Enhanced Corn Mapping with Height-Spectral Gaussian Mixture Modeling DOI

Guilong Xiao,

Jianxi Huang, Xuecao Li

и другие.

Опубликована: Янв. 1, 2024

Язык: Английский

Integrating GEDI, Sentinel-2, and Sentinel-1 imagery for tree crops mapping DOI Creative Commons
Esmaeel Adrah, Jesse P. Wong, He Yin

и другие.

Remote Sensing of Environment, Год журнала: 2025, Номер 319, С. 114644 - 114644

Опубликована: Фев. 11, 2025

Язык: Английский

Процитировано

2

A framework for montane forest canopy height estimation via integrating deep learning and multi-source remote sensing data DOI Creative Commons
Hongbin Luo, Guanglong Ou,

Cairong Yue

и другие.

International Journal of Applied Earth Observation and Geoinformation, Год журнала: 2025, Номер unknown, С. 104474 - 104474

Опубликована: Март 1, 2025

Язык: Английский

Процитировано

1

A Segment Anything Model based weakly supervised learning method for crop mapping using Sentinel-2 time series images DOI Creative Commons
Jialin Sun, Shuai Yan, Xiaochuang Yao

и другие.

International Journal of Applied Earth Observation and Geoinformation, Год журнала: 2024, Номер 133, С. 104085 - 104085

Опубликована: Авг. 10, 2024

Язык: Английский

Процитировано

7

Vegetable Fields Mapping in Northeast China Based on Phenological Features DOI Creative Commons
Jialin Hu, Huimin Lu,

Kaishan Song

и другие.

Agronomy, Год журнала: 2025, Номер 15(2), С. 307 - 307

Опубликована: Янв. 26, 2025

Developing vegetable agriculture is crucial for ensuring a balanced dietary structure and promoting nutritional health. However, remote sensing extraction in open-field planting areas faces several challenges, such as the mixing of target crops with natural vegetation caused by differences climate conditions practices, which hinders development large-scale field mapping. This paper proposes classification method based on phenological characteristics (VPC), takes into account spatiotemporal heterogeneity cultivation Northeast China. We used two-step strategy. First, Sentinel-2 satellite images land use data were utilized to identify optimal time key indicators detection crop growth. Second, spectral analysis was integrated three machine learning classifiers, leveraged features extracted from accurately vegetable-growing areas. combined approach enabled generation high-precision map. The research findings reveal consistent year-by-year increase area vegetables 2019 2023. overall accuracy (OA) results ranges 0.81 0.93, Kappa coefficient 0.83. Notably, this first 10 m resolution regional map China, marking significant advancement economic

Язык: Английский

Процитировано

0

Enhancing soil organic carbon prediction by unraveling the role of crop residue coverage using interpretable machine learning DOI Creative Commons
Yi Dong, Xinting Wang, Sheng Wang

и другие.

Geoderma, Год журнала: 2025, Номер 455, С. 117225 - 117225

Опубликована: Фев. 21, 2025

Язык: Английский

Процитировано

0

Integrating isolation forest and deep learning for reliability check of land vector patch types DOI
Shengli Wang, Nanshan Zheng,

Yihu Zhu

и другие.

International Journal of Remote Sensing, Год журнала: 2025, Номер unknown, С. 1 - 30

Опубликована: Апрель 16, 2025

Язык: Английский

Процитировано

0

Enhancing Crop Type Mapping in Data-Scarce Regions Through Transfer Learning: A Case Study of the Hexi Corridor DOI Creative Commons
Juliane Mai, Qisheng Feng, Shuai Fu

и другие.

Remote Sensing, Год журнала: 2025, Номер 17(9), С. 1494 - 1494

Опубликована: Апрель 23, 2025

Timely and accurate crop mapping is crucial for providing essential data support agricultural production management. Reliable ground truth samples form the foundation using remote sensing imagery, a task that presents significant challenges in regions with limited sample availability. To address this issue, study evaluates instance-based transfer learning methods, Hexi Corridor as case to explore strategies areas scarce samples. High-confidence pixels from United States Cropland Data Layer (CDL), along high-density time series derived Sentinel-1, Sentinel-2, Landsat-8 satellite well key vegetation indices, were selected training source domain. Various algorithms, including Random Forest (RF), Extreme Gradient Boosting (XGBoost), TrAdaBoost, employed knowledge domain target type mapping. The results demonstrated during process only data—without utilizing any data—the overall classification accuracy reached 73.88%, optimal accuracies maize alfalfa at 88.97% 85.23%, respectively. As gradually incorporated, total all models ranged 0.77 0.92, F1-scores ranging 0.76 showing consistent improvement model performance. This highlights feasibility of employing Corridor, demonstrating its potential reduce labeling costs valuable reference

Язык: Английский

Процитировано

0

Forest Canopy Height Retrieval Model Based on a Dual Attention Mechanism Deep Network DOI Open Access
Zongze Zhao, Baogui Jiang, Hongtao Wang

и другие.

Forests, Год журнала: 2024, Номер 15(7), С. 1132 - 1132

Опубликована: Июнь 28, 2024

Accurate estimation of forest canopy height is crucial for biomass inversion, carbon storage assessment, and forestry management. However, deep learning methods are underutilized compared to machine learning. This paper introduces the convolutional neural network–bidirectional long short-term memory (CNN-BiLSTM) model proposes a Convolutional Neural network–spatial channel attention–bidirectional (CNN-SCA-BiLSTM) model, incorporating dual attention mechanisms richer feature extraction. A dataset comprising vegetation indices data from regions in Luoyang, specifically within 8–20 m range, used comparative analysis multiple models, with accuracy evaluated based on mean absolute error (MAE), root square (RMSE), coefficient determination (R2). The results demonstrate that (1) CNN-BiLSTM exhibits strong potential (MAE = 1.6554 m, RMSE 2.2393 R2 0.9115) (2) CNN-SCA-BiLSTM while slightly less efficient (<1%), demonstrates improved performance. It reduces MAE by 0.3047 0.6420 increases value 0.0495. Furthermore, utilized generate map 5.2332 7.0426 m) Henan Yellow River Basin year 2022. primarily distributed around 5–20 approaching levels global maps 4.0 6.0 m).

Язык: Английский

Процитировано

3

Regional Scale Inversion of Chlorophyll Content of Dendrocalamus giganteus by Multi-Source Remote Sensing DOI Open Access
Cuifen Xia, Wenwu Zhou, Qingtai Shu

и другие.

Forests, Год журнала: 2024, Номер 15(7), С. 1211 - 1211

Опубликована: Июль 12, 2024

The spectrophotometer method is costly, time-consuming, laborious, and destructive to the plant. Samples will be lost during transportation process, can only obtain sample point data. This poses a challenge estimation of chlorophyll content at regional level. In this study, in order improve accuracy, new collaborative inversion using Landsat 8 Global Ecosystem Dynamics Investigation (GEDI) proposed. Specifically, data set combined with preprocessed two remote-sensing (RS) factors construct three regression models support vector machine (SVM), BP neural network (BP) random forest (RF), better model selected for inversion. addition, ordinary Kriging (OK) used interpolate GEDI attribute into surface modeling. results showed following: (1) single plant was y = 0.1373x1.7654. (2) optimal semi-variance function pai, pgap_theta pgap_theta_a3 are exponential models. (3) top correlations between RS were B2_3_SM, B2_3_HO, B2_5_EN pgap_theta, pgap_theta_a3. (4) combination imagery resulted highest modeling RF had best performance, R2, RMSE P values 0.94, 0.18 g/m2 83.32%, respectively. study shows that it reliable use images retrieve Dendrocalamus giganteus (D. giganteus), revealing potential multi-source ecological parameters.

Язык: Английский

Процитировано

3

The novel triangular spectral indices for characterizing winter wheat drought DOI Creative Commons
Xuan Fu, Hui Liu,

JingHao Xue

и другие.

International Journal of Applied Earth Observation and Geoinformation, Год журнала: 2024, Номер 134, С. 104151 - 104151

Опубликована: Сен. 13, 2024

Язык: Английский

Процитировано

1