[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-1444":3,"related-1444":104},{"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,"primary_topic":31,"secondary_topics":37,"classification":44,"search_phrases":72,"slug":75,"view_count":76,"doi":77,"paper":78,"created_at":103},1444,"An Improved Transformer-KAN Model for Soybean Mapping Based on Multi-Temporal Remote Sensing Data","https:\u002F\u002Fdoi.org\u002F10.3390\u002Fagriculture16171895","Accurate and transferable soybean mapping is essential for agricultural monitoring and area verification, yet conventional Transformer models can be limited in modeling complex nonlinear phenological relationships and maintaining training stability. This study proposes an improved Transformer-KAN model for multi-temporal Sentinel-2 data. Temporal positional encoding, multi-head self-attention with relative positional bias, and a Pre-LayerNorm residual structure are introduced to strengthen phenological sequence modeling, while FastKAN replaces the conventional MLP-based feed-forward network to enhance nonlinear feature representation. The model was trained using 2023 samples from Hailun City and directly evaluated in Bozhou, McLean, and Cass without retraining or fine-tuning. Cross-year transferability was further evaluated by applying the Hailun-trained model to data from Bozhou and McLean from 2021 to 2025. The proposed model achieved overall accuracies of 0.975, 0.983, 0.958, and 0.966 in Hailun, Bozhou, McLean, and Cass, respectively, with corresponding Kappa coefficients of 0.895, 0.887, 0.883, and 0.916. Complexity analysis showed that Transformer-KAN required 0.8142 M parameters and 1.1967 M FLOPs, with an average inference time of 3.0191 ms per sample, compared with 0.6130 M parameters, 0.7971 M FLOPs, and 1.4009 ms per sample for the conventional Transformer, indicating that the improved feature representation was accompanied by increased computational complexity. Cross-year classification performance remained generally high from 2021 to 2025, although interannual variations were observed due to differences in crop growth conditions, phenological timing, and image acquisition quality. Discrepancies between remote-sensing-derived and officially reported soybean areas were mainly related to differences in statistical definitions and residual classification uncertainties, rather than model transferability. Overall, Transformer-KAN provides accurate and transferable soybean mapping and can serve as a spatially explicit complement to official agricultural statistics.","准确且可迁移的大豆制图对于农业监测和面积核查至关重要，然而传统Transformer模型在建模复杂非线性物候关系及维持训练稳定性方面可能存在局限。本研究提出一种改进的Transformer-KAN模型，用于多时相Sentinel-2数据。引入时间位置编码、带相对位置偏置的多头自注意力机制以及Pre-LayerNorm残差结构，以增强物候序列建模能力，同时采用FastKAN替代传统基于MLP的前馈网络，以提升非线性特征表达能力。该模型利用2023年海伦市样本进行训练，并在北州、麦克莱恩和卡斯直接评估，无需重新训练或微调。跨年度迁移能力进一步通过将海伦训练模型应用于2021至2025年北州和麦克莱恩的数据进行评估。所提模型在海伦、北州、麦克莱恩和卡斯的总体精度分别为0.975、0.983、0.958和0.966，对应的Kappa系数分别为0.895、0.887、0.883和0.916。复杂度分析表明，Transformer-KAN需要0.8142 M参数和1.1967 M FLOPs，平均每样本推理时间为3.0191 ms，而传统Transformer为0.6130 M参数、0.7971 M FLOPs和每样本1.4009 ms，说明改进的特征表示伴随着计算复杂度的增加。2021至2025年跨年度分类性能总体保持较高水平，但由于作物生长条件、物候时序和影像获取质量的差异，存在年际波动。遥感反演大豆面积与官方统计面积之间的差异主要与统计口径差异和残余分类不确定性有关，而非模型迁移能力所致。总体而言，Transformer-KAN能够提供准确且可迁移的大豆制图结果，可作为官方农业统计的空间显式补充。",null,"Agriculture","2026-09-01T00:00:00Z","论文",10,false,79,{"impact":17,"substance":18,"depth":19,"authority":20,"freshness":21,"relevant":22,"comment":23},18,22,20,12,7,1,"提出改进的Transformer-KAN模型用于大豆遥感制图，跨区域跨年份验证精度高，对农业监测有实质贡献。",[25],{"name":10,"url":6},[27,28,29,30],"智慧农业","农业人工智能","大豆","遥感",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},"A4","soybean-oilseeds","大豆与油料","\u002Ftopic\u002Ffield-crops\u002Fsoybean-oilseeds","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":14,"version":46,"method":47,"evidence":48},"classified","2026-10-22.2","rules",{"candidates":49,"agricultural_evidence":52},[50,57,62,67],{"code":32,"score":51,"supported":52,"title":53,"abstract":55,"tags":56},15,true,[54],"soybean",[29,54],[29],{"code":39,"score":20,"supported":52,"title":58,"abstract":60,"tags":61},[59],"remote sensing",[30],[30],{"code":63,"score":22,"supported":14,"title":64,"abstract":65,"tags":66},"I1",[],[],[27],{"code":68,"score":22,"supported":14,"title":69,"abstract":70,"tags":71},"I2",[],[],[28],[73,74],"农业人工智能 智慧农业 大豆 遥感","农业人工智能 智慧农业","农业人工智能智慧农业大豆遥感-1444",3,"10.3390\u002Fagriculture16171895",{"doi":77,"openalex_id":79,"authors":80,"venue":10,"cited_by_count":95,"oa_url":6,"card":96,"direction":100,"ingested_from":102},"W7204914742",[81,84,86,89,92],{"name":82,"orcid":83},"Chulin Pan","https:\u002F\u002Forcid.org\u002F0000-0003-3976-465X",{"name":85,"orcid":9},"Jiachen Ju",{"name":87,"orcid":88},"Hongpeng Guo","https:\u002F\u002Forcid.org\u002F0000-0003-0625-8929",{"name":90,"orcid":91},"Yufeng Jiang","https:\u002F\u002Forcid.org\u002F0000-0002-8783-9786",{"name":93,"orcid":94},"Shuang Xu","https:\u002F\u002Forcid.org\u002F0000-0002-3470-0609",0,{"tldr":97,"method":98,"finding":99,"direction":100,"opportunity":101},"提出改进的Transformer-KAN模型，基于多时相遥感数据实现高精度、可迁移的大豆种植区制图。","结合时间位置编码、相对位置偏置、Pre-LayerNorm和FastKAN替换M","模型在四个区域精度达0.958-0.983，跨年迁移稳定，但计算复杂度高于传统Transformer","农业遥感与作物表型","可探索降低模型复杂度或轻量化设计，以平衡精度与效率，并研究跨区域物候差异对迁移的影响。","openalex","2026-09-02T23:30:53.720188Z",{"total":105,"page":22,"page_size":105,"items":106},6,[107,167,254,329,417,470],{"id":108,"title":109,"url":110,"summary":111,"summary_zh":112,"content":9,"source_name":113,"source_url":110,"published_at":114,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":115,"score_detail":116,"sources":120,"tags":122,"primary_topic":125,"secondary_topics":126,"classification":128,"search_phrases":143,"slug":146,"view_count":76,"doi":147,"paper":148,"created_at":166},3501,"A Dual-Phenological-Characteristic Weighting Method to Reconcile Time Discrepancies in Soybean Phenology Estimation from MODIS NDVI Time Series","https:\u002F\u002Fdoi.org\u002F10.3390\u002Frs18193300","Accurate large-scale monitoring of crop phenology is essential for optimizing agricultural management. Remote sensing has been widely used for estimating crop phenological stages, yet time discrepancies often exist between remotely sensed phenological metrics and ground-observed growth stages. Moreover, phenological parameters derived from different characterization models exhibit varying degrees of deviation from field observations. The primary goal of this study was to develop a novel method that fully exploits the deviation patterns of diverse phenological parameters to enhance the accuracy of soybean phenology retrieval. To this end, we extracted 11 phenological parameters for six key growth stages—emerged, blooming, pod-setting, turning yellow, dropping leaf, and harvest—of soybean across 16 U.S. states using MODIS NDVI (normalized difference vegetation index) time-series data from 2000 to 2020, employing GU-, curvature-, and derivative-based phenological modeling methods. The study design centered on proposing a dual-phenological-characteristic weighting (DPCW) method that leverages the deviation features of different phenological parameters relative to ground-observed growth stages, generating composite phenological characteristics by pairing two distinct parameters. The key innovation of this paper is the use of dual-feature weighting to improve the correspondence between satellite-derived phenometrics and field observations, offering an alternative to conventional phenological estimation. The results demonstrated that the optimal DPCW-based combinations for the six growth stages were SOS (start of season) and GREEN, SOS and POS (peak of season), MATURITY and POS, EOS (end of season) and SENES (senescence), RD (recession date) and DD (downturn date), and EOS and DORM (dormancy), respectively. The coefficient of determination (R2) between the retrieved transition dates and ground observations exceeded 0.65 for most stages, with the emerged stage improving to 0.47 from 0.052 and 0.357 of the unadjusted and offset-adjusted benchmarks. The average root mean square error (RMSE) was less than 5 days in most cases, representing a reduction of over 40%, with the most substantial improvement at the turning yellow stage, where RMSE dropped from 12.8 days to 2.8 days. A strength of this study lies in its multi-state, multi-decade validation, demonstrating the robustness and temporal consistency of the DPCW method within the major U.S. soybean-growing region. However, a limitation is that the method’s performance may vary with different satellite sensors or crop types, warranting further investigation. The proposed approach is expected to enhance the accuracy of remote sensing-based crop phenology monitoring and offers an effective alternative for calibrating remotely sensed phenological parameters.","准确的大尺度作物物候监测对于优化农业管理至关重要。遥感已被广泛用于估算作物物候阶段，但遥感物候指标与地面观测生育阶段之间常存在时间差异。此外，不同特征化模型衍生的物候参数与田间观测之间存在不同程度的偏差。本研究的主要目标是开发一种新方法，充分利用多种物候参数的偏差模式，以提高大豆物候反演精度。为此，我们利用2000—2020年MODIS NDVI（归一化差异植被指数）时间序列数据，采用基于GU、曲率和导数的方法提取了美国16个州大豆六个关键生育阶段——出苗、开花、结荚、黄化、落叶和收获——的11个物候参数。研究设计的核心是提出一种双物候特征加权（dual-phenological-characteristic weighting，DPCW）方法，该方法利用不同物候参数相对于地面观测生育阶段的偏差特征，通过配对两个不同参数生成复合物候特征。本文的关键创新在于利用双特征加权提高卫星衍生物候指标与田间观测之间的对应关系，为传统物候估算提供了一种替代方案。结果表明，六个生育阶段基于DPCW的最优组合分别为SOS（生长季开始）与GREEN、SOS与POS（生长季峰值）、MATURITY与POS、EOS（生长季结束）与SENES（衰老）、RD（衰退日期）与DD（下降日期）以及EOS与DORM（休眠）。反演得到的转换日期与地面观测之间的决定系数（R²）在大多数阶段超过0.65，其中出苗阶段从基准的0.052和偏移调整后的0.357提高至0.47。大多数情况下平均均方根误差（RMSE）小于5天，降幅超过40%，其中黄化阶段改善最为显著，RMSE从12.8天降至2.8天。本研究的一个优势在于其多州、多年代际验证，证明了DPCW方法在美国主要大豆种植区内的稳健性和时间一致性。","Remote Sensing","2026-09-24T00:00:00Z",81,{"impact":17,"substance":18,"depth":17,"authority":117,"freshness":118,"relevant":22,"comment":119},14,9,"提出双物候特征加权方法，用MODIS NDVI长时序数据校正大豆物候遥感估算偏差，方法新颖、验证扎实，对农情遥感监测有参考价值。",[121],{"name":113,"url":110},[27,29,123,30,124],"农情监测","作物表型",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[127],{"code":39,"slug":40,"name":41,"path":42,"parent_code":43},{"status":45,"locked":14,"version":46,"method":47,"evidence":129},{"candidates":130,"agricultural_evidence":52},[131,135,139],{"code":32,"score":51,"supported":52,"title":132,"abstract":133,"tags":134},[54],[29,54],[29],{"code":39,"score":118,"supported":52,"title":136,"abstract":137,"tags":138},[],[30,59],[124,123,30],{"code":63,"score":22,"supported":14,"title":140,"abstract":141,"tags":142},[],[],[27],[144,145],"MODIS NDVI 大豆 物候","美国大豆 遥感 物候监测","MODISNDVI大豆物候-3501","10.3390\u002Frs18193300",{"doi":147,"openalex_id":149,"authors":150,"venue":113,"cited_by_count":95,"oa_url":110,"card":161,"direction":100,"ingested_from":102},"W7214203102",[151,153,156,159],{"name":152,"orcid":9},"Qiuxiang Yi",{"name":154,"orcid":155},"Siting Chen","https:\u002F\u002Forcid.org\u002F0000-0003-3468-9320",{"name":157,"orcid":158},"Fumin Wang","https:\u002F\u002Forcid.org\u002F0000-0002-5078-358X",{"name":160,"orcid":9},"Qinyan Zhu",{"tldr":162,"method":163,"finding":164,"direction":100,"opportunity":165},"提出双物候特征加权法，校正MODIS NDVI大豆物候估计与地面观测的时间偏差。","用MODIS NDVI 2000-2020数据，结合GU、曲率、导数三类物候模型","多数生育期R²超0.65，RMSE多小于5天，降幅超40%，转黄期RMSE从12.8天降至2.8天。","可将该加权校正思路迁移到其他作物与多源遥感数据，并探索自适应权重与深度学习融合的物候反演。","2026-09-25T23:30:31.075499Z",{"id":168,"title":169,"url":170,"summary":171,"summary_zh":172,"content":9,"source_name":113,"source_url":170,"published_at":173,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":174,"score_detail":175,"sources":178,"tags":180,"primary_topic":183,"secondary_topics":184,"classification":190,"search_phrases":215,"slug":218,"view_count":219,"doi":220,"paper":221,"created_at":253},2320,"Genotype-Aware Prediction of Soybean Seed Composition from Multimodal UAV Imagery of the Standing Crop","https:\u002F\u002Fdoi.org\u002F10.3390\u002Frs18183121","Geospatial artificial intelligence (GeoAI) integrates multimodal remote sensing with deep learning to model complex agricultural systems at scale. Within this framework, accurate and non-destructive prediction of seed composition from in-season standing crops is essential for breeding and precision agriculture. This study developed an end-to-end convolutional neural network (CNN) framework to estimate eight seed traits (protein, oil, sucrose, fiber, starch, ash, complex and simple carbohydrates) from UAV-based multisensor imagery and associated genotype and phenological metadata. A total of 372 soybean samples were collected over two growing seasons (2020–2021) from two fields in Missouri, with UAV flights capturing multispectral (MSI), thermal (THR), and LiDAR (LDR) data at four time points spanning vegetative to reproductive growth stages. CNN models were trained in single- and multi-date configurations, incorporating genotype (GEN) and days after sowing (DAS) as additional features. The highest accuracy was achieved for sucrose (R2 = 0.80), followed by simple carbohydrate (R2 = 0.70) and starch (R2 = 0.55), with notable gains from GEN and DAS. Multi-date models incorporating earlier acquisitions often matched or outperformed later or all-date combinations. Among modalities, MSI provided the most robust estimates, with limited added value from LDR or THR. Unlike feature-based pipelines prone to multicollinearity, this image-to-trait approach enables automated, scalable prediction of soybean seed composition for in-season, field-level assessment.","地理空间人工智能（GeoAI）将多模态遥感与深度学习相结合，以在大尺度上对复杂农业系统进行建模。在该框架下，对当季未收获作物进行准确且无损的种子成分预测，对于育种和精准农业至关重要。本研究开发了一种端到端的卷积神经网络（CNN）框架，利用基于无人机（UAV）的多传感器影像以及相关的基因型和物候元数据，估算八种种子性状（蛋白质、油分、蔗糖、纤维、淀粉、灰分、复杂碳水化合物和简单碳水化合物）。研究在2020—2021两个生长季内，从密苏里州两块田地共采集372份大豆样本，无人机飞行在从营养生长期到生殖生长期的四个时间点获取了多光谱（MSI）、热红外（THR）和激光雷达（LDR）数据。CNN模型以单日期和多日期配置进行训练，并将基因型（GEN）和播种后天数（DAS）作为附加特征纳入。蔗糖的估算精度最高（R² = 0.80），其次为简单碳水化合物（R² = 0.70）和淀粉（R² = 0.55），且GEN和DAS的加入带来了显著提升。纳入早期采集数据的多日期模型往往与后期或全日期组合的表现相当，甚至更优。在各模态中，MSI提供了最稳健的估算结果，而LDR或THR的附加价值有限。与易受多重共线性影响的特征工程流程不同，这种从图像到性状的方法能够实现大豆种子成分的自动化、可扩展预测，用于当季田块尺度的评估。","2026-09-11T00:00:00Z",80,{"impact":17,"substance":18,"depth":17,"authority":117,"freshness":176,"relevant":22,"comment":177},8,"基于多模态无人机影像与基因型信息的大豆籽粒成分无损预测研究，方法新颖、数据扎实，对智慧育种与精准农业有参考价值。",[179],{"name":113,"url":170},[27,181,28,30,182],"无人机","大豆育种",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[185,189],{"code":68,"slug":186,"name":187,"path":188,"parent_code":43},"agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai",{"code":39,"slug":40,"name":41,"path":42,"parent_code":43},{"status":45,"locked":14,"version":46,"method":47,"evidence":191},{"candidates":192,"agricultural_evidence":52},[193,197,204,209],{"code":32,"score":51,"supported":52,"title":194,"abstract":195,"tags":196},[54],[29,54],[182],{"code":68,"score":13,"supported":52,"title":198,"abstract":199,"tags":203},[],[200,201,202],"深度学习","deep learning","artificial intelligence",[28],{"code":39,"score":13,"supported":52,"title":205,"abstract":206,"tags":208},[],[30,59,207],"multispectral",[30],{"code":63,"score":21,"supported":52,"title":210,"abstract":211,"tags":214},[],[212,213],"精准农业","precision agriculture",[27],[216,217],"农业人工智能 大豆育种 智慧农业 无人机","农业人工智能 大豆育种","农业人工智能大豆育种智慧农业无人机-2320",4,"10.3390\u002Frs18183121",{"doi":220,"openalex_id":222,"authors":223,"venue":113,"cited_by_count":95,"oa_url":170,"card":248,"direction":100,"ingested_from":102},"W7212223520",[224,227,230,233,236,238,240,243,246],{"name":225,"orcid":226},"Vasit Sagan","https:\u002F\u002Forcid.org\u002F0000-0003-4375-2096",{"name":228,"orcid":229},"Kristen Rhodes","https:\u002F\u002Forcid.org\u002F0000-0002-8469-5043",{"name":231,"orcid":232},"Sourav Bhadra","https:\u002F\u002Forcid.org\u002F0000-0002-5832-4695",{"name":234,"orcid":235},"Haireti Alifu","https:\u002F\u002Forcid.org\u002F0000-0002-7369-6657",{"name":237,"orcid":9},"Aviskar Giri",{"name":239,"orcid":9},"Ashutosh Pawar",{"name":241,"orcid":242},"Bishal Roy","https:\u002F\u002Forcid.org\u002F0000-0001-9912-5505",{"name":244,"orcid":245},"Supria Sarkar","https:\u002F\u002Forcid.org\u002F0000-0001-9467-5786",{"name":247,"orcid":9},"Felix Fritschi",{"tldr":249,"method":250,"finding":251,"direction":100,"opportunity":252},"用无人机多模态影像与基因型数据训练CNN，预测大豆八种种子成分。","372份大豆样本，多光谱\u002F热红外\u002FLiDAR影像加基因型与播后天数，端到端CNN","蔗糖预测最优(R²=0.80)，基因型与播期提升精度，多光谱最稳健。","可探索基因型-表型-环境交互建模，并迁移至其他作物与多时相早期预测。","2026-09-13T23:30:22.776177Z",{"id":255,"title":256,"url":257,"summary":258,"summary_zh":259,"content":9,"source_name":260,"source_url":257,"published_at":261,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":262,"score_detail":263,"sources":267,"tags":269,"primary_topic":273,"secondary_topics":274,"classification":276,"search_phrases":300,"slug":303,"view_count":219,"doi":304,"paper":305,"created_at":328},4998,"Unmanned aerial vehicle time‐series phenotyping illuminates genotype‐specific responses to off‐target dicamba exposure in soybean","https:\u002F\u002Fdoi.org\u002F10.1002\u002Fppj2.70118","Abstract Dicamba, a synthetic auxin herbicide, is widely used for postemergence weed control. However, the high potential for volatility and spray drift raises concerns about damage to nontarget dicot crops lacking the dicamba monooxygenase transgene, which provides bioengineered resistance. Our objective was to evaluate the practical application of unmanned aerial vehicle (UAV)‐based multispectral imaging for quantifying off‐target dicamba injury, using non‐genetically modified (GM) soybean ( Glycine max (L.) Merr.) genotypes with known tolerant (T) and sensitive (S) responses to provide a reference framework. We conducted field experiments in 2021 and 2022 using varying dicamba rates. Temporal visual injury estimates (0–5 scale) were recorded approximately every 5 days for 6–8 weeks. Significant differences were detected between T and S lines, especially with 1\u002F500 single and weekly repeated treatments, which provided the largest separation. To augment the precision of visual injury assessment, UAV‐based multispectral imagery from each of multiple flights was used to calculate four vegetation indices (VIs): green leaf index (GLI), normalized excess green index (ExG2), normalized green red difference index (NGRDI), and red edge normalized difference vegetation index (RENDVI). These VIs captured differences between T and S lines, indicating dicamba injury can be quantified through remote sensing across contrasting soybean genotypes. Strong negative correlations were observed between visual injury scores and VI scores across treatments at each flight date, with overall correlations for GLI, ExG2, NGRDI, and RENDVI being −0.43, −0.43, −0.44, and −0.39 in 2021 and −0.54, −0.54, −0.64, and −0.57 in 2022. These results support the utility of UAV‐derived VIs for detecting and quantifying off‐target herbicide injury in soybean, facilitating scientific research efforts to develop non‐GM‐tolerant genotypes.","摘要 麦草畏（dicamba）是一种合成生长素类除草剂，广泛用于苗后杂草防除。然而，其高挥发性和喷雾飘移潜力引发了人们对非靶标双子叶作物的担忧，这些作物缺乏提供生物工程抗性的麦草畏单加氧酶（dicamba monooxygenase）转基因。本研究旨在评估基于无人机（UAV）的多光谱成像在量化非靶标麦草畏药害中的实际应用，利用对麦草畏具有已知耐受（T）和敏感（S）响应的非转基因（GM）大豆（Glycine max (L.) Merr.）基因型作为参考框架。我们于2021年和2022年开展了田间试验，使用不同的麦草畏剂量。在约6–8周内，每约5天记录一次时序视觉药害评估（0–5级）。在T和S品系之间检测到显著差异，尤其是在1\u002F500单次和每周重复处理下，差异最为明显。为提高视觉药害评估的精度，利用多次飞行获取的无人机多光谱影像计算了4种植被指数（VIs）：绿叶指数（GLI）、归一化超绿指数（ExG2）、归一化绿红差异指数（NGRDI）和红边归一化差异植被指数（RENDVI）。这些植被指数捕捉到了T和S品系之间的差异，表明麦草畏药害可通过遥感在不同大豆基因型间进行量化。在各飞行日期的所有处理中，视觉药害评分与植被指数评分之间均观察到强负相关，2021年GLI、ExG2、NGRDI和RENDVI的总体相关系数分别为−0.43、−0.43、−0.44和−0.39，2022年分别为−0.54、−0.54、−0.64和−0.57。这些结果支持无人机衍生植被指数在大豆非靶标除草剂药害检测和量化中的实用性，有助于推动开发非转基因耐受基因型的科学研究工作。","The Plant Phenome Journal","2026-10-14T00:00:00Z",77,{"impact":264,"substance":18,"depth":17,"authority":265,"freshness":176,"relevant":22,"comment":266},16,13,"该研究利用无人机多光谱时序表型定量评估大豆非靶标除草剂药害，方法新颖、数据扎实，对智慧农业遥感监测具有参考价值。",[268],{"name":260,"url":257},[27,29,270,271,272],"无人机遥感","植被指数","除草剂药害",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[275],{"code":39,"slug":40,"name":41,"path":42,"parent_code":43},{"status":45,"locked":14,"version":46,"method":47,"evidence":277},{"candidates":278,"agricultural_evidence":52},[279,283,289,296],{"code":32,"score":51,"supported":52,"title":280,"abstract":281,"tags":282},[54],[29,54],[29],{"code":39,"score":284,"supported":52,"title":285,"abstract":286,"tags":288},11,[],[30,287,59,207],"无人机多光谱",[270,271],{"code":290,"score":105,"supported":52,"title":291,"abstract":292,"tags":295},"F4",[],[293,294],"除草","weed control",[],{"code":63,"score":22,"supported":14,"title":297,"abstract":298,"tags":299},[],[],[27],[301,302],"无人机 多光谱 大豆 药害","麦草畏 飘移 大豆 遥感","无人机多光谱大豆药害-4998","10.1002\u002Fppj2.70118",{"doi":304,"openalex_id":306,"authors":307,"venue":260,"cited_by_count":95,"oa_url":257,"card":323,"direction":100,"ingested_from":102},"W7221759072",[308,311,314,317,320],{"name":309,"orcid":310},"Seda Ozer","https:\u002F\u002Forcid.org\u002F0000-0002-1644-2394",{"name":312,"orcid":313},"Brian W. Diers","https:\u002F\u002Forcid.org\u002F0000-0003-3584-5495",{"name":315,"orcid":316},"Dean E. Riechers","https:\u002F\u002Forcid.org\u002F0000-0002-6081-5629",{"name":318,"orcid":319},"Caio Canella Vieira","https:\u002F\u002Forcid.org\u002F0000-0001-5428-3753",{"name":321,"orcid":322},"Alper Adak","https:\u002F\u002Forcid.org\u002F0000-0002-2737-8041",{"tldr":324,"method":325,"finding":326,"direction":100,"opportunity":327},"用无人机多光谱时序表型量化非转基因大豆受除草剂飘移药害的基因型差异。","2021-2022田间试验，无人机多光谱影像计算GLI、ExG2、NGRDI、R","植被指数与目测药害评分显著负相关，能区分耐性与敏感基因型，2022年相关性更强。","可拓展到多作物、多除草剂飘移的无人机早期诊断，并融合机器学习构建药害分级预警模型。","2026-10-15T23:30:20.537887Z",{"id":330,"title":331,"url":332,"summary":333,"summary_zh":334,"content":9,"source_name":335,"source_url":332,"published_at":336,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":115,"score_detail":337,"sources":339,"tags":341,"primary_topic":344,"secondary_topics":350,"classification":353,"search_phrases":385,"slug":388,"view_count":76,"doi":389,"paper":390,"created_at":416},2318,"Estimation of field-scale crop evapotranspiration from the mechanistic SIF-ET model using the UAV hyperspectral imagery","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.eja.2026.128332","Accurate estimation of evapotranspiration (ET) is critical for agricultural water management and understanding land-atmosphere interactions. Conventional thermal and optical remote sensing is a powerful tool for ET estimation, but existing methods remain constrained by empirical parameterization, calibration complexity, and limited applicability. As a direct indicator of vegetation photosynthesis, solar-induced chlorophyll fluorescence (SIF) enables consistent and accurate ET estimation through a mechanistically grounded framework that does not require site-specific empirical calibration of the SIF-GPP relationship. However, current SIF-based ET estimation mainly rely on coarse-resolution satellite data and its capacity to resolve high-resolution ET dynamics over farmland remains unexplored. To address this, the mechanistic light response model and water-carbon relationship was integrated to construct a SIF-ET model at field-scale using UAV narrow-band hyperspectra imagery. The principal conclusions are: (1) High-resolution crop (canola,soybean and wheat) canopy SIF 740 at 1-nm SR was retrieved from UAV hyperspectra imagery using the Photochemistry and Energy Fluxes (SCOPE) model; (2)The daily ET estimates from SIF-ET model were validated against the in-situ measurements across multiple growth stages and water-nitrogen treatments, with wheat exhibiting the highest precision (R²=0.72–0.94, RMSE = 0.04–0.11 mm\u002Fd, MAPE = 1.70–5.89%), followed by canola and soybean; (3) Cumulative ET errors across repeated UAV observations within each growth stage remained within acceptable bounds, indicating the stability of SIF-ET model in capturing long-term ET dynamics. Overall, the SIF-ET model from UAV hyperspectra imagery revealed fine-scale cropland ET spatial-temporal heterogeneity. These findings provide support for precision irrigation, rational water-nitrogen management, and stress diagnostics under changing environmental conditions.","准确估算蒸散发（ET）对农业水资源管理和理解陆气相互作用至关重要。传统的热红外与光学遥感是估算ET的有力工具，但现有方法仍受限于经验参数化、校准复杂性和适用性有限等问题。作为植被光合作用的直接指示因子，日光诱导叶绿素荧光（SIF）能够通过机理明确的框架实现一致且准确的ET估算，且无需对SIF-GPP关系进行站点特定的经验校准。然而，当前基于SIF的ET估算主要依赖粗分辨率卫星数据，其解析农田高分辨率ET动态的能力尚未得到探索。为此，本研究整合机理光响应模型与水碳关系，利用无人机窄波段高光谱影像构建了田块尺度的SIF-ET模型。主要结论如下：（1）利用光化学与能量通量（SCOPE）模型，从无人机高光谱影像中反演了1 nm光谱分辨率下油菜、大豆和小麦的高分辨率冠层SIF₇₄₀；（2）基于SIF-ET模型的日ET估算值在多个生育期和水氮处理下与原位观测进行了验证，其中小麦精度最高（R²=0.72–0.94，RMSE=0.04–0.11 mm\u002Fd，MAPE=1.70–5.89%），油菜和大豆次之；（3）各生育期内重复无人机观测的累积ET误差均在可接受范围内，表明SIF-ET模型在捕捉长期ET动态方面具有稳定性。总体而言，基于无人机高光谱影像的SIF-ET模型揭示了精细尺度的农田ET时空异质性。这些发现为精准灌溉、合理水氮管理以及变化环境条件下的胁迫诊断提供了支持。","European Journal of Agronomy","2026-09-12T00:00:00Z",{"impact":17,"substance":18,"depth":17,"authority":117,"freshness":118,"relevant":22,"comment":338},"该研究基于无人机高光谱与SIF-ET机理模型实现田块尺度作物蒸散发高精度估算，方法新颖、验证充分，对精准灌溉与水氮管理有实质参考价值，值得进入每日精选。",[340],{"name":335,"url":332},[27,181,30,342,343],"精准灌溉","蒸散发",{"code":345,"slug":346,"name":347,"path":348,"parent_code":349},"H3","irrigation","水资源与灌溉","\u002Ftopic\u002Fenvironment\u002Firrigation","H",[351,352],{"code":39,"slug":40,"name":41,"path":42,"parent_code":43},{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},{"status":45,"locked":14,"version":46,"method":47,"evidence":354},{"candidates":355,"agricultural_evidence":52},[356,364,369,374,381],{"code":345,"score":357,"supported":52,"title":358,"abstract":360,"tags":363},19,[359],"evapotranspiration",[361,362,343,346,359],"灌溉","水资源",[342,343],{"code":39,"score":51,"supported":52,"title":365,"abstract":367,"tags":368},[366],"hyperspectral",[30,59],[30],{"code":32,"score":118,"supported":52,"title":370,"abstract":371,"tags":373},[],[29,372,54],"油菜",[],{"code":375,"score":105,"supported":52,"title":376,"abstract":377,"tags":380},"A2",[],[378,379],"小麦","wheat",[],{"code":63,"score":22,"supported":14,"title":382,"abstract":383,"tags":384},[],[],[27],[386,387],"智慧农业 精准灌溉 无人机 蒸散发","智慧农业 精准灌溉","智慧农业精准灌溉无人机蒸散发-2318","10.1016\u002Fj.eja.2026.128332",{"doi":389,"openalex_id":391,"authors":392,"venue":335,"cited_by_count":95,"oa_url":332,"card":411,"direction":100,"ingested_from":102},"W7212242584",[393,395,397,400,403,405,408],{"name":394,"orcid":9},"Ruiqi Du",{"name":396,"orcid":9},"Yonghong Zhang",{"name":398,"orcid":399},"Youzhen Xiang","https:\u002F\u002Forcid.org\u002F0000-0002-8268-1609",{"name":401,"orcid":402},"Fucang Zhang","https:\u002F\u002Forcid.org\u002F0000-0002-6659-3262",{"name":404,"orcid":9},"Jian Gao",{"name":406,"orcid":407},"Qiliang Yang","https:\u002F\u002Forcid.org\u002F0000-0002-3274-2119",{"name":409,"orcid":410},"Xianghui Lu","https:\u002F\u002Forcid.org\u002F0000-0003-0638-3068",{"tldr":412,"method":413,"finding":414,"direction":100,"opportunity":415},"利用无人机高光谱影像构建田间尺度SIF-ET机理模型，实现作物蒸散发高精度估算。","无人机窄波段高光谱结合SCOPE模型反演SIF，耦合光响应与水碳关系构建SIF-","小麦估算精度最高（R²=0.72–0.94），模型能稳定捕捉多生育期蒸散发动态与空间异质性。","可探索多作物多环境下的SIF-ET普适性，并融合热红外与机器学习提升胁迫诊断能力。","2026-09-13T23:30:22.677041Z",{"id":418,"title":419,"url":420,"summary":421,"summary_zh":422,"content":9,"source_name":423,"source_url":420,"published_at":424,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":425,"score_detail":426,"sources":429,"tags":431,"primary_topic":434,"secondary_topics":435,"classification":437,"search_phrases":454,"slug":457,"view_count":95,"doi":458,"paper":459,"created_at":469},5570,"Evaluation of Counterfactual Explanation Robustness for Computer Vision Agriculture Yield-Forecast Models Under Satellite Cloud-Mask Gaps: An Interpretability Reliability Bound for Cooperative Storage Decisions","https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23545079","Counterfactual explanations are increasingly proposed as the interface through which cooperative storage managers interrogate computer-vision yield-forecast models, yet their reliability under the satellite cloud-mask gaps that routinely afflict agricultural remote sensing has not been bounded. The bound is derived from a Lipschitz continuity assumption on the forecast model and a masking-operator perturbation analysis, and it is expressed as a function of the cloud-mask gap fraction, the model's local Lipschitz constant, and the counterfactual search tolerance. A replication-based evaluation of the before-after-control-impact estimator was conducted to characterise the operating characteristics of the design under a declared variance structure (between-cluster SD = 6.0, residual SD = 4.5, intraclass correlation = 0.64), with power estimated across 2000 design cells. The reliability bound degrades non-linearly as the gap fraction rises, and that the intraclass correlation of 0.64 implies that cooperative-level aggregation of counterfactual recourse is necessary to preserve explanation stability. The contribution is a decision-relevant reliability bound that storage cooperatives can use to set a maximum tolerable cloud-mask gap before counterfactual recourse is suspended.","反事实解释日益被提出作为合作式仓储管理者质询计算机视觉产量预测模型的接口，然而在农业遥感中经常出现的卫星云掩膜缺口条件下，其可靠性尚未得到界定。该界基于预测模型上的Lipschitz连续性假设和掩膜算子扰动分析推导得出，并表示为云掩膜缺口比例、模型局部Lipschitz常数和反事实搜索容差的函数。对前后对照影响估计量进行了基于重复的评估，以表征该设计在给定方差结构（组间标准差=6.0，残差标准差=4.5，组内相关系数=0.64）下的运行特征，并在2000个设计单元中估计了检验效能。可靠性界随缺口比例上升呈非线性退化，且组内相关系数0.64意味着需要对反事实救济进行合作社层面的聚合，以保持解释的稳定性。本文的贡献在于提供了一个与决策相关的可靠性界，仓储合作社可据此设定在暂停反事实救济之前可容忍的最大云掩膜缺口。","Zenodo (CERN European Organization for Nuclear Research)","2026-10-21T00:00:00Z",67,{"impact":176,"substance":19,"depth":427,"authority":20,"freshness":13,"relevant":22,"comment":428},17,"该论文提出反事实解释在卫星云掩膜缺口下的可靠性边界，方法新颖且结论可用于合作社决策，但属细分方法学研究，产业影响有限。",[430],{"name":423,"url":420},[27,28,432,30,433],"产量预测","可解释性",{"code":68,"slug":186,"name":187,"path":188,"parent_code":43},[436],{"code":39,"slug":40,"name":41,"path":42,"parent_code":43},{"status":45,"locked":14,"version":46,"method":47,"evidence":438},{"candidates":439,"agricultural_evidence":52},[440,446,450],{"code":68,"score":20,"supported":52,"title":441,"abstract":443,"tags":445},[442],"computer vision",[444],"计算机视觉",[28],{"code":39,"score":176,"supported":52,"title":447,"abstract":448,"tags":449},[],[30,59],[432,30],{"code":63,"score":22,"supported":14,"title":451,"abstract":452,"tags":453},[],[],[27],[455,456],"农业人工智能 产量预测 可解释性 智慧农业","农业人工智能 产量预测","农业人工智能产量预测可解释性智慧农业-5570","10.5281\u002Fzenodo.23545079",{"doi":458,"openalex_id":460,"authors":461,"venue":423,"cited_by_count":95,"oa_url":420,"card":464,"direction":100,"ingested_from":102},"W7223114002",[462],{"name":463,"orcid":9},"Kadiatou Barry",{"tldr":465,"method":466,"finding":467,"direction":100,"opportunity":468},"推导卫星云掩膜缺口下反事实解释可靠性的理论界，并评估其对合作社储粮决策的影响。","基于Lipschitz连续性假设与掩膜扰动分析推导可靠性界，并用2000个设计单","缺口比例上升使可靠性非线性下降；ICC=0.64表明需合作社层级聚合以保持解释稳定。","可研究不同作物区云掩膜统计特性下的自适应解释暂停阈值与聚合策略。","2026-10-22T11:20:39.240791Z",{"id":471,"title":472,"url":473,"summary":474,"summary_zh":475,"content":9,"source_name":476,"source_url":473,"published_at":477,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":478,"score_detail":479,"sources":481,"tags":483,"primary_topic":485,"secondary_topics":489,"classification":491,"search_phrases":523,"slug":526,"view_count":95,"doi":527,"paper":528,"created_at":541},5394,"Rice Yield Prediction Using Multi-Modal Regression Learning with Optimized Ensemble Models","https:\u002F\u002Fdoi.org\u002F10.3390\u002Feng7100572","Accurate rice yield estimation is important for agricultural planning, resource allocation, and food security assessment, but yield depends on interacting climatic, vegetation, remote sensing, seasonal, and crop growth factors. This study presents a Multi-Modal Regression Learning (MMRL) framework for continuous rice yield prediction. The framework integrates rainfall, temperature, humidity, normalized difference vegetation index (NDVI), VH synthetic aperture radar (SAR) backscatter, season, rice cultivation area, and leaf wetness information. Predictor relevance is assessed before model development, and multiple regression learners are trained using a common feature set. Their validation predictions are combined through non-negative ensemble weights constrained to sum to one, thereby producing a single optimized prediction. Experiments were performed on a modeled spatio-temporal agricultural dataset representing Karnataka conditions, with 10,000 observations used for the regression comparison. The proposed MMRL ensemble achieved an R2 of 0.9515, an RMSE of 0.6334 tons\u002Fha, an MAE of 0.5075 tons\u002Fha, and an MAPE of 9.81%, outperforming the evaluated individual regression models. Diagnostic plots further indicated close agreement between predicted and reference yields across most of the evaluated range. The MMRL workflow was also evaluated on a separately reconstructed synthetic dataset generated with fully specified distributions, sampling rules, noise terms, a yield generation equation, and fixed random seeds.","准确的水稻产量估算对农业规划、资源配置和粮食安全评估具有重要意义，但产量取决于气候、植被、遥感、季节和作物生长等多种因素的相互作用。本研究提出了一种用于连续水稻产量预测的多模态回归学习（Multi-Modal Regression Learning，MMRL）框架。该框架整合了降雨量、温度、湿度、归一化植被指数（NDVI）、VH合成孔径雷达（SAR）后向散射、季节、水稻种植面积和叶片湿润信息。在模型开发之前评估预测变量的相关性，并使用共同特征集训练多个回归学习器。其验证预测结果通过约束为和等于1的非负集成权重进行组合，从而生成单一优化预测。实验在代表卡纳塔克邦条件的建模时空农业数据集上进行，使用10,000个观测值进行回归比较。所提出的MMRL集成模型实现了R²为0.9515、RMSE为0.6334吨\u002F公顷、MAE为0.5075吨\u002F公顷、MAPE为9.81%，优于所评估的各个回归模型。诊断图进一步表明，在大多数评估范围内，预测产量与参考产量高度一致。MMRL工作流程还在一个单独重建的合成数据集上进行了评估，该数据集使用完全指定的分布、采样规则、噪声项、产量生成方程和固定随机种子生成。","Eng—Advances in Engineering","2026-10-18T00:00:00Z",78,{"impact":17,"substance":18,"depth":17,"authority":20,"freshness":176,"relevant":22,"comment":480},"多模态回归集成方法在水稻估产上取得较高精度，方法新颖、数据规模明确，对农业信息化有参考价值，但属学术论文且数据集为模拟数据，产业影响有限。",[482],{"name":476,"url":473},[27,28,484,432,30],"水稻",{"code":486,"slug":487,"name":484,"path":488,"parent_code":36},"A1","rice","\u002Ftopic\u002Ffield-crops\u002Frice",[490],{"code":39,"slug":40,"name":41,"path":42,"parent_code":43},{"status":45,"locked":14,"version":46,"method":47,"evidence":492},{"candidates":493,"agricultural_evidence":52},[494,498,502,509,515,519],{"code":486,"score":51,"supported":52,"title":495,"abstract":496,"tags":497},[487],[484,487],[484],{"code":39,"score":176,"supported":52,"title":499,"abstract":500,"tags":501},[],[30,59],[432,30],{"code":503,"score":105,"supported":52,"title":504,"abstract":505,"tags":508},"A8",[],[506,507],"粮食安全","food security",[],{"code":510,"score":76,"supported":52,"title":511,"abstract":512,"tags":514},"J1",[],[513],"农业规划",[],{"code":63,"score":22,"supported":14,"title":516,"abstract":517,"tags":518},[],[],[27],{"code":68,"score":22,"supported":14,"title":520,"abstract":521,"tags":522},[],[],[28],[524,525],"水稻 产量预测 多模态","NDVI SAR 水稻估产","水稻产量预测多模态-5394","10.3390\u002Feng7100572",{"doi":527,"openalex_id":529,"authors":530,"venue":476,"cited_by_count":95,"oa_url":473,"card":536,"direction":100,"ingested_from":102},"W7222942064",[531,533],{"name":532,"orcid":9},"Nayana Venkatesh",{"name":534,"orcid":535},"Evangelin Geetha Devaraj","https:\u002F\u002Forcid.org\u002F0000-0002-7253-4314",{"tldr":537,"method":538,"finding":539,"direction":100,"opportunity":540},"构建多模态回归学习框架，融合气候、遥感与作物因子，实现水稻产量连续预测。","融合降雨、温度、NDVI、SAR后向散射等多源特征，用非负权重集成多个回归器。","集成模型R²达0.9515，RMSE为0.6334吨\u002F公顷，优于各单一回归模型。","可引入真实田间多时相遥感与深度时序模型，验证合成数据结论在真实产区的泛化性。","2026-10-20T23:30:19.292634Z"]