[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-3494":3,"related-3494":69},{"id":4,"title":5,"url":6,"summary":7,"summary_zh":8,"content":9,"source_name":10,"source_url":6,"published_at":11,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":15,"score_detail":16,"sources":24,"tags":26,"search_phrases":32,"slug":35,"view_count":36,"doi":37,"paper":38,"created_at":68},3494,"Rice Cropping Pattern Mapping Using GEE-Based Sentinel-1\u002F2 Time-Series Imagery and Machine Learning","https:\u002F\u002Fdoi.org\u002F10.3390\u002Frs18193302","Accurate mapping of rice cropping patterns is fundamental to sustainable agricultural management and regional food security assessment. In this study, we developed a high-precision mapping framework for major rice cropping patterns in the Yangtze River Delta region, China, using the Google Earth Engine (GEE) cloud platform. A multi-temporal and multi-source feature set was constructed by integrating Sentinel-1 radar backscatter (VV\u002FVH polarization), Sentinel-2 optical indices, including the normalized difference vegetation index (NDVI) and land surface water index (LSWI), and topographic factors (DEM and slope), to classify three predominant cropping systems, namely, wheat–rice rotation, double rice, and rapeseed–rice rotation. A total of 548 field-surveyed sample points collected during the 2024–2025 growing season were used for model training and validation. Three classical classifiers—Random Forest (RF), A Gradient Boosting Tree (GBTREE), and a Support Vector Machine (SVM)—were systematically compared. The ablation experiment demonstrated that the fusion of Sentinel-1 and Sentinel-2 outperformed both the Sentinel-1-only and Sentinel-2-only configurations across all three classifiers. Among them, the GBTREE achieved the highest overall accuracy (93.8%), Kappa coefficient (0.87), and macro-average F1 score (88.3%) in this specific experiment. Notably, it also performed best on the more challenging double-rice class. The SHapley Additive exPlanations (SHAP)-based feature importance analysis revealed that multi-temporal NDVI phenological features were the primary drivers of classification accuracy, while radar backscatter and water indices provided essential complementary information, and topographic factors served as spatial constraints at the regional scale. The spatial distribution derived from the GBTREE classification exhibited clear patterns: wheat–rice rotation dominated the northern plains (northern Jiangsu, northern Anhui, and the Hangjiahu Plain); double rice was concentrated in the southern Zhejiang hills and scattered valley plains; and rapeseed–rice rotation showed a scattered, mosaic distribution. Overall, this study demonstrates that integrating multi-source remote sensing data on the GEE platform with the GBTREE classifier enables effective and scalable high-precision mapping of rice cropping patterns in complex agricultural landscapes. This approach provides a reliable technical foundation for regional agricultural structure analysis, crop rotation assessment, and sustainable agricultural monitoring.","准确绘制水稻种植模式图是实现可持续农业管理和区域粮食安全评估的基础。本研究基于Google Earth Engine（GEE）云平台，构建了长江三角洲地区主要水稻种植模式的高精度制图框架。通过整合Sentinel-1雷达后向散射（VV\u002FVH极化）、Sentinel-2光学指数（包括归一化差异植被指数NDVI和地表水体指数LSWI）以及地形因子（DEM和坡度），构建了多时相、多源特征集，用于分类三种主要种植制度，即麦–稻轮作、双季稻和油–稻轮作。利用2024—2025年生长季采集的548个实地调查样点进行模型训练与验证。系统比较了三种经典分类器——随机森林（RF）、梯度提升树（GBTREE）和支持向量机（SVM）。消融实验表明，Sentinel-1与Sentinel-2的融合在三种分类器中均优于仅使用Sentinel-1或仅使用Sentinel-2的配置。其中，GBTREE在本实验中取得了最高的总体精度（93.8%）、Kappa系数（0.87）和宏平均F1分数（88.3%）。值得注意的是，其在更具挑战性的双季稻类别上同样表现最佳。基于SHapley加法解释（SHAP）的特征重要性分析表明，多时相NDVI物候特征是分类精度的主要驱动因素，雷达后向散射和水体指数提供了必要的补充信息，而地形因子则在区域尺度上起到空间约束作用。基于GBTREE分类得到的空间分布呈现出清晰的格局：麦–稻轮作主导北部平原（苏北、皖北和杭嘉湖平原）；双季稻集中于浙南丘陵和零散的河谷平原；油–稻轮作则呈零散镶嵌状分布。总体而言，本研究表明，在GEE平台上整合多源遥感数据与GBTREE分类器，能够对复杂农业景观中的水稻种植模式进行有效且可扩展的高精度制图。该方法为区域农业结构分析和作物",null,"Remote Sensing","2026-09-24T00:00:00Z","论文",10,false,79,{"impact":17,"substance":18,"depth":19,"authority":20,"freshness":21,"relevant":22,"comment":23},16,22,18,14,9,1,"基于GEE与Sentinel-1\u002F2时序影像结合机器学习实现长三角水稻种植模式高精度制图，方法扎实、结论可靠，对农业遥感监测有参考价值。",[25],{"name":10,"url":6},[27,28,29,30,31],"智慧农业","水稻","机器学习","遥感","作物分类",[33,34],"长三角 水稻 种植模式 遥感","Sentinel-1 Sentinel-2 水稻制图","长三角水稻种植模式遥感-3494",0,"10.3390\u002Frs18193302",{"doi":37,"openalex_id":39,"authors":40,"venue":10,"cited_by_count":36,"oa_url":6,"card":61,"direction":65,"ingested_from":67},"W7214147197",[41,44,47,50,52,55,58],{"name":42,"orcid":43},"Xuan Li","https:\u002F\u002Forcid.org\u002F0000-0001-5509-2385",{"name":45,"orcid":46},"Lintao Chen","https:\u002F\u002Forcid.org\u002F0009-0000-6558-1289",{"name":48,"orcid":49},"Lin Chen","https:\u002F\u002Forcid.org\u002F0000-0002-9270-1626",{"name":51,"orcid":9},"Chao Su",{"name":53,"orcid":54},"Hoi Leong Lee","https:\u002F\u002Forcid.org\u002F0000-0002-4984-2183",{"name":56,"orcid":57},"Ruci Wang","https:\u002F\u002Forcid.org\u002F0000-0001-7049-7006",{"name":59,"orcid":60},"Xuguang Tang","https:\u002F\u002Forcid.org\u002F0009-0008-3494-8867",{"tldr":62,"method":63,"finding":64,"direction":65,"opportunity":66},"基于GEE融合Sentinel-1\u002F2时序与地形特征，用机器学习高精度制图长三角水稻种植模式。","GEE平台、Sentinel-1\u002F2时序特征、DEM、548个实地样本、RF\u002FG","GBTREE精度最高（总体93.8%、Kappa 0.87），双季稻识别最好；NDVI物候特征贡献最","农业遥感与作物表型","可探索样本稀缺区迁移学习与多作物轮作模式泛化制图，并耦合产量与碳核算。","openalex","2026-09-25T23:30:30.511522Z",{"total":70,"page":22,"page_size":70,"items":71},6,[72,115,178,216,254,282],{"id":73,"title":74,"url":75,"summary":76,"summary_zh":77,"content":9,"source_name":78,"source_url":75,"published_at":79,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":80,"score_detail":81,"sources":85,"tags":87,"search_phrases":90,"slug":93,"view_count":36,"doi":94,"paper":95,"created_at":114},3328,"Yield ranking of spring wheat breeding lines absent from model training within two contrasting seasons in northern Kazakhstan","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.atech.2026.102588","Reliable assessment of yield models in breeding trials requires evaluation on lines that have not contributed observations to model development. We examined this problem in an advanced spring wheat yield nursery in northern Kazakhstan during the 2024 and 2025 seasons. Of 300 plots, 176 from 22 entries formed the model development set, while 124 plots representing 31 breeding lines were completely excluded from training. Eleven variables were retained from 82 candidates using the training data alone; the candidates were derived from UAV multispectral imagery, ERA5-Land reanalysis, LiDAR-derived static plot microtopography, phenology and maturity group. LightGBM yielded a pooled R² of 0.587 for the excluded lines; resampling breeding lines within season gave a mean R² of 0.564 with a 95 % confidence interval from 0.252 to 0.776. The pooled value partly reflected the strong contrast between seasons rather than discrimination among lines within a season. Random division of plots produced an R² of 0.890, indicating that apparent predictive accuracy depended strongly on validation design. Agreement between observed and predicted line rankings was moderate, with Spearman correlations of 0.612 in 2024 and 0.650 in 2025, although uncertainty was wide because only 14 and 17 independent lines were available. Six modelling approaches produced overlapping uncertainty intervals, with no clear evidence of superiority, and the complete combination of data sources was not separable from simpler configurations retaining the multispectral block. A separate analysis in which screening, selection and fitting used 2024 data alone and evaluation used 2025 data showed a severe loss of performance, with R² = −8.81 and a Spearman correlation of −0.082 that was indistinguishable from zero. Poor transfer persisted among entries grown in both seasons, so genotype novelty alone was insufficient to explain the failure. Breeding lines absent from training could therefore be ranked with moderate consistency when both observed seasonal regimes were represented during model development, whereas prediction in an unobserved season remained unreliable. The study covers one nursery complex and two seasons, so transfer across locations and broader conditions remains to be established.","在育种试验中，要对产量模型进行可靠评估，必须在未参与模型开发的品系上进行评价。我们在2024年和2025年生长季于哈萨克斯坦北部的一个高级春小麦产量圃中考察了这一问题。在300个小区中，来自22个品系的176个小区构成模型开发集，而代表31个育种品系的124个小区则完全排除在训练之外。仅使用训练数据，从82个候选变量中保留了11个变量；这些候选变量来源于无人机多光谱影像、ERA5-Land再分析数据、LiDAR衍生的静态小区微地形、物候和成熟期组。LightGBM对被排除品系给出的合并R²为0.587；在生长季内对育种品系进行重采样得到的平均R²为0.564，95%置信区间为0.252至0.776。合并值部分反映了生长季之间的强烈差异，而非生长季内品系之间的区分能力。对小区进行随机划分得到的R²为0.890，表明表观预测精度在很大程度上取决于验证设计。观测品系排名与预测品系排名之间的一致性为中等，2024年和2025年的Spearman相关系数分别为0.612和0.650，但由于仅有14个和17个独立品系可用，不确定性范围较宽。六种建模方法产生了相互重叠的不确定性区间，没有明确证据表明哪一种更优，并且完整的数据源组合与保留多光谱模块的较简单配置无法区分。另一项分析中，筛选、选择和拟合仅使用2024年数据，而评估使用2025年数据，结果显示性能严重下降，R² = −8.81，Spearman相关系数为−0.082，与零无法区分。在两个生长季均种植的品系之间，较差的迁移性依然存在，因此仅用品系新颖性不足以解释这种失败。因此，当模型开发过程中涵盖了所观测到的两种生长季情形时，未参与训练的育种品系可以以中等一致性进行排名，而在未观测生长季中的预测仍然不可靠。本研究仅涵盖一个圃系复合体和两个生长季，因此跨地点和更广泛条件下的迁移性仍有待确立。","Smart Agricultural Technology","2026-09-22T00:00:00Z",74,{"impact":82,"substance":18,"depth":19,"authority":83,"freshness":21,"relevant":22,"comment":84},12,13,"基于无人机多光谱与气象数据的春小麦育种品系产量预测研究，验证设计严谨、结论审慎，对智慧育种与遥感估产有参考价值，但属单点试验、地域性强，未达重大突破层级。",[86],{"name":78,"url":75},[27,88,29,89,30],"产量预测","小麦育种",[91,92],"哈萨克斯坦 春小麦 产量预测","UAV 多光谱 育种试验","哈萨克斯坦春小麦产量预测-3328","10.1016\u002Fj.atech.2026.102588",{"doi":94,"openalex_id":96,"authors":97,"venue":78,"cited_by_count":36,"oa_url":75,"card":109,"direction":65,"ingested_from":67},"W7214044328",[98,101,104,107],{"name":99,"orcid":100},"Dastan Yelubayev","https:\u002F\u002Forcid.org\u002F0000-0001-5358-7982",{"name":102,"orcid":103},"TIMUR SAVIN","https:\u002F\u002Forcid.org\u002F0000-0002-3550-647X",{"name":105,"orcid":106},"Ismail Tokbergenov","https:\u002F\u002Forcid.org\u002F0000-0002-0656-9914",{"name":108,"orcid":9},"Bakhtiyar Zhanzakov",{"tldr":110,"method":111,"finding":112,"direction":65,"opportunity":113},"评估春小麦育种品系产量模型在未参与训练品系上的跨季预测能力。","无人机多光谱、ERA5-Land、LiDAR与物候数据，LightGBM等六种模","两季均参与训练时品系排名中等一致，但预测未观测季节完全失效。","需研究跨地点、跨年份可迁移的表型预测模型与验证设计，避免随机划分高估精度。","2026-09-24T23:30:03.201775Z",{"id":116,"title":117,"url":118,"summary":119,"summary_zh":120,"content":9,"source_name":121,"source_url":118,"published_at":79,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":122,"score_detail":123,"sources":126,"tags":128,"search_phrases":131,"slug":134,"view_count":36,"doi":135,"paper":136,"created_at":177},3278,"Research on the optimal modeling path for inversion of Pb content in rice leaves based on hyperspectral data of ground objects and machine learning and cross-scale remote sensing monitoring","https:\u002F\u002Fdoi.org\u002F10.1007\u002Fs10661-026-15959-x","Research on the optimal modeling path for inversion of Pb content in rice leaves based on hyperspectral data of ground objects and machine learning and cross-scale remote sensing monitoring。Environmental Monitoring and Assessment","基于地物高光谱数据与机器学习的稻叶铅含量反演最优建模路径及跨尺度遥感监测研究。环境监测与评估","Environmental Monitoring and Assessment",64,{"impact":124,"substance":19,"depth":17,"authority":83,"freshness":21,"relevant":22,"comment":125},8,"基于地物高光谱与机器学习的水稻叶片铅含量反演建模研究，方法有创新但属细分领域学术进展，公共影响有限。",[127],{"name":121,"url":118},[27,129,28,29,130],"农业遥感","高光谱遥感",[132,133],"水稻叶片 铅含量 高光谱 反演","稻米 重金属 遥感 监测","水稻叶片铅含量高光谱反演-3278","10.1007\u002Fs10661-026-15959-x",{"doi":135,"openalex_id":137,"authors":138,"venue":121,"cited_by_count":36,"oa_url":9,"card":172,"direction":65,"ingested_from":67},"W7214027889",[139,142,145,148,151,154,156,158,161,164,166,168,170],{"name":140,"orcid":141},"Zhenlong Zhang","https:\u002F\u002Forcid.org\u002F0009-0008-2354-9123",{"name":143,"orcid":144},"Zhe Wang","https:\u002F\u002Forcid.org\u002F0000-0003-1266-7251",{"name":146,"orcid":147},"Chengxia Wang","https:\u002F\u002Forcid.org\u002F0009-0001-0820-2462",{"name":149,"orcid":150},"Wenxue Lin","https:\u002F\u002Forcid.org\u002F0000-0002-8245-9063",{"name":152,"orcid":153},"Jingyan Zhang","https:\u002F\u002Forcid.org\u002F0009-0004-4567-5316",{"name":155,"orcid":9},"Ying Luo",{"name":157,"orcid":9},"Jiaqian Zhang",{"name":159,"orcid":160},"Kai Ye","https:\u002F\u002Forcid.org\u002F0000-0002-2851-6741",{"name":162,"orcid":163},"Yiming Chen","https:\u002F\u002Forcid.org\u002F0000-0002-8121-3109",{"name":165,"orcid":9},"Chaoliang Peng",{"name":167,"orcid":9},"Duan Tian",{"name":169,"orcid":9},"Weihao Wang",{"name":171,"orcid":9},"Jiaxin Liu",{"tldr":173,"method":174,"finding":175,"direction":65,"opportunity":176},"研究基于地面高光谱与机器学习反演水稻叶片铅含量，并探索跨尺度遥感监测的最优建模路径。","地面高光谱数据结合机器学习建模，开展跨尺度遥感监测。","明确了水稻叶片铅含量反演的最优建模路径，实现跨尺度遥感监测。","可探索多尺度遥感数据融合与迁移学习，提升重金属胁迫反演的普适性与精度。","2026-09-23T23:30:19.291361Z",{"id":179,"title":180,"url":181,"summary":182,"summary_zh":183,"content":9,"source_name":184,"source_url":181,"published_at":185,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":186,"score_detail":187,"sources":189,"tags":191,"search_phrases":194,"slug":197,"view_count":36,"doi":198,"paper":199,"created_at":215},3164,"Spatiotemporal Deep Learning for Rice Plant Height Estimation from Multi-Temporal UAV RGB Imagery","https:\u002F\u002Fdoi.org\u002F10.3390\u002Fagriculture16182034","Accurate plant height estimation is important for monitoring crop growth and supporting precision agricultural management. Manual measurements are labor-intensive, while LiDAR-based methods are expensive and require complex processing. UAV photogrammetry provides a lower-cost alternative but remains challenging in flooded rice paddies because of canopy deformation and difficulties in terrain extraction. This study proposes Rice-STNet, a spatiotemporal deep learning framework for end-to-end rice plant height estimation using multi-temporal UAV RGB imagery. Rice-STNet integrates a convolutional neural network for spatial feature extraction, Time2Vec for temporal encoding, and a gated recurrent unit network for modeling temporal dependencies across observation dates. The framework was evaluated using field data collected from rice paddies over two growing seasons. Rice-STNet achieved an R2 of 0.97, a root mean squared error of 1.97 cm, and a mean absolute error of 1.14 cm. It outperformed random forest, support vector regression, a CNN-only baseline, and a UAV photogrammetry-based point-cloud approach. In addition, the framework generated high-resolution plant height maps for field-scale analysis of spatial growth variability. These results underscore the importance of jointly modeling spatial and temporal characteristics for continuously evolving crop traits. The proposed framework offers an accurate, scalable, and non-destructive solution for large-scale crop phenotyping and precision agriculture.","准确的株高估算对于监测作物生长和支持精准农业管理具有重要意义。人工测量劳动强度大，而基于激光雷达（LiDAR）的方法成本高昂且需要复杂的处理。无人机摄影测量提供了一种成本较低的替代方案，但在淹水稻田中仍面临挑战，原因在于冠层变形和地形提取困难。本研究提出了Rice-STNet，一种时空深度学习框架，用于利用多时相无人机RGB影像进行端到端水稻株高估算。Rice-STNet集成了用于空间特征提取的卷积神经网络、用于时间编码的Time2Vec，以及用于建模观测日期之间时间依赖关系的门控循环单元网络。该框架利用两个生长季从稻田采集的田间数据进行了评估。Rice-STNet取得了R²为0.97、均方根误差为1.97 cm、平均绝对误差为1.14 cm的结果。其性能优于随机森林、支持向量回归、仅使用CNN的基线方法以及基于无人机摄影测量的点云方法。此外，该框架生成了高分辨率株高图，用于田块尺度空间生长变异性分析。这些结果凸显了联合建模空间与时间特征对于持续变化的作物性状的重要性。所提出的框架为大规模作物表型分析和精准农业提供了一种准确、可扩展且非破坏性的解决方案。","Agriculture","2026-09-21T00:00:00Z",80,{"impact":19,"substance":18,"depth":19,"authority":83,"freshness":21,"relevant":22,"comment":188},"提出时空深度学习框架Rice-STNet，用多时相无人机RGB影像实现水稻株高高精度估算，方法新颖、数据跨两个生长季，对作物表型与精准农业有实用价值。",[190],{"name":184,"url":181},[27,192,28,30,193],"农业人工智能","作物表型",[195,196],"无人机 RGB 水稻株高","Rice-STNet 水稻表型","无人机RGB水稻株高-3164","10.3390\u002Fagriculture16182034",{"doi":198,"openalex_id":200,"authors":201,"venue":184,"cited_by_count":36,"oa_url":181,"card":210,"direction":65,"ingested_from":67},"W7213887432",[202,205,207],{"name":203,"orcid":204},"Weiguo Wang","https:\u002F\u002Forcid.org\u002F0009-0003-4028-9363",{"name":206,"orcid":9},"Noboru Noguchi",{"name":208,"orcid":209},"Liangliang Yang","https:\u002F\u002Forcid.org\u002F0000-0002-5055-3987",{"tldr":211,"method":212,"finding":213,"direction":65,"opportunity":214},"提出Rice-STNet时空深度学习框架，用多时相无人机RGB影像估算水稻株高。","CNN提取空间特征，Time2Vec编码时间，GRU建模时序依赖，两季稻田数据验","R²达0.97、RMSE 1.97cm，优于随机森林、SVR、纯CNN及点云方法。","可迁移至其他作物与多源遥感融合，探索轻量化模型及实时田间部署。","2026-09-22T23:30:18.545958Z",{"id":217,"title":218,"url":219,"summary":220,"summary_zh":221,"content":9,"source_name":222,"source_url":219,"published_at":223,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":224,"score_detail":225,"sources":229,"tags":231,"search_phrases":233,"slug":236,"view_count":36,"doi":237,"paper":238,"created_at":253},3009,"AgriMAC: An Attention Based Multimodal Deep Clustering Framework for Rice Health Assessment","https:\u002F\u002Fdoi.org\u002F10.22266\u002Fijies2026.1031.06","Rice is Indonesia's staple crop, yet its productivity has declined in recent years because pest and disease outbreaks remain difficult to detect at an early stage.Existing precision agriculture approaches commonly process Internet of Things (IoT) sensor data and remote sensing imagery independently and often rely on supervised learning, requiring large amounts of labeled data.Meanwhile, multispectral drone imagery producing the Normalized Difference Vegetation Index (NDVI) provides richer information on crop physiological conditions than RGB-based vegetation indices.This study proposes Agricultural Multimodal Attention Clustering (AgriMAC), an unsupervised framework that integrates UAV derived NDVI imagery, 7-in-1 IoT soil sensor measurements, and historical weather data from the Open-Meteo API for rice field condition monitoring.Each modality is encoded using a dedicated autoencoder and fused through an entropy-regularized attention mechanism before Deep Embedded Clustering is performed.To reduce the influence of crop growth stage, the IoT representation is residualized using growth-phase statistics estimated exclusively from the training fold, enabling the discovered clusters to represent within-phase agronomic conditions rather than crop age.Experiments conducted under a grouped leave-one-field-out protocol produced a Silhouette Score of 0.465 ± 0.048, a Davies Bouldin Index of 0.807 ± 0.036, and a Calinski Harabasz Index of 195 ± 27.The learned groups also showed low normalized mutual information with growth phase (0.079 ± 0.043) and near chance phase decodability (balanced accuracy = 0.554 ± 0.042), indicating that they are only weakly associated with crop growth stage.The learned attention weights identified IoT soil measurements (0.570 ± 0.024) as the dominant modality, while NDVI imagery (0.210 ± 0.014) and weather information (0.220 ± 0.014) provided complementary spatial and temporal context.Overall, AgriMAC provides an interpretable and leakage-aware framework for multimodal clustering of rice field conditions.Although its clustering performance is comparable to that of a capacity-matched IoT-only model, it additionally quantifies the contribution of each sensing modality through attention weights and explicitly mitigates the growth-phase confound, making it suitable for field level agronomic condition monitoring and spatial decision support in precision agriculture.","水稻是印度尼西亚的主要作物，但近年来其生产力有所下降，因为病虫害暴发在早期阶段仍难以检测。现有的精准农业方法通常独立处理物联网（IoT）传感器数据和遥感影像，且往往依赖监督学习，需要大量标注数据。与此同时，生成归一化植被指数（NDVI）的多光谱无人机影像比基于RGB的植被指数能提供更丰富的作物生理状况信息。本研究提出农业多模态注意力聚类（AgriMAC），这是一个无监督框架，整合了无人机获取的NDVI影像、七合一IoT土壤传感器测量数据以及来自Open-Meteo API的历史天气数据，用于稻田状况监测。每种模态均使用专用自编码器进行编码，并通过熵正则化注意力机制进行融合，随后执行深度嵌入聚类。为减少作物生长阶段的影响，IoT表征利用仅从训练折估计的生长阶段统计量进行残差化处理，使发现的聚类能够表征阶段内的农艺状况而非作物年龄。在分组留一田块协议下进行的实验产生了0.465 ± 0.048的轮廓系数、0.807 ± 0.036的Davies-Bouldin指数和195 ± 27的Calinski-Harabasz指数。学习到的分组还显示出与生长阶段的低归一化互信息（0.079 ± 0.043）以及接近随机的阶段可解码性（平衡准确率 = 0.554 ± 0.042），表明它们与作物生长阶段仅存在弱关联。学习到的注意力权重将IoT土壤测量（0.570 ± 0.024）识别为主导模态，而NDVI影像（0.210 ± 0.014）和天气信息（0.220 ± 0.014）则提供了互补的空间和时间背景。总体而言，AgriMAC为稻田状况的多模态聚类提供了一个可解释且感知数据泄漏的框架。尽管其聚类性能与容量匹配的仅IoT模型相当，但它还通过注意力权重量化了每种传感模态的贡献，并明确减轻了生长阶段混杂因素，使其适用于田块级农艺状况监测和精准农业中的空间决策支持。","International journal of intelligent engineering and systems","2026-09-19T00:00:00Z",72,{"impact":82,"substance":226,"depth":227,"authority":83,"freshness":21,"relevant":22,"comment":228},21,17,"提出无监督多模态注意力聚类框架，融合无人机NDVI、IoT土壤与气象数据评估水稻健康，方法新颖且实验严谨，对精准农业田间监测有参考价值。",[230],{"name":222,"url":219},[27,192,28,232,30],"多模态融合",[234,235],"AgriMAC 水稻 多模态聚类","无人机 NDVI 水稻 病害监测","AgriMAC水稻多模态聚类-3009","10.22266\u002Fijies2026.1031.06",{"doi":237,"openalex_id":239,"authors":240,"venue":222,"cited_by_count":36,"oa_url":219,"card":247,"direction":251,"ingested_from":67},"W7213619014",[241,243,245],{"name":242,"orcid":9},"Nurfadhilah Mardianti Andini",{"name":244,"orcid":9},"Mike Yuliana",{"name":246,"orcid":9},"Moch. Zen Samsono Hadi",{"tldr":248,"method":249,"finding":250,"direction":251,"opportunity":252},"提出无监督多模态聚类框架AgriMAC，融合无人机NDVI、IoT土壤与气象数据评估水稻健康。","各模态自编码器编码，熵正则注意力融合，深度嵌入聚类，按生长阶段残差化。","聚类性能与仅IoT模型相当，但注意力权重可解释模态贡献并弱化生长阶段混淆。","智慧农业 \u002F 农业物联网","可探索注意力融合机制在更多作物与传感器组合下的泛化性，并引入时序动态聚类。","2026-09-20T23:30:08.419613Z",{"id":255,"title":256,"url":257,"summary":258,"summary_zh":9,"content":9,"source_name":259,"source_url":9,"published_at":260,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":261,"score_detail":262,"sources":264,"tags":266,"search_phrases":269,"slug":272,"view_count":36,"doi":9,"paper":273,"created_at":281},2855,"UAV无人机高光谱图像土壤盐度制图(湿度校正)——MDPI Agronomy","https:\u002F\u002Fwww.mdpi.com\u002F2073-4395\u002F16\u002F18\u002F1812","研究评估了6种光谱变换方法(原始反射率Ref、一阶导数FDR、PDS、OSC、FDR+PDS、FDR+OSC),结合3种机器学习算法(KNN、SVR、MLP)。进一步开发了集成这些基础学习者的Stacking集成模型,以提高湿度干扰下土壤盐度反演的精度。结果表明,Stacking模型在评估模型中达到最高的精度和稳定性。FDR+OSC-Stacking组合实现最佳验证性能,R²p=0.87,RMSEP=0.67 mS·cm⁻¹,RPD=2.93。FDR+OSC-Stacking组合成功应用于UAV高光谱图像,用于EC1:5的空间制图。来自吉林大学。","MDPI Agronomy","2026-09-15T00:00:00Z",75,{"impact":17,"substance":226,"depth":227,"authority":83,"freshness":124,"relevant":22,"comment":263},"方法组合新颖、验证指标扎实的无人机高光谱盐分制图研究，属细分领域实质进展，值得精选。",[265],{"name":259,"url":257},[27,267,29,30,268],"无人机","土壤盐渍化",[270,271],"土壤盐渍化 智慧农业 机器学习 无人机","土壤盐渍化 智慧农业","土壤盐渍化智慧农业机器学习无人机-2855",{"doi":9,"openalex_id":9,"authors":274,"venue":9,"cited_by_count":36,"oa_url":9,"card":275,"direction":65,"ingested_from":280},[],{"tldr":276,"method":277,"finding":278,"direction":65,"opportunity":279},"用无人机高光谱结合Stacking集成模型实现湿度干扰下的土壤盐度制图。","6种光谱变换与KNN、SVR、MLP及Stacking集成，基于UAV高光谱数据","FDR+OSC-Stacking最优，R²p=0.87、RMSEP=0.67 mS·cm⁻¹、RPD","可探索多时相\u002F多传感器融合与迁移学习，提升不同湿度与区域下盐度反演泛化性。","agent","2026-09-18T00:03:30.822732Z",{"id":283,"title":284,"url":285,"summary":286,"summary_zh":287,"content":9,"source_name":288,"source_url":285,"published_at":289,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":290,"score_detail":291,"sources":294,"tags":296,"search_phrases":298,"slug":301,"view_count":22,"doi":302,"paper":303,"created_at":323},2659,"Application of Remote Sensing and Machine Learning in Sustainable Agriculture","https:\u002F\u002Fdoi.org\u002F10.3390\u002Fsu18189472","Global agriculture is undergoing a period of profound transformation, driven by the need to increase food production in a context characterized by climate change, the degradation of natural resources and increasing pressure on agricultural ecosystems [...]","全球农业正经历一场深刻变革，其驱动力是在气候变化、自然资源退化以及农业生态系统压力日益增大的背景下提高粮食产量的需求。[...]","Sustainability","2026-09-16T00:00:00Z",77,{"impact":19,"substance":292,"depth":227,"authority":83,"freshness":21,"relevant":22,"comment":293},20,"发表于核心期刊的遥感与机器学习综述，方法视角新颖、时效性强，对智慧农业技术路线有参考价值，值得进入每日精选。",[295],{"name":288,"url":285},[27,192,29,297,30],"可持续农业",[299,300],"农业人工智能 可持续农业 智慧农业 机器学习","农业人工智能 可持续农业","农业人工智能可持续农业智慧农业机器学习-2659","10.3390\u002Fsu18189472",{"doi":302,"openalex_id":304,"authors":305,"venue":288,"cited_by_count":36,"oa_url":285,"card":318,"direction":65,"ingested_from":67},"W7213301703",[306,309,312,315],{"name":307,"orcid":308},"Mihai Valentin Herbei","https:\u002F\u002Forcid.org\u002F0000-0002-3884-3658",{"name":310,"orcid":311},"Ana-Cornelia Badea","https:\u002F\u002Forcid.org\u002F0000-0003-4521-5403",{"name":313,"orcid":314},"Aleksandar Ristić","https:\u002F\u002Forcid.org\u002F0000-0003-0979-3345",{"name":316,"orcid":317},"Paul Sestraș","https:\u002F\u002Forcid.org\u002F0000-0002-8554-0924",{"tldr":319,"method":320,"finding":321,"direction":65,"opportunity":322},"综述遥感与机器学习在可持续农业中的应用现状与前景。","综述遥感数据与机器学习方法在农业中的应用。","遥感结合机器学习可提升农业监测与可持续管理能力。","可探索多源遥感与可解释机器学习融合，用于小农户精准决策与碳核算。","2026-09-16T23:30:28.640661Z"]