A5047206939- Remote Sensing in Agriculture
- Smart Agriculture and AI
- Land Use and Ecosystem Services
- Spectroscopy and Chemometric Analyses
- Climate change impacts on agriculture
- Synthetic Aperture Radar (SAR) Applications and Techniques
- Flood Risk Assessment and Management
- Remote Sensing and Land Use
- Domain Adaptation and Few-Shot Learning
- Time Series Analysis and Forecasting
- Genetic Mapping and Diversity in Plants and Animals
- Healthcare Systems and Challenges
- Crop Yield and Soil Fertility
- Remote-Sensing Image Classification
- Geophysical Methods and Applications
- Irrigation Practices and Water Management
- Machine Learning and ELM
- Mechanisms of cancer metastasis
- Soil Moisture and Remote Sensing
- Topic Modeling
- Anomaly Detection Techniques and Applications
- Natural Language Processing Techniques
- Clinical Nutrition and Gastroenterology
- Diverse Topics in Contemporary Research
- Groundwater and Watershed Analysis
Chinese Academy of Sciences
2022-2025
Aerospace Information Research Institute
2022-2025
State Key Laboratory of Remote Sensing Science
2022-2025
State Key Laboratory of Information Engineering in Surveying Mapping and Remote Sensing
2019-2022
Wuhan University
2019-2022
Yield prediction is essential in food security, trade, and field management. However, due to the associated complex formation mechanisms of yield, accurate timely yield remains challenging remote sensing-based crop monitoring domains. In this study, a framework soybean integrating extreme gradient boosting (XGBoost) multidimensional feature engineering was developed at county level United States using publicly available datasets. Excellent accuracy values were obtained for over 959 counties...
The application of machine learning in crop yield prediction has gained considerable traction, yet uncertainties persist regarding the impact trends on these predictions and differences between detrending methods. In our study, we utilized extreme gradient boosting (XGBoost) to scrutinize effects no trend processing (NTP), input year as a feature (IYF), average (IAYF), linear (ILYF), global method (GDT) maize soybean Midwestern United States. Based findings, compared with that NTP,...
Forty percent of global food production relies upon irrigation, which accounts for 70% total freshwater use. Thus, the mapping cropland irrigation plays a significant role in agricultural water management and estimating production. However, current spaceborne irrigated is highly reliant its spectral behavior, often has high uncertainty lacks information about method irrigation. Deep learning (DL) allows classification according to unique spatial patterns, such as central pivot system (CPIS)....
Traditional methods for crop data collection are labor-intensive, inefficient and, more costly compared to remote sensing (RS) techniques. This study aims identify key climatic variables influencing maize and wheat yields develop predictive models while also evaluating the performance of CropWatch cloud yield prediction model (CW_YPM) in major agricultural regions Ethiopia. Climate from 54 meteorological stations spanning 2000–2021 were analyzed. RS data, including NDVI MODIS at 250 m...
The quick and accurate extraction of water bodies from images is imperative for land resources management, ecological protection, flood disaster prevention. However, the prevalent methods body SAR imagery have major issues depending on expert experience, poor retention water-land boundaries, a high false alarm rate. In this paper, focusing these issues, we study effectiveness using only simple polarimetric decomposition components commonly used machine learning classifiers Gaofen-3(GF-3)...
The combination of transfer learning and remote sensing image processing technology can effectively improve the automation level information extraction from a time series. However, in polarimetric synthetic aperture radar (PolSAR) time-series images, existing methods often cannot make full use relying too much on labeled samples target domain. Furthermore, speckle noise inherent (SAR) imagery aggravates difficulty manual selection samples, so these have meeting requirements large data...
This study introduces a novel approach to improve crop classification accuracy in airborne synthetic aperture radar (SAR) time-series imagery, focusing on overcoming the challenges posed by incidence angle effect. The aims innovate integration of transfer learning and variational mode decomposition techniques. Transfer effectively addresses disparities data distribution caused varying angles encountered SAR. Variational extracts robust temporal features, significantly reducing sensitivity...
The information extraction of polarimetric synthetic aperture radar (PolSAR) images typically requires a great number training samples; however, the samples from historical are less reusable due to distribution differences. Consequently, there is significant manual cost collecting when processing new images. In this paper, address problem, we propose novel active transfer learning method, which combines and deep forest model perform learning. main idea proposed method gradually improve...
Abstract. Spatial-temporal distribution information on global crop production is of crucial for studying food security and promoting sustainable agricultural development. However, the presently available datasets related to this subject are characterized by coarse resolution discontinuous time spans. To tackle these problems, we have integrated multiple data sources, including statistical data, gridded agroclimatic indicator agronomic land surface satellite products ground develop a...
Airborne SAR is an important data source for crop mapping and has applications in agricultural monitoring food safety. However, the incidence-angle effects of airborne imagery decrease accuracy. An active pairwise constraint learning method (APCL) proposed constrained time-series clustering to address this problem. APCL constructs two types instance-level constraints based on incidence angles samples a non-iterative batch-mode selection scheme: must-link constraint, which links objects same...
地块作为农业耕作的最小单元,对其精准识别是国土资源监测、耕地利用监测的需要。现有的方法多使用手工勾绘的方式获取,耗时费力,成本高昂,并且无法实现实时、近实时更新。本文设计了一种基于空间注意力机制与多任务学习的地块分割模型—Field-Net。模型基于UNet架构,增加了空间注意力机制,并采用多任务学习的策略,在语义分割的基础上增加了边界、像素到地块边界的距离等任务。在山东省东营市利津县对模型的性能进行了测试,结果发现耕地地块识别的交并比达到了87.05%,总体精度为92.23%。Field-Net模型的性能优于几种高性能的深度学习框架,交并比较Link-Net模型高出0.26%,较DeepLab...
Abstract The global gridded crop production dataset at 10 km resolution from 2010 to 2020 (GGCP10) for maize, wheat, rice, and soybean was developed address limitations of existing datasets characterized by coarse discontinuous time spans. GGCP10 generated using a series adaptively trained data-driven spatial estimation models integrating multiple data sources, including statistical data, agroclimatic indicator agronomic land surface satellite products, ground data. These were based on...