A Multisource Dynamic Fusion Network for Urban Functional Zone Identification on Remote Sensing, POI, and Building Footprint DOI Creative Commons
H Qiao, Huiping Jiang, Gang Yang

et al.

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, Journal Year: 2024, Volume and Issue: 17, P. 10583 - 10599

Published: Jan. 1, 2024

Urban functional zones (UFZ) identification with remote sensing imagery (RSI) is attracting increasing attention in urban planning and resource allocation areas, etc. The UFZ a comprehensive unit comprising geographical, how to effectively integrate the RSI points of interest (POI) different physical socioeconomic characteristics important promising. However, there are two challenges for identification. On one hand, closely related buildings, most current methods lack an in-depth understanding building semantics. Therefore, efficient integration footprint (FT) data deserves further investigation. other these RSI, POI, FT heterogeneous; leverage complementary information among highly heterogeneous modalities enhance urban. To solve above challenges, this study introduces end-to-end deep learning-based multi-source dynamic fusion network on FT. In proposed method, adaptive weight interactive module (AW-IFM) designed comprehensively sources. addition, multi-scale feature focus (MS-FFM) extract image features emphasize critical characteristics. This method was applied classification Ningbo, Zhejiang Province, China, experimental results demonstrate competitive performance.

Language: Английский

Remote sensing of diverse urban environments: From the single city to multiple cities DOI Creative Commons
Gang Chen, Yuyu Zhou, James A. Voogt

et al.

Remote Sensing of Environment, Journal Year: 2024, Volume and Issue: 305, P. 114108 - 114108

Published: March 14, 2024

Language: Английский

Citations

15
SwinTFNet: Dual-stream Transformer with Cross Attention Fusion for Land Cover Classification DOI
Bo Ren, Bo Liu, Biao Hou

et al.

IEEE Geoscience and Remote Sensing Letters, Journal Year: 2024, Volume and Issue: 21, P. 1 - 5

Published: Jan. 1, 2024

Land cover classification (LCC) is an important application in remote sensing data interpretation. As two common sources, SAR images can be regarded as effective complement to optical images, which will reduce the influence caused by single-modal data. But LCC methods are focusing on designing advanced network architectures process Few works have been oriented toward improving segmentation performance through fusing multi-modal In order deeply integrate and features, we propose SwinTFNet, a dual-stream deep fusion network. Through global context modeling capability of Transformer structure, SwinTFNet models teleconnections between pixels other regions cloud for better prediction regions. addition, Cross-Attention Fusion Module (CAFM) proposed fuse features from Experimental results show that our method improves greatly clouded compared with excellent achieves best

Language: Английский

Citations

4
Land use cover changes abated terrestrial ecosystem carbon sink in China during the past four decades DOI Creative Commons

Xueqing Jiang,

Jinxun Liu, Changhui Peng

et al.

All Earth, Journal Year: 2025, Volume and Issue: 37(1), P. 1 - 14

Published: Jan. 15, 2025

Changes in land use and cover can strongly affect terrestrial carbon balance, which turn the calculation of sinks that will keep future temperature within desired limits. Understanding how changes influence is challenging. Here, we simulated net balance across China with full consideration between 1981 2020 using dynamic global vegetation model. The results indicated sink ecosystem have grown steadily particularly since 2001, average values primary productivity, productivity biome were 3317 TgC • yr−1, 325 yr−1 70 yr−1. However, during period, cumulatively reduced by 1,353.00 TgC, 1,290.71 226.93 TgC. Land created a source effect abated 1981. Our findings may help guide policies to regulate order achieve neutrality future.

Language: Английский

Citations

0
Learning Global Context and Fine Structures for Enhanced Hyperspectral Subpixel Mapping DOI
Wen Zhou, Ailong Ma, Da He

et al.

IEEE Geoscience and Remote Sensing Letters, Journal Year: 2025, Volume and Issue: 22, P. 1 - 5

Published: Jan. 1, 2025

Language: Английский

Citations

0
Revealing the fireworks-set-off pattern of spatial multi-function expansion across cities leveraging big geodata – a case of the Greater Bay Area, China DOI Creative Commons
Ku Gao, Xiaomei Yang, Zhihua Wang

et al.

International Journal of Digital Earth, Journal Year: 2025, Volume and Issue: 18(1)

Published: Jan. 21, 2025

Language: Английский

Citations

0
Spatial-temporal patterns of cultivated land expansion and intensification in Africa from 2000 to 2020 DOI Creative Commons
Mengxi Wang, Cong Wang, Qiong Hu

et al.

International Journal of Agricultural Sustainability, Journal Year: 2025, Volume and Issue: 23(1)

Published: March 12, 2025

Language: Английский

Citations

0
Decreased and Fragmented Greenspaces in and around Rural Residential Areas of Eastern China in the Process of Urbanization DOI

W Li,

Jun Wang,

Yuan Luo

et al.

Remote Sensing Applications Society and Environment, Journal Year: 2025, Volume and Issue: unknown, P. 101518 - 101518

Published: March 1, 2025

Language: Английский

Citations

0
DLAN: A Dual Attention Network for Effective Land Cover Classification in Remote Sensing DOI
Muhammad Fayaz, L. Minh Dang, Hyeonjoon Moon

et al.

Knowledge-Based Systems, Journal Year: 2025, Volume and Issue: unknown, P. 113620 - 113620

Published: April 1, 2025

Language: Английский

Citations

0
HR-UVFormer: A Top-Down and Multimodal Hierarchical Extraction Approach for Urban Villages DOI
Xin Tan, Qingyan Meng, Fei Zhao

et al.

IEEE Transactions on Geoscience and Remote Sensing, Journal Year: 2024, Volume and Issue: 62, P. 1 - 15

Published: Jan. 1, 2024

Urban Villages (UVs) renovation has been incorporated into the Sustainable Development Goals (SDGs) as a result of inequality issue among residents garnering substantial social attention. However, existing deep-learning techniques for UVs extraction have limited to single spatial scale (e.g., patch-level or pixel-level extraction), leading inadequate precision and integrity in their outcomes. To overcome this limitation, our study introduces HR-UVFormer, top-down multimodal hierarchical approach that extracts from coarse (patch) fine granularity (pixel), aiming enhance internal completeness boundary accuracy results. The can effectively fuse features building footprints (BF)) with remote sensing images (RSI) extraction. Shenzhen results indicate coarse-scale achieves an overall (OA) 98.79%, fine-grained mean Intersection over Union (mIoU) 93.60%. Furthermore, ablation experiments demonstrate notable 7.14% improvement mIoU strategy compared traditional pixel-based strategy, fusion BF RSI yields further improvements 2.78% 0.65% OA mIoU, respectively. This finding confirms synergistic effect between extraction, which analyzed study. Additionally, proposed model outperforms other deep learning models exhibits potential support more modal POI). Finally, experimental dataset code be publicly accessed at https://github.com/q1310546582/HR-UVFormer-code.

Language: Английский

Citations

3
A Noval Super-Resolution Model for 10-m Mangrove Mapping With Landsat-5 DOI
Wei Chen, Jinyan Tian, Jie Song

et al.

IEEE Transactions on Geoscience and Remote Sensing, Journal Year: 2024, Volume and Issue: 62, P. 1 - 12

Published: Jan. 1, 2024

Existing temporal mangrove products are at a 30-m resolution from Landsat, facing challenges such as unclear delineation of community edges, difficulty in identifying creeks and open spaces within communities, ineffective recognition small patches. Therefore, there is an urgent need to produce higher (e.g., 10-m) with particularly considering the absence available Sentinel imagery before 2015. To this end, we propose novel super-resolution model that incorporating Residual Channel Attention Networks (RCAN) Texture Transformer Network (TTSR) generate 10-m Landsat-5, namely RCAN-TTSR. RCAN TTSR play crucial roles different perspectives process, respectively. accurately transfers texture information Sentinel-2 Landsat by computing correlation between them. On other hand, assigns weights multiple low-frequency features number high-frequency derived raw bands imagery, thus achieving better outcomes. The results demonstrate images produced significantly outperform existing models terms PSNR SSIM metrics. Furthermore, random forest classifier was employed for mapping. Compared products, our map shows mapping accuracy finer spatial details.

Language: Английский

Citations

3