |
| WOS记录号 | WOS:001187559300001 |
| 出版者 | TAYLOR & FRANCIS LTD |
| 资助机构 | National Natural Science Foundation of China |
| 源URL | [http://ir.igsnrr.ac.cn/handle/311030/203700]  |
| 专题 | 中国科学院地理科学与资源研究所 |
| 通讯作者 | Huang, Liang |
| 作者单位 | 1.Kunming Univ Sci & Technol, Fac Land Resources Engn, Kunming, Peoples R China 2.Yunnan Prov Dept Educ, Key Lab Plateau Remote Sensing, Kunming, Peoples R China 3.Chinese Acad Sci, Inst Geog Sci & Nat Resources Res, Beijing, Peoples R China |
| 推荐引用方式 GB/T 7714 | Wang, Min,Huang, Liang,Tang, Bo-Hui,et al. TDSCCNet: twin-depthwise separable convolution connect network for change detection with heterogeneous images[J]. GEOCARTO INTERNATIONAL,2024,39(1):21. |
| APA | Wang, Min,Huang, Liang,Tang, Bo-Hui,Le, Weipeng,&Tian, Qiuyuan.(2024).TDSCCNet: twin-depthwise separable convolution connect network for change detection with heterogeneous images.GEOCARTO INTERNATIONAL,39(1),21. |
| MLA | Wang, Min,et al."TDSCCNet: twin-depthwise separable convolution connect network for change detection with heterogeneous images".GEOCARTO INTERNATIONAL 39.1(2024):21. |
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。