Environmental Monitoring and Assessment, Год журнала: 2024, Номер 196(12)
Опубликована: Ноя. 21, 2024
Язык: Английский
Environmental Monitoring and Assessment, Год журнала: 2024, Номер 196(12)
Опубликована: Ноя. 21, 2024
Язык: Английский
Journal of Environmental Management, Год журнала: 2025, Номер 376, С. 124443 - 124443
Опубликована: Фев. 8, 2025
Язык: Английский
Процитировано
1ISPRS Journal of Photogrammetry and Remote Sensing, Год журнала: 2025, Номер 225, С. 440 - 460
Опубликована: Май 15, 2025
Язык: Английский
Процитировано
0Applied Sciences, Год журнала: 2025, Номер 15(11), С. 6292 - 6292
Опубликована: Июнь 3, 2025
Precise identification of water bodies in agricultural watersheds is crucial for irrigation, resource management, and flood disaster prevention. However, the spectral noise caused by complex light shadow interference quality differences, combined with diverse shapes high computational cost image processing, severely limits accuracy body recognition watersheds. This paper proposed a lightweight efficient learnable Kalman filter Deformable Convolutional Attention Network (LKF-DCANet). The encoder built using shallow Channel Attention-Enhanced Convolution module (CADCN), while decoder combines Additive Token Mixer (CATM) (LKF) to achieve adaptive suppression enhance global context modeling. Additionally, feature-based knowledge distillation strategy employed further improve representational capacity model. Experimental results show that LKF-DCANet achieves an Intersection over Union (IoU) 85.95% only 0.22 M parameters on public dataset. When transferred self-constructed UAV dataset, it IoU 96.28%, demonstrating strong generalization ability. All experiments are conducted RGB optical imagery, confirming offers highly versatile solution segmentation precision agriculture.
Язык: Английский
Процитировано
0Agronomy, Год журнала: 2024, Номер 14(12), С. 2986 - 2986
Опубликована: Дек. 15, 2024
Accurate mapping of tea plantations is crucial for agricultural management and economic planning, yet it poses a significant challenge due to the complex variable nature cultivation landscapes. This study presents high-precision approach in Anji County, Zhejiang Province, China, utilizing multi-source remote sensing data advanced deep learning models. We employed combination Sentinel-2 optical imagery, Sentinel-1 synthetic aperture radar digital elevation models capture rich spatial, spectral, temporal characteristics plantations. Three models, namely U-Net, SE-UNet, Swin-UNet, were constructed trained semantic segmentation Cross-validation point-based accuracy assessment methods used evaluate performance The results demonstrated that Swin-UNet model, transformer-based capturing long-range dependencies global context superior feature extraction, outperformed others, achieving an overall 0.993 F1-score 0.977 when using multi-temporal data. integration with slightly improved classification accuracy, particularly areas affected by cloud cover, highlighting complementary imagery all-weather monitoring. also analyzed influence terrain factors, such as elevation, slope, aspect, on plantation mapping. It was found at higher altitudes or north-facing slopes exhibited improves increasing likely simpler land cover types tea’s preference shade. findings this research not only provide valuable insights into precision but contribute broader application
Язык: Английский
Процитировано
0Environmental Monitoring and Assessment, Год журнала: 2024, Номер 196(12)
Опубликована: Ноя. 21, 2024
Язык: Английский
Процитировано
0