[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-2996":3,"related-2996":86},{"id":4,"title":5,"url":6,"summary":7,"summary_zh":8,"content":8,"source_name":9,"source_url":8,"published_at":10,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":14,"score_detail":15,"sources":23,"tags":25,"primary_topic":31,"secondary_topics":37,"classification":44,"search_phrases":70,"slug":73,"view_count":74,"doi":8,"paper":75,"created_at":85},2996,"基于高光谱遥感和跨区域迁移学习的冬小麦叶绿素含量评估","https:\u002F\u002Fwww.toutiao.com\u002Farticle\u002F7686128266574430720","甘肃农业大学李怡悦等与中国农业科学院作物科学研究所肖永贵、孟亚雄研究员团队合作，针对高光谱遥感跨区域叶绿素评估中因\"域偏移\"导致的模型泛化性能下降问题，提出一种稳健自适应迁移学习框架（RATL），通过自适应特征选择、特征权重调整与域适应训练3模块协同实现跨区域知识有效迁移。基于2023—2024年冬小麦灌浆后期新乡与周口两地共1491份冠层高光谱与叶片SPAD数据，设置4种场景比较RATL与直接迁移、迁移成分分析和相关对齐之间的精度。",null,"智慧农业(中英文)2026,8(4):47-59","2026-09-16T13:43:00Z","论文",10,false,76,{"impact":16,"substance":17,"depth":18,"authority":19,"freshness":20,"relevant":21,"comment":22},16,22,18,14,6,1,"方法新颖、数据扎实的跨区域叶绿素遥感评估研究，对智慧农业遥感应用有参考价值。",[24],{"name":9,"url":6},[26,27,28,29,30],"智慧农业","遥感","冬小麦","迁移学习","叶绿素监测",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},"A2","wheat","小麦","\u002Ftopic\u002Ffield-crops\u002Fwheat","A",[38],{"code":39,"slug":40,"name":41,"path":42,"parent_code":43},"I3","remote-sensing","遥感与农情监测","\u002Ftopic\u002Fdigital-agriculture\u002Fremote-sensing","I",{"status":45,"locked":13,"version":46,"method":47,"evidence":48},"classified","2026-10-22.2","rules",{"candidates":49,"agricultural_evidence":52},[50,56,60,65],{"code":32,"score":51,"supported":52,"title":53,"abstract":54,"tags":55},12,true,[34],[34],[28],{"code":39,"score":51,"supported":52,"title":57,"abstract":58,"tags":59},[27],[27],[27],{"code":61,"score":21,"supported":13,"title":62,"abstract":63,"tags":64},"I1",[],[],[26],{"code":66,"score":21,"supported":13,"title":67,"abstract":68,"tags":69},"I2",[],[],[29],[71,72],"冬小麦 叶绿素 高光谱遥感","跨区域迁移学习 冬小麦","冬小麦叶绿素高光谱遥感-2996",3,{"doi":8,"openalex_id":8,"authors":76,"venue":8,"cited_by_count":77,"oa_url":8,"card":78,"direction":82,"ingested_from":84},[],0,{"tldr":79,"method":80,"finding":81,"direction":82,"opportunity":83},"提出RATL迁移学习框架，解决高光谱遥感跨区域冬小麦叶绿素评估的域偏移问题。","基于新乡与周口1491份冠层高光谱与SPAD数据，对比RATL与直接迁移、TCA","RATL通过自适应特征选择、权重调整与域适应训练，有效提升跨区域叶绿素评估泛化精度。","农业遥感与作物表型","可探索多作物、多生育期及不同传感器间的迁移学习，构建通用跨域叶绿素评估模型。","agent","2026-09-20T00:03:07.685193Z",{"total":20,"page":21,"page_size":20,"items":87},[88,167,237,306,389,485],{"id":89,"title":90,"url":91,"summary":92,"summary_zh":93,"content":8,"source_name":94,"source_url":91,"published_at":95,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":96,"score_detail":97,"sources":100,"tags":102,"primary_topic":104,"secondary_topics":105,"classification":111,"search_phrases":142,"slug":145,"view_count":146,"doi":147,"paper":148,"created_at":166},4433,"PROSAIL-DNN-TL: A Hyperspectral Inversion Framework for Winter Wheat Leaf Chlorophyll Content Based on Fine-Tuned Transfer Learning","https:\u002F\u002Fdoi.org\u002F10.3390\u002Fagronomy16191971","Accurate and efficient estimation of leaf chlorophyll content (LCC) is essential for crop physiological monitoring and precision agriculture. Current remote sensing approaches for LCC estimation mainly rely on machine learning and deep learning methods, which exhibit strong nonlinear modelling capabilities but require sufficient and representative samples. Radiative transfer models (RTMs) can simulate canopy spectral response based on physical mechanisms and generate large-scale datasets, thereby overcoming the limitations of purely data-driven approaches. However, their applications under diverse environmental conditions remain challenging due to uncertainties caused by model parameterization and discrepancies between simulated and observed data. This study proposed a hybrid framework, PROSAIL-DNN-TL, integrating the PROSAIL RTM with full-network fine-tuned transfer learning for winter wheat LCC estimation. First, 85,680 simulated samples generated by PROSAIL were used to pre-train a deep neural network (DNN), and field observations (n = 270) were subsequently used for full-network fine-tuning. A conventional DNN model trained only with field data and machine learning models based on different feature selection strategies were established for comparison. The results showed that: (1) UVE-selected bands were distributed across visible, red-edge, and near-infrared regions; CARS retained more bands in the visible region, whereas SPA mainly selected near-infrared bands; (2) RR and SVR achieved higher accuracy than KNN, with the full-spectrum and CARS feature sets showing superior performance and SPA showing the lowest accuracy; (3) PROSAIL-DNN-TL achieved reliable LCC estimation performance, with an average test R2 of 0.79 across 500 repeated experiments, and showed improved performance compared with conventional machine learning models; (4) PROSAIL-DNN-TL achieved reliable predictive performance when evaluated using datasets collected during the 2024 and 2025 growing seasons, with R2 values of 0.75 and 0.92, respectively. Overall, the proposed hybrid framework integrates physical knowledge from RTMs with data-driven modelling through transfer learning, providing a potential approach for crop chlorophyll monitoring. However, the SPAD-to-LCC conversion used in this study was not specifically calibrated for the instrument and crop, and therefore, the absolute LCC values should be interpreted with caution.","准确、高效地估算叶片叶绿素含量（leaf chlorophyll content，LCC）对作物生理监测和精准农业至关重要。当前遥感估算LCC的方法主要依赖机器学习和深度学习方法，这类方法具有较强的非线性建模能力，但需要充足且具有代表性的样本。辐射传输模型（radiative transfer models，RTMs）能够基于物理机制模拟冠层光谱响应并生成大规模数据集，从而克服纯数据驱动方法的局限性。然而，由于模型参数化带来的不确定性以及模拟数据与观测数据之间的差异，其在不同环境条件下的应用仍面临挑战。本研究提出了一种混合框架PROSAIL-DNN-TL，将PROSAIL辐射传输模型与全网络微调迁移学习相结合，用于冬小麦LCC估算。首先，利用PROSAIL生成的85 680个模拟样本预训练深度神经网络（deep neural network，DNN），随后利用田间观测数据（n = 270）进行全网络微调。同时建立了仅使用田间数据训练的传统DNN模型以及基于不同特征选择策略的机器学习模型作为对比。结果表明：（1）UVE选择的波段分布在可见光、红边和近红外区域；CARS在可见光区域保留了更多波段，而SPA主要选择近红外波段；（2）RR和SVR的精度高于KNN，全光谱和CARS特征集表现较优，SPA精度最低；（3）PROSAIL-DNN-TL取得了可靠的LCC估算性能，在500次重复实验中平均测试R²为0.79，且相比传统机器学习模型表现出更优的性能；（4）利用2024年和2025年生长季采集的数据集进行评估时，PROSAIL-DNN-TL均表现出可靠的预测性能，R²分别为0.75和0.92。总体而言，所提出的混合框架通过迁移学习将辐射传输模型的物理知识与数据驱动建模相结合，为作物叶绿素监测提供了一种潜在途径。然而，本研究中使用的SPAD到LCC的转换并未针对仪器和作物进行专门校准，因此绝对LCC值应谨慎解读。","Agronomy","2026-10-07T00:00:00Z",81,{"impact":18,"substance":17,"depth":18,"authority":19,"freshness":98,"relevant":21,"comment":99},9,"将PROSAIL辐射传输模型与深度神经网络迁移学习结合，用8万余模拟样本预训练、270个田间样本微调，冬小麦叶绿素反演精度可靠，方法新颖且数据规模扎实，对作物生理监测与精准农业有参考价值。",[101],{"name":94,"url":91},[26,103,28,29,30],"农业遥感",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[106,107],{"code":39,"slug":40,"name":41,"path":42,"parent_code":43},{"code":66,"slug":108,"name":109,"path":110,"parent_code":43},"agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai",{"status":45,"locked":13,"version":46,"method":47,"evidence":112},{"candidates":113,"agricultural_evidence":52},[114,120,127,135],{"code":32,"score":115,"supported":52,"title":116,"abstract":118,"tags":119},26,[117,33],"winter wheat",[34,117,33],[28],{"code":39,"score":121,"supported":52,"title":122,"abstract":124,"tags":126},15,[123],"hyperspectral",[27,125],"remote sensing",[103],{"code":66,"score":12,"supported":52,"title":128,"abstract":129,"tags":134},[],[130,131,132,133],"深度学习","机器学习","machine learning","deep learning",[29],{"code":61,"score":136,"supported":52,"title":137,"abstract":138,"tags":141},7,[],[139,140],"精准农业","precision agriculture",[26],[143,144],"PROSAIL 冬小麦 叶绿素","高光谱 叶绿素 反演","PROSAIL冬小麦叶绿素-4433",5,"10.3390\u002Fagronomy16191971",{"doi":147,"openalex_id":149,"authors":150,"venue":94,"cited_by_count":77,"oa_url":91,"card":160,"direction":82,"ingested_from":165},"W7220844556",[151,153,155,158],{"name":152,"orcid":8},"Xueqing Zhu",{"name":154,"orcid":8},"Ying Nian",{"name":156,"orcid":157},"Jikai Liu","https:\u002F\u002Forcid.org\u002F0000-0003-2365-5221",{"name":159,"orcid":8},"Xinwei Li",{"tldr":161,"method":162,"finding":163,"direction":82,"opportunity":164},"提出PROSAIL-DNN-TL框架，用PROSAIL模拟数据预训练DNN并全网络微调，反演冬小麦叶","PROSAIL辐射传输模型生成85680样本预训练DNN，270个田间样本全网络","500次重复实验平均测试R²为0.79，2024和2025年生长季验证R²分别为0.75和0.92，","SPAD到LCC转换未针对仪器和作物标定，可研究跨传感器、跨作物、跨生长季的迁移学习泛化与绝对叶绿素标定。","openalex","2026-10-08T23:30:29.383957Z",{"id":168,"title":169,"url":170,"summary":171,"summary_zh":172,"content":8,"source_name":173,"source_url":170,"published_at":174,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":175,"score_detail":176,"sources":180,"tags":182,"primary_topic":185,"secondary_topics":186,"classification":189,"search_phrases":215,"slug":218,"view_count":74,"doi":219,"paper":220,"created_at":236},1996,"An Interpretable BO-TCBDA Deep Learning Framework for Winter Wheat Yield Estimation Using Multi-Source Remote Sensing Data","https:\u002F\u002Fdoi.org\u002F10.3390\u002Frs18183061","Reliable crop yield estimation is fundamental to food security and efficient agricultural management. However, current deep learning models still face limitations in selecting and integrating multi-source features, and their high predictive accuracy is often accompanied by limited interpretability. This study introduces a Bayesian Optimization–Temporal Convolutional Network–Bidirectional Long Short-Term Memory–Dual Attention (BO-TCBDA) deep learning framework for winter wheat yield estimation. Using Henan Province, China, as the study area, county-level winter wheat yield from 2013 to 2022 was estimated using the Enhanced Vegetation Index (EVI), Leaf Area Index (LAI), Solar-Induced Chlorophyll Fluorescence (SIF), and climate data. The proposed model was compared with five commonly used machine learning and deep learning models. BO-TCBDA achieved the best performance, with an R2 of 0.823 and an RMSE of 561.26 kg\u002Fha. SIF improved the predictive performance of all models, with statistically significant gains observed in the deep learning models. The dual-attention mechanism provided interpretable insights by revealing relatively balanced contributions among the input features and highlighting the grain-filling stage through temporal attention. Furthermore, SHAP-based cross-validation analysis identified T12, corresponding to the latter part of the jointing stage, as the period with the highest contribution to yield prediction. The model also achieved an R2 of approximately 0.80 about 25 days before harvest. Overall, BO-TCBDA provides an accurate and interpretable approach for county-level winter wheat yield estimation and supports regional food security assessments and precision agriculture.","可靠的作物产量估算对于粮食安全和高效农业管理至关重要。然而，当前的深度学习模型在多源特征的选择与整合方面仍存在局限，且高预测精度往往伴随着有限的可解释性。本研究提出了一种基于贝叶斯优化–时间卷积网络–双向长短期记忆–双重注意力（BO-TCBDA）的深度学习框架，用于冬小麦产量估算。以中国河南省为研究区域，利用增强型植被指数（EVI）、叶面积指数（LAI）、太阳诱导叶绿素荧光（SIF）及气候数据，对2013至2022年县级冬小麦产量进行了估算。将所提模型与五种常用的机器学习和深度学习模型进行了比较。BO-TCBDA取得了最佳性能，其决定系数（R²）为0.823，均方根误差（RMSE）为561.26千克\u002F公顷。SIF提升了所有模型的预测性能，其中在深度学习模型中观察到了统计学上显著的增益。双重注意力机制通过揭示输入特征间相对均衡的贡献，并借助时间注意力突出灌浆期，提供了可解释性的见解。此外，基于SHAP的交叉验证分析识别出T12时段（对应拔节期后期）对产量预测的贡献最大。该模型在收获前约25天时，R²亦达到约0.80。总体而言，BO-TCBDA为县级冬小麦产量估算提供了一种准确且可解释的方法，并支持区域粮食安全评估和精准农业实践。","Remote Sensing","2026-09-08T00:00:00Z",73,{"impact":18,"substance":17,"depth":18,"authority":177,"freshness":178,"relevant":21,"comment":179},13,2,"提出可解释的深度学习框架，结合多源遥感数据提升冬小麦估产精度与可解释性，对精准农业有实质贡献。",[181],{"name":173,"url":170},[26,183,27,184,28],"农业人工智能","产量估算",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[187,188],{"code":66,"slug":108,"name":109,"path":110,"parent_code":43},{"code":39,"slug":40,"name":41,"path":42,"parent_code":43},{"status":45,"locked":13,"version":46,"method":47,"evidence":190},{"candidates":191,"agricultural_evidence":52},[192,196,200,204,208],{"code":32,"score":115,"supported":52,"title":193,"abstract":194,"tags":195},[117,33],[34,117,33],[28],{"code":66,"score":18,"supported":52,"title":197,"abstract":198,"tags":199},[133],[130,131,132,133],[183],{"code":39,"score":12,"supported":52,"title":201,"abstract":202,"tags":203},[125],[],[184,27],{"code":61,"score":136,"supported":52,"title":205,"abstract":206,"tags":207},[],[139,140],[26],{"code":209,"score":20,"supported":52,"title":210,"abstract":211,"tags":214},"A8",[],[212,213],"粮食安全","food security",[],[216,217],"农业人工智能 产量估算 智慧农业 冬小麦","农业人工智能 产量估算","农业人工智能产量估算智慧农业冬小麦-1996","10.3390\u002Frs18183061",{"doi":219,"openalex_id":221,"authors":222,"venue":173,"cited_by_count":77,"oa_url":170,"card":230,"direction":82,"ingested_from":165},"W7211938917",[223,225,227],{"name":224,"orcid":8},"Anqi Xue",{"name":226,"orcid":8},"Shufang Tian",{"name":228,"orcid":229},"Tingyan Fu","https:\u002F\u002Forcid.org\u002F0000-0003-4207-4211",{"tldr":231,"method":232,"finding":233,"direction":234,"opportunity":235},"提出BO-TCBDA深度学习框架，融合多源遥感数据估算冬小麦产量，兼具高精度与可解释性。","贝叶斯优化、TCN、BiLSTM、双注意力机制，结合EVI、LAI、SIF及气候","BO-TCBDA性能最优（R²=0.823），SIF提升预测，拔节后期贡献最大，收获前25天可预测。","农业人工智能与决策模型","可探索将双注意力与SHAP结合用于其他作物或区域，或开发实时预警系统，提升模型泛化与实用性。","2026-09-09T23:30:22.411579Z",{"id":238,"title":239,"url":240,"summary":241,"summary_zh":242,"content":8,"source_name":243,"source_url":240,"published_at":244,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":245,"score_detail":246,"sources":249,"tags":251,"primary_topic":255,"secondary_topics":256,"classification":258,"search_phrases":277,"slug":280,"view_count":77,"doi":281,"paper":282,"created_at":305},5367,"Integration of multi-source UAV remote sensing features and derived features for winter wheat aboveground biomass estimation","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.atech.2026.102656","Accurate estimation of winter wheat ( Triticum aestivum L.) aboveground biomass (AGB) is important for crop growth monitoring and field management. However, unmanned aerial vehicle (UAV)-based AGB estimation remains challenging due to variations in canopy conditions across growth stages, and different remote sensing features can capture complementary characteristics related to AGB estimation. This study integrated multi-source UAV-derived features, including vegetation indices (VIs), texture features (TFs), light detection and ranging (LiDAR)-derived features, and derived features (DFs) generated from the original feature sets, with four machine learning algorithms for winter wheat AGB estimation. Results showed that TFs provided better prediction performance than other single-source feature sets. Compared with single-source feature sets, the integration of VIs, TFs, LiDAR-derived features, and DFs further improved AGB estimation, and the random forest (RF) model achieved the best performance among the evaluated models (R² = 0.923, RMSE = 1.731 Mg ha⁻¹). Incremental R² analysis based on a stage-only baseline model showed that the RF model explained additional variation in AGB beyond growth-stage information, with an incremental R² of 0.479. Stage-specific evaluation indicated that model performance varied among growth stages, with the highest accuracy observed at the booting stage (R² = 0.632, RMSE = 1.425 Mg ha⁻¹). Feature contribution analysis based on the full-feature fusion model using all growth-stage data identified canopy coverage and several DFs as important predictors, while stage-specific analyses showed differences in dominant predictors across growth stages. The integration of multi-source UAV-derived features and DFs with machine learning provides an effective approach for winter wheat AGB estimation, while model performance remains strongly dependent on growth stage.","准确估算冬小麦（Triticum aestivum L.）地上生物量（aboveground biomass, AGB）对作物生长监测和田间管理具有重要意义。然而，由于不同生育期冠层条件存在差异，且不同遥感特征能够捕捉与AGB估算相关的互补信息，基于无人机（unmanned aerial vehicle, UAV）的AGB估算仍面临挑战。本研究将多源无人机衍生特征，包括植被指数（vegetation indices, VIs）、纹理特征（texture features, TFs）、激光雷达（light detection and ranging, LiDAR）衍生特征以及由原始特征集生成的衍生特征（derived features, DFs），与四种机器学习算法相结合，用于冬小麦AGB估算。结果表明，TFs的预测性能优于其他单一来源特征集。与单一来源特征集相比，融合VIs、TFs、LiDAR衍生特征和DFs进一步提高了AGB估算精度，其中随机森林（random forest, RF）模型在所评估的模型中表现最佳（R² = 0.923，RMSE = 1.731 Mg ha⁻¹）。基于仅含生育期信息的基线模型进行的增量R²分析表明，RF模型在生育期信息之外解释了AGB的额外变异，增量R²为0.479。分生育期评估表明，模型性能在不同生育期之间存在差异，其中在孕穗期观察到最高精度（R² = 0.632，RMSE = 1.425 Mg ha⁻¹）。基于使用全部生育期数据的全特征融合模型进行的特征贡献分析识别出冠层覆盖度和若干DFs为重要预测因子，而分生育期分析则表明不同生育期的主导预测因子存在差异。将多源无人机衍生特征和DFs与机器学习相结合，为冬小麦AGB估算提供了一种有效方法，但模型性能仍强烈依赖于生育期。","Smart Agricultural Technology","2026-10-17T00:00:00Z",77,{"impact":16,"substance":17,"depth":18,"authority":177,"freshness":247,"relevant":21,"comment":248},8,"该研究融合无人机多源遥感特征与衍生特征，用随机森林估算冬小麦地上生物量，方法新颖、结论可靠，对作物长势监测有参考价值。",[250],{"name":243,"url":240},[26,252,253,28,254],"无人机","遥感监测","生物量估算",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[257],{"code":39,"slug":40,"name":41,"path":42,"parent_code":43},{"status":45,"locked":13,"version":46,"method":47,"evidence":259},{"candidates":260,"agricultural_evidence":52},[261,265,269,273],{"code":32,"score":115,"supported":52,"title":262,"abstract":263,"tags":264},[117,33],[34,117,33],[28],{"code":39,"score":16,"supported":52,"title":266,"abstract":267,"tags":268},[125],[27,125],[254,253],{"code":66,"score":20,"supported":52,"title":270,"abstract":271,"tags":272},[],[131,132],[],{"code":61,"score":21,"supported":13,"title":274,"abstract":275,"tags":276},[],[],[26],[278,279],"无人机 多源遥感 冬小麦 生物量","冬小麦 地上生物量 随机森林","无人机多源遥感冬小麦生物量-5367","10.1016\u002Fj.atech.2026.102656",{"doi":281,"openalex_id":283,"authors":284,"venue":243,"cited_by_count":77,"oa_url":240,"card":300,"direction":82,"ingested_from":165},"W7222806066",[285,287,289,291,294,296,298],{"name":286,"orcid":8},"Guangtao Wang",{"name":288,"orcid":8},"Sumei Yao",{"name":290,"orcid":8},"Yanzheng Xie",{"name":292,"orcid":293},"Anzhen Qin","https:\u002F\u002Forcid.org\u002F0000-0002-2550-0015",{"name":295,"orcid":8},"Xiangdong Chen",{"name":297,"orcid":8},"Yuhao Zhang",{"name":299,"orcid":8},"Lina Jiang",{"tldr":301,"method":302,"finding":303,"direction":82,"opportunity":304},"融合无人机多源遥感特征与衍生特征，用机器学习估算冬小麦地上生物量。","无人机植被指数、纹理、LiDAR特征及衍生特征结合四种机器学习算法建模。","多源特征融合优于单一特征，随机森林最优（R²=0.923），精度随生育期变化。","可探索生育期自适应建模与特征选择，提升跨期生物量估算的稳健性与可迁移性。","2026-10-20T23:30:03.106267Z",{"id":307,"title":308,"url":309,"summary":310,"summary_zh":311,"content":8,"source_name":312,"source_url":309,"published_at":313,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":314,"score_detail":315,"sources":319,"tags":321,"primary_topic":323,"secondary_topics":324,"classification":327,"search_phrases":357,"slug":360,"view_count":21,"doi":361,"paper":362,"created_at":388},3863,"Species identification of crops and weeds in wheat fields using hyperspectral data and machine learning algorithms","https:\u002F\u002Fdoi.org\u002F10.31489\u002F2026feb3\u002F137-151","Based on establishing the relationship between the morphological and anatomical properties of plant tissues and the recognition efficiency of the machine learning algorithm, a comprehensive assessment of the limitations of ML algorithms (using Random Forest as an example) in the multiclass identification of crops and weeds in the agroecosystems of northeastern Kazakhstan was carried out for the first time. The study substantiates the principal biological causes manifested in the optical overlap of classes in the hyperspectral feature space. Data collection was performed with a hyperspectral sensor in the VIS-NIR range. The analysis of spectral profiles was carried out based on variance distribution assessment using principal component analysis and the construction of reflectance curves. Classification of the extracted data was performed using the Random Forest algorithm. Spring wheat (Triticum aestivum) crops at different phenological phases and five dominant weed species (Euphorbia virgata, Descurainia sophia, Vicia hirsuta, Fallopia convolvulus, Echinochloa crus-galli) were selected as research objects. The algorithm demonstrated \"crop-weed\" segregation efficiency with an overall classification accuracy on test data of 72.94%. High identification accuracy was revealed for wheat (MCC=0.994) due to its spectral stability and pronounced red edge features (680-730 nm). Reduced accuracy in intra-class weed separation is driven by species-specific anatomy: thick cuticles in Euphorbia virgata increase reflectance, complex structure in Vicia hirsuta causes spectral dispersion, and mesophyll similarity between Echinochloa crus-galli and grasses leads to optical overlap with crop features. Thus, the model enables reliable crop-weed differentiation, a key requirement for precision agriculture and targeted agrochemical application.","在建立植物组织形态解剖特性与机器学习算法识别效率之间关系的基础上，首次对机器学习算法（以随机森林为例）在哈萨克斯坦东北部农业生态系统中作物与杂草多类识别方面的局限性进行了综合评估。研究证实了在高光谱特征空间中表现为类别光谱重叠的主要生物学成因。数据采集使用VIS-NIR波段的高光谱传感器完成。光谱剖面分析基于主成分分析的方差分布评估及反射率曲线构建进行。提取数据的分类采用随机森林算法完成。选取不同物候期的春小麦（Triticum aestivum）作物及五种优势杂草物种（Euphorbia virgata、Descurainia sophia、Vicia hirsuta、Fallopia convolvulus、Echinochloa crus-galli）作为研究对象。该算法在测试数据上实现了“作物-杂草”分离，总体分类精度为72.94%。小麦因其光谱稳定性及显著的红边特征（680-730 nm）而表现出较高的识别精度（MCC=0.994）。杂草类内分离精度降低由物种特异性解剖结构所致：Euphorbia virgata的厚角质层增加了反射率，Vicia hirsuta的复杂结构导致光谱离散，而Echinochloa crus-galli与禾本科植物之间的叶肉相似性导致其与作物特征发生光谱重叠。因此，该模型能够实现可靠的作物-杂草区分，这是精准农业和靶向农化施用的关键要求。","Fundamental and Experimental Biology","2026-09-29T00:00:00Z",70,{"impact":51,"substance":316,"depth":317,"authority":51,"freshness":98,"relevant":21,"comment":318},20,17,"高光谱结合机器学习实现麦田作物与杂草识别，方法新颖、结论具体，对精准农业有参考价值，但属细分领域研究，影响范围有限。",[320],{"name":312,"url":309},[26,131,34,322,27],"杂草识别",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[325,326],{"code":66,"slug":108,"name":109,"path":110,"parent_code":43},{"code":39,"slug":40,"name":41,"path":42,"parent_code":43},{"status":45,"locked":13,"version":46,"method":47,"evidence":328},{"candidates":329,"agricultural_evidence":52},[330,334,338,342,346,352],{"code":32,"score":121,"supported":52,"title":331,"abstract":332,"tags":333},[33],[34,33],[34],{"code":66,"score":121,"supported":52,"title":335,"abstract":336,"tags":337},[132],[131,132],[131],{"code":39,"score":51,"supported":52,"title":339,"abstract":340,"tags":341},[123],[123],[27],{"code":61,"score":136,"supported":52,"title":343,"abstract":344,"tags":345},[],[139,140],[26],{"code":347,"score":74,"supported":52,"title":348,"abstract":349,"tags":351},"H6",[],[350],"农业生态",[],{"code":353,"score":21,"supported":13,"title":354,"abstract":355,"tags":356},"F4",[],[],[322],[358,359],"高光谱 小麦 杂草 识别","随机森林 作物 杂草 分类","高光谱小麦杂草识别-3863","10.31489\u002F2026feb3\u002F137-151",{"doi":361,"openalex_id":363,"authors":364,"venue":312,"cited_by_count":77,"oa_url":8,"card":383,"direction":234,"ingested_from":165},"W7214771122",[365,368,371,374,377,380],{"name":366,"orcid":367},"Rimma Meyramovna Ualiyeva","https:\u002F\u002Forcid.org\u002F0000-0003-3551-5007",{"name":369,"orcid":370},"Mariya M. Kaverina","https:\u002F\u002Forcid.org\u002F0000-0002-3838-7656",{"name":372,"orcid":373},"Александра Викторовна Осипова","https:\u002F\u002Forcid.org\u002F0000-0002-3947-5252",{"name":375,"orcid":376},"E. Koppayev","https:\u002F\u002Forcid.org\u002F0009-0004-0476-9017",{"name":378,"orcid":379},"М. А. Чайзабекова","https:\u002F\u002Forcid.org\u002F0009-0004-0468-956X",{"name":381,"orcid":382},"Sayan Berikovich Zhangazin","https:\u002F\u002Forcid.org\u002F0000-0002-5813-8331",{"tldr":384,"method":385,"finding":386,"direction":82,"opportunity":387},"利用高光谱数据与随机森林识别哈萨克斯坦东北部麦田作物与杂草，并分析误判的生物学原因。","VIS-NIR高光谱传感器采集数据，PCA分析光谱，随机森林分类春小麦与五种杂草","作物-杂草总体分类精度72.94%，小麦识别精度高（MCC=0.994），杂草种间分离受解剖结构影响","可探索融合形态解剖特征与深度学习，提升杂草种间分类精度，服务精准施药。","2026-09-30T23:30:54.952093Z",{"id":390,"title":391,"url":392,"summary":393,"summary_zh":394,"content":8,"source_name":395,"source_url":392,"published_at":396,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":397,"score_detail":398,"sources":400,"tags":402,"primary_topic":404,"secondary_topics":405,"classification":407,"search_phrases":432,"slug":435,"view_count":74,"doi":436,"paper":437,"created_at":484},3354,"Optimizing canopy nitrogen retrieval in wheat from high-resolution UAV hyperspectral data using RTM-based Gaussian process regression","https:\u002F\u002Fdoi.org\u002F10.1080\u002F01904167.2026.2736627","The canopy nitrogen content (CNC) is one of the vital crop health indicators directly influencing the crop growth, yield, and nutritional status. Unmanned aerial vehicles (UAVs) that acquired hyperspectral imagery opened a new path for nondestructive, accurate, and timely estimation of CNC, facilitating the adoption of smart agro practices in precision agriculture for site-specific nutrient management. With the purpose of processing such hyperspectral data into CNC maps, two hybrid retrieval approaches, protein-based (CNCprot) and chlorophyll-based (CNCchl) models, were built by combining a radiative transfer model with a machine learning Gaussian process regression (GPR) algorithm. The GPR retrieval models were trained using synthetic spectral data from the PROSAIL-PRO model, and the sampling was optimized using the Euclidean distance-based diversity (EBD) active learning (AL) technique. The CNC models were applied to a UAV-borne hyperspectral image in the spectral range of 400-1000 nm with an ultrahigh spatial resolution of 4 cm, acquired over the experimental wheat field of Indian Council of Agricultural Research (ICAR) -Indian Agricultural Research Institute (IARI), New Delhi, India. The GPR models yielded good prediction accuracies with R2 values of 0.74 and 0.56 and NRMSE of 13.62 and 23.82% for CNCchl and CNCprot, respectively. The consistent prediction accuracies and low associated uncertainties revealed that CNC was more accurately mapped using chlorophyll-based CNC retrieval applied to a Vis-NIR-based hyperspectral data. These promising results demonstrated the possibility of applying hybrid GPR models at ultrahigh resolution and the operational viability of UAV datasets for regular CNC monitoring.","冠层氮含量（CNC）是直接影响作物生长、产量和营养状况的重要作物健康指标之一。无人机（UAV）获取的高光谱影像为无损、精准、及时地估算CNC开辟了新途径，有助于在精准农业中采用智能农业实践进行定点养分管理。为了将此类高光谱数据处理为CNC分布图，本研究通过将辐射传输模型与机器学习高斯过程回归（GPR）算法相结合，构建了两种混合反演方法：基于蛋白质的模型（CNCprot）和基于叶绿素的模型（CNCchl）。GPR反演模型使用PROSAIL-PRO模型生成的合成光谱数据进行训练，并采用基于欧氏距离的多样性（EBD）主动学习（AL）技术优化采样。CNC模型被应用于一幅无人机载高光谱影像，光谱范围为400-1000 nm，空间分辨率高达4 cm，该影像获取自印度农业研究理事会（ICAR）-印度农业研究所（IARI）位于印度新德里的实验麦田。GPR模型取得了良好的预测精度，CNCchl和CNCprot的R²值分别为0.74和0.56，NRMSE分别为13.62%和23.82%。一致的预测精度和较低的相关不确定性表明，将基于叶绿素的CNC反演应用于基于可见光-近红外（Vis-NIR）的高光谱数据能够更准确地制图CNC。这些有前景的结果证明了在超高分辨率下应用混合GPR模型的可能性，以及无人机数据集用于常规CNC监测的可行性。","Journal of Plant Nutrition","2026-09-23T00:00:00Z",79,{"impact":18,"substance":17,"depth":18,"authority":177,"freshness":247,"relevant":21,"comment":399},"将辐射传输模型与高斯过程回归结合，用4厘米超高分辨率无人机高光谱实现小麦冠层氮素精准反演，方法新颖、结论可靠，对精准农业养分管理有实操参考价值。",[401],{"name":395,"url":392},[26,252,34,27,403],"氮素监测",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[406],{"code":39,"slug":40,"name":41,"path":42,"parent_code":43},{"status":45,"locked":13,"version":46,"method":47,"evidence":408},{"candidates":409,"agricultural_evidence":52},[410,414,418,422,426],{"code":32,"score":51,"supported":52,"title":411,"abstract":412,"tags":413},[33],[33],[34],{"code":39,"score":51,"supported":52,"title":415,"abstract":416,"tags":417},[123],[123],[27],{"code":61,"score":136,"supported":52,"title":419,"abstract":420,"tags":421},[],[139,140],[26],{"code":66,"score":20,"supported":52,"title":423,"abstract":424,"tags":425},[],[131,132],[],{"code":427,"score":74,"supported":52,"title":428,"abstract":429,"tags":431},"F3",[],[430],"养分管理",[],[433,434],"无人机 高光谱 小麦 氮素","PROSAIL 高斯过程回归 小麦","无人机高光谱小麦氮素-3354","10.1080\u002F01904167.2026.2736627",{"doi":436,"openalex_id":438,"authors":439,"venue":395,"cited_by_count":77,"oa_url":8,"card":478,"direction":483,"ingested_from":165},"W7214104831",[440,442,444,447,450,452,454,457,460,463,465,468,470,472,475],{"name":441,"orcid":8},"R. G. Rejith",{"name":443,"orcid":8},"Rabi N. Sahoo",{"name":445,"orcid":446},"Shalini Gakhar","https:\u002F\u002Forcid.org\u002F0000-0001-5717-1714",{"name":448,"orcid":449},"Jochem Verrelst","https:\u002F\u002Forcid.org\u002F0000-0002-6313-2081",{"name":451,"orcid":8},"Amrita Bhandari",{"name":453,"orcid":8},"Tarun Kondraju",{"name":455,"orcid":456},"Rajeev Ranjan","https:\u002F\u002Forcid.org\u002F0000-0003-2233-9147",{"name":458,"orcid":459},"Mahesh C. Meena","https:\u002F\u002Forcid.org\u002F0000-0001-5386-883X",{"name":461,"orcid":462},"Abir Dey","https:\u002F\u002Forcid.org\u002F0000-0002-5009-9518",{"name":464,"orcid":8},"Joydeep Mukherjee",{"name":466,"orcid":467},"Sudhir Kumar","https:\u002F\u002Forcid.org\u002F0000-0002-1089-7435",{"name":469,"orcid":8},"Mahesh Kumar",{"name":471,"orcid":8},"Raju Dhandapani",{"name":473,"orcid":474},"Anchal Dass","https:\u002F\u002Forcid.org\u002F0000-0003-3909-1803",{"name":476,"orcid":477},"Viswanathan Chinnusamy","https:\u002F\u002Forcid.org\u002F0000-0003-2174-9064",{"tldr":479,"method":480,"finding":481,"direction":82,"opportunity":482},"用PROSAIL-PRO与高斯过程回归从无人机高光谱数据反演小麦冠层氮含量。","PROSAIL-PRO合成光谱训练GPR，EBD主动学习采样，4cm无人机高光谱","叶绿素基模型精度更高，R²达0.74，NRMSE为13.62%，优于蛋白基模型。","可探索多生育期、多品种迁移能力及主动学习采样策略对反演精度的提升空间。","智慧农业 \u002F 农业物联网","2026-09-24T23:30:10.311254Z",{"id":486,"title":487,"url":488,"summary":489,"summary_zh":490,"content":8,"source_name":243,"source_url":488,"published_at":491,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":492,"score_detail":493,"sources":495,"tags":497,"primary_topic":500,"secondary_topics":501,"classification":503,"search_phrases":524,"slug":527,"view_count":528,"doi":529,"paper":530,"created_at":549},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。","2026-09-22T00:00:00Z",74,{"impact":51,"substance":17,"depth":18,"authority":177,"freshness":98,"relevant":21,"comment":494},"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。",[496],{"name":243,"url":488},[26,498,131,499,27],"产量预测","小麦育种",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[502],{"code":39,"slug":40,"name":41,"path":42,"parent_code":43},{"status":45,"locked":13,"version":46,"method":47,"evidence":504},{"candidates":505,"agricultural_evidence":52},[506,510,516,520],{"code":32,"score":121,"supported":52,"title":507,"abstract":508,"tags":509},[33],[34,33],[499],{"code":39,"score":247,"supported":52,"title":511,"abstract":512,"tags":515},[],[513,514],"无人机多光谱","multispectral",[498,27],{"code":61,"score":21,"supported":13,"title":517,"abstract":518,"tags":519},[],[],[26],{"code":66,"score":21,"supported":13,"title":521,"abstract":522,"tags":523},[],[],[131],[525,526],"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",4,"10.1016\u002Fj.atech.2026.102588",{"doi":529,"openalex_id":531,"authors":532,"venue":243,"cited_by_count":77,"oa_url":488,"card":544,"direction":82,"ingested_from":165},"W7214044328",[533,536,539,542],{"name":534,"orcid":535},"Dastan Yelubayev","https:\u002F\u002Forcid.org\u002F0000-0001-5358-7982",{"name":537,"orcid":538},"TIMUR SAVIN","https:\u002F\u002Forcid.org\u002F0000-0002-3550-647X",{"name":540,"orcid":541},"Ismail Tokbergenov","https:\u002F\u002Forcid.org\u002F0000-0002-0656-9914",{"name":543,"orcid":8},"Bakhtiyar Zhanzakov",{"tldr":545,"method":546,"finding":547,"direction":82,"opportunity":548},"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"]