[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-3550":3,"related-3550":105},{"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":32,"secondary_topics":38,"classification":44,"search_phrases":76,"slug":79,"view_count":80,"doi":81,"paper":82,"created_at":104},3550,"Asymmetric Vegetation Responses to Flood Exposure in the Chi River Basin: A Multi-Temporal Remote Sensing and Machine Learning Investigation","https:\u002F\u002Fdoi.org\u002F10.3390\u002Fsym18101606","Flooding and drought alternate in many tropical floodplains, but how vegetation responds to both hazards over multiple years remains poorly quantified. We examined vegetation–flood associations in the Chi River Basin, Northeast Thailand, drawing on Sentinel-2 imagery from 2020, 2023, and 2024, flood records from 2017, 2018, 2021, and 2022, and machine learning methods across 230,911 point-based spatial units. Nine forms of asymmetry emerged from the analysis. The most striking was recovery asymmetry: areas that flooded at least twice between 2017 and 2022 gained vegetation during 2020–2023 (mean NDVI change = +0.0270), whereas areas with little or no flood exposure lost vegetation (mean change = −0.0430). This gap was statistically significant (Cohen’s d = 0.5586, p \u003C 0.001) and suggests that repeated flooding may buffer vegetation against subsequent drought. Agricultural land declined less (−0.0316) than forest (−0.0937; ANOVA F = 1717.7, p \u003C 0.001). Baseline vegetation condition, measured as NDVI in 2020, contributed 32.9% to model importance, more than twice the contribution of elevation (14.5%). Spectral indices together accounted for 76.6% of importance, compared with 23.5% for topographic variables. The 2023 El Niño year produced the largest difference between high-flood and low-flood areas (+0.0364); because only one year per ENSO phase was available, we treat this as a case-based comparison rather than a general ENSO response. Threshold analysis identified two distinct values: an operational cut-off at NDVI = 0.05 (overall accuracy 82.67%) and an ecological transition around 0.25–0.30. Spatial clustering was weak but significant (Moran’s I = 0.2179, p \u003C 0.001). Spatial block cross-validation gave lower accuracy (0.597) than random cross-validation (0.627), pointing to spatial autocorrelation in the data. High-flood areas had 1.67 times the vulnerability index of low-flood areas (0.4306 vs. 0.2573). These patterns support differentiated management: elevation-based zoning, warning systems calibrated to local flood regimes, and focused interventions at hotspots.","在许多热带洪泛平原，洪水与干旱交替发生，但植被如何在多年间同时响应这两种灾害，仍缺乏充分的定量研究。我们以泰国东北部栖河（Chi River）流域为研究区，考察植被与洪水的关联，数据来源包括2020年、2023年和2024年的Sentinel-2影像，2017年、2018年、2021年和2022年的洪水记录，并采用机器学习方法，覆盖230 911个基于点的空间单元。分析中出现了九种不对称形式。其中最显著的是恢复不对称：2017—2022年间至少发生两次洪水的区域，在2020—2023年间植被增加（NDVI平均变化=+0.0270），而洪水暴露很少或没有洪水暴露的区域则植被减少（平均变化=−0.0430）。这一差异具有统计显著性（Cohen’s d=0.5586，p\u003C0.001），表明反复洪水可能缓冲植被对随后干旱的响应。农田的下降幅度（−0.0316）小于森林（−0.0937；ANOVA F=1717.7，p\u003C0.001）。以2020年NDVI衡量的基线植被状况对模型重要性的贡献为32.9%，是海拔贡献（14.5%）的两倍多。光谱指数合计占重要性的76.6%，而地形变量占23.5%。2023年厄尔尼诺年（El Niño year）高洪水区与低洪水区之间的差异最大（+0.0364）；由于每个ENSO相位仅有1年数据，我们将其视为基于个案的比较，而非一般性的ENSO响应。阈值分析识别出两个不同的值：操作性截断值为NDVI=0.05（总体精度82.67%），生态过渡值约为0.25~0.30。空间聚类较弱但显著（Moran’s I=0.2179，p\u003C0.001）。空间分块交叉验证的精度（0.597）低于随机交叉验证（0.627），表明数据中存在空间自相关。高洪水区的脆弱性指数是低洪水区的1.67倍（0.4306对0.2573）。这些格局支持差异化治理：基于海拔的分区、根据当地洪水情势校准的预警系统，以及在热点区域的重点干预。",null,"Symmetry","2026-09-25T00:00:00Z","论文",10,false,70,{"impact":17,"substance":18,"depth":19,"authority":20,"freshness":21,"relevant":22,"comment":23},8,22,18,13,9,1,"基于Sentinel-2与机器学习的多时相洪涝-植被响应研究，数据规模大、结论具体，对农业遥感与灾害风险管理有参考价值，但属区域案例研究，公共影响有限。",[25],{"name":10,"url":6},[27,28,29,30,31],"智慧农业","农业人工智能","遥感监测","洪涝灾害","植被恢复",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},"I2","agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai","I",[39],{"code":40,"slug":41,"name":42,"path":43,"parent_code":37},"I3","remote-sensing","遥感与农情监测","\u002Ftopic\u002Fdigital-agriculture\u002Fremote-sensing",{"status":45,"locked":14,"version":46,"method":47,"evidence":48},"classified","2026-10-22.2","rules",{"candidates":49,"agricultural_evidence":52},[50,58,63,71],{"code":33,"score":51,"supported":52,"title":53,"abstract":55,"tags":57},15,true,[54],"machine learning",[56,54],"机器学习",[28],{"code":40,"score":13,"supported":52,"title":59,"abstract":61,"tags":62},[60],"remote sensing",[],[31,29],{"code":64,"score":65,"supported":52,"title":66,"abstract":67,"tags":70},"H7",7,[],[68,69],"干旱","drought",[30],{"code":72,"score":22,"supported":14,"title":73,"abstract":74,"tags":75},"I1",[],[],[27],[77,78],"湄公河支流 流域 遥感 洪水","Sentinel-2 NDVI 洪涝 植被","湄公河支流流域遥感洪水-3550",3,"10.3390\u002Fsym18101606",{"doi":81,"openalex_id":83,"authors":84,"venue":10,"cited_by_count":96,"oa_url":6,"card":97,"direction":101,"ingested_from":103},"W7214283372",[85,87,90,93],{"name":86,"orcid":9},"Jiradech Majandang",{"name":88,"orcid":89},"Patiwat Littidej","https:\u002F\u002Forcid.org\u002F0000-0002-1024-547X",{"name":91,"orcid":92},"Benjamabhorn Pumhirunroj","https:\u002F\u002Forcid.org\u002F0009-0009-6607-5594",{"name":94,"orcid":95},"D. C. Slack","https:\u002F\u002Forcid.org\u002F0000-0003-0324-2163",0,{"tldr":98,"method":99,"finding":100,"direction":101,"opportunity":102},"利用多时相遥感和机器学习揭示泰国湄公河支流流域植被对洪水的非对称响应。","Sentinel-2影像、洪水记录与机器学习，分析23万个空间单元。","重复洪水区植被增加，低洪水区减少，表明洪水可缓冲干旱影响。","农业遥感与作物表型","可探索洪水-干旱交替下植被恢复机制，并开发基于阈值的差异化预警系统。","openalex","2026-09-26T23:30:29.514361Z",{"total":106,"page":22,"page_size":106,"items":107},6,[108,191,288,339,415,505],{"id":109,"title":110,"url":111,"summary":112,"summary_zh":113,"content":9,"source_name":114,"source_url":111,"published_at":115,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":116,"score_detail":117,"sources":121,"tags":123,"primary_topic":126,"secondary_topics":132,"classification":135,"search_phrases":164,"slug":167,"view_count":22,"doi":168,"paper":169,"created_at":190},4808,"Early Detection of Apple Scab in Orchards and Mildew in Vineyards Using A Novel Deep Learning Model and Multispectral Imagery","https:\u002F\u002Fdoi.org\u002F10.15832\u002Fankutbd.1898342","Agricultural diseases are a significant problem for growers as they cause substantial damage to fruits and leaves, leading to yield losses. Therefore, developing systems for early detection of plant diseases is a crucial issue. In this study, a new deep learning-based hybrid model is presented for the detection and segmentation of plant diseases, specifically apple scab disease in apple orchards and mildew disease in vineyards. Multispectral imagery data collected in agricultural areas of Tekirdag province in Turkey via drones between May 2023 and December 2024 were used in developing the early disease detection system. The datasets contain 3625 images of apple orchards and 5625 images of vineyards. DJI Phantom 4 Multispectral drone, equipped with five spectral bands (Blue, Green, Red, Red Edge, and Near Infrared), was utilized to capture both spectral richness and temporal progression. This study introduces a novel deep learning architecture developed specifically for time series-based disease detection and segmentation using multispectral drone imagery. Changes in plants over time provide important information about the occurrence of plant diseases. For this reason, time series were used in this study to extract features from the changes in plants over time. The proposed model architecture was specifically designed to leverage time series information, enabling robust tracking of crop health changes over time. The model trained with the dataset collected for apple diseases achieved an average Intersection Over Unit (IoU) value of 82.05%, while the model trained with the dataset collected for vineyard diseases reached an IoU value of 79.67%. The findings highlight the effectiveness of integrating drone-acquired multispectral time-series data with advanced deep learning models for scalable, accurate, and timely plant disease detection in precision agriculture.","农业病害是种植者面临的重大问题，会对果实和叶片造成严重损害，导致产量损失。因此，开发植物病害早期检测系统是一个至关重要的问题。本研究提出了一种基于深度学习的新型混合模型，用于植物病害的检测与分割，具体针对苹果园中的苹果黑星病和葡萄园中的霜霉病。该早期病害检测系统的开发使用了2023年5月至2024年12月期间通过无人机在土耳其泰基尔达省农业区域采集的多光谱影像数据。数据集包含3625幅苹果园图像和5625幅葡萄园图像。所用DJI Phantom 4多光谱无人机配备五个光谱波段（蓝、绿、红、红边和近红外），用于捕捉光谱丰富度和时间变化过程。本研究提出了一种专门开发的深度学习架构，用于基于时间序列的多光谱无人机影像病害检测与分割。植物随时间的变化提供了关于植物病害发生的重要信息。因此，本研究使用时间序列从植物随时间的变化中提取特征。所提出的模型架构专门设计以利用时间序列信息，从而能够稳健地跟踪作物健康状况随时间的变化。使用苹果病害数据集训练的模型达到了82.05%的平均交并比（IoU）值，而使用葡萄园病害数据集训练的模型达到了79.67%的IoU值。研究结果凸显了将无人机获取的多光谱时间序列数据与先进深度学习模型相结合，在精准农业中实现可扩展、准确且及时的植物病害检测方面的有效性。","Tarım Bilimleri Dergisi","2026-10-12T00:00:00Z",77,{"impact":118,"substance":119,"depth":19,"authority":20,"freshness":21,"relevant":22,"comment":120},16,21,"该研究提出基于多光谱时序影像的新型深度学习模型，实现果园与葡萄园病害早期检测，数据规模大、方法新颖，对精准农业病害预警具有参考价值。",[122],{"name":114,"url":111},[27,124,28,29,125],"无人机","病害预警",{"code":127,"slug":128,"name":129,"path":130,"parent_code":131},"B2","fruit","果树","\u002Ftopic\u002Fhorticulture\u002Ffruit","B",[133,134],{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},{"code":40,"slug":41,"name":42,"path":43,"parent_code":37},{"status":45,"locked":14,"version":46,"method":47,"evidence":136},{"candidates":137,"agricultural_evidence":52},[138,146,152,158],{"code":127,"score":139,"supported":52,"title":140,"abstract":142,"tags":145},17,[141],"apple",[143,144,141],"苹果","葡萄",[],{"code":33,"score":51,"supported":52,"title":147,"abstract":149,"tags":151},[148],"deep learning",[150,148],"深度学习",[28],{"code":40,"score":153,"supported":52,"title":154,"abstract":156,"tags":157},12,[155],"multispectral",[155],[29],{"code":72,"score":65,"supported":52,"title":159,"abstract":160,"tags":163},[],[161,162],"精准农业","precision agriculture",[27],[165,166],"苹果黑星病 无人机 多光谱","葡萄园 霜霉病 深度学习","苹果黑星病无人机多光谱-4808","10.15832\u002Fankutbd.1898342",{"doi":168,"openalex_id":170,"authors":171,"venue":114,"cited_by_count":96,"oa_url":183,"card":184,"direction":189,"ingested_from":103},"W7221558726",[172,175,178,180],{"name":173,"orcid":174},"Batuhan Aşıroğlu","https:\u002F\u002Forcid.org\u002F0000-0003-0767-6348",{"name":176,"orcid":177},"Aydın Kağan Turkoglu","https:\u002F\u002Forcid.org\u002F0009-0001-8383-8807",{"name":179,"orcid":9},"Fatih Sinan Bir",{"name":181,"orcid":182},"Sibel Senan","https:\u002F\u002Forcid.org\u002F0000-0001-6773-0428","https:\u002F\u002Fdergipark.org.tr\u002Fen\u002Fdownload\u002Farticle-file\u002F5761152",{"tldr":185,"method":186,"finding":187,"direction":101,"opportunity":188},"提出一种基于无人机多光谱时序影像的深度学习混合模型，实现苹果黑星病和葡萄霉病的早期检测与分割。","DJI Phantom 4多光谱无人机采集时序影像，构建时序深度学习分割模型。","苹果病IoU达82.05%，葡萄病达79.67%，验证多光谱时序数据有效性。","可探索多作物、多病害的时序多光谱通用模型，并融合气象与地面数据提升早期预警能力。","智慧农业 \u002F 农业物联网","2026-10-13T23:30:10.242467Z",{"id":192,"title":193,"url":194,"summary":195,"summary_zh":196,"content":9,"source_name":197,"source_url":194,"published_at":198,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":199,"score_detail":200,"sources":202,"tags":204,"primary_topic":208,"secondary_topics":209,"classification":211,"search_phrases":260,"slug":263,"view_count":96,"doi":264,"paper":265,"created_at":287},5582,"AI-Based RGB Imaging and 14-Channel Multispectral Food Quality Scanner","https:\u002F\u002Fdoi.org\u002F10.22214\u002Fijraset.2026.85255","Food-quality assessment plays an important role in food safety, agricultural processing, storage, and supply-chain management. Conventional inspection methods depend on visual judgement or laboratory-based testing, which may require considerable time, specialised equipment, and manual effort. Recent developments in computer vision, multispectral sensing, and artificial intelligence (AI) provide opportunities for automated, non-destructive food-quality assessment. This paper reviews the application of RGB imaging, multispectral sensing, machine learning, and deep learning for fruit-ripeness estimation, surfacedefect detection, grain classification, and food-authenticity assessment. The reviewed literature demonstrates that RGB imaging can extract visible features such as colour, shape, and texture, whereas multispectral sensing provides wavelength-dependent information associated with food characteristics. Published studies have reported promising results in tomato-ripeness classification, apple-defect detection, rice-grain analysis, and honey-adulteration screening. Based on these findings, a compact food-quality scanner integrating an RGB camera, a 14-channel multispectral sensor, controlled illumination, and an embedded processing unit is proposed. The review discusses sensor integration, data preprocessing, feature fusion, classification methods, evaluation metrics, and implementation challenges. It also identifies the need for broader datasets, standardised illumination, independent validation, and resource-efficient AI models to improve the reliability of portable food-inspection systems.","食品质量评估在食品安全、农产品加工、储运和供应链管理中发挥着重要作用。传统检测方法依赖人工目视判断或实验室检测，往往需要较长时间、专用设备和人工投入。近年来，计算机视觉、多光谱传感和人工智能（AI）的发展为自动化、无损食品质量评估提供了新的机遇。本文综述了RGB成像、多光谱传感、机器学习和深度学习在果实成熟度估计、表面缺陷检测、谷物分类和食品真实性评估中的应用。所综述的文献表明，RGB成像能够提取颜色、形状和纹理等可见特征，而多光谱传感则可提供与食品特性相关的波长依赖信息。已发表研究在番茄成熟度分类、苹果缺陷检测、稻米籽粒分析和蜂蜜掺假筛查等方面取得了良好结果。基于上述研究结果，本文提出一种集成了RGB相机、14通道多光谱传感器、受控照明和嵌入式处理单元的紧凑型食品质量扫描仪。综述讨论了传感器集成、数据预处理、特征融合、分类方法、评价指标和实现挑战，并指出需要更广泛的数据集、标准化照明、独立验证和资源高效的AI模型，以提高便携式食品检测系统的可靠性。","International Journal for Research in Applied Science and Engineering Technology","2026-10-21T00:00:00Z",65,{"impact":153,"substance":19,"depth":118,"authority":21,"freshness":13,"relevant":22,"comment":201},"综述性论文，提出集成RGB与14通道多光谱的便携食品质量扫描仪方案，方法框架清晰但缺乏自有实验数据，属细分领域技术进展，可作参考但非重大突破。",[203],{"name":197,"url":194},[27,28,205,206,207],"无损检测","多光谱成像","食品质量检测",{"code":40,"slug":41,"name":42,"path":43,"parent_code":37},[210],{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},{"status":45,"locked":14,"version":46,"method":47,"evidence":212},{"candidates":213,"agricultural_evidence":52},[214,218,225,232,239,246,250,256],{"code":40,"score":153,"supported":52,"title":215,"abstract":216,"tags":217},[155],[155],[206],{"code":33,"score":13,"supported":52,"title":219,"abstract":220,"tags":224},[],[150,56,221,54,148,222,223],"计算机视觉","artificial intelligence","computer vision",[28],{"code":226,"score":65,"supported":52,"title":227,"abstract":228,"tags":231},"E4",[],[229,230],"食品安全","food safety",[205],{"code":233,"score":106,"supported":52,"title":234,"abstract":235,"tags":238},"A1",[],[236,237],"稻米","rice",[],{"code":240,"score":106,"supported":52,"title":241,"abstract":242,"tags":245},"B1",[],[243,244],"番茄","tomato",[],{"code":127,"score":106,"supported":52,"title":247,"abstract":248,"tags":249},[],[143,141],[],{"code":251,"score":80,"supported":52,"title":252,"abstract":253,"tags":255},"E1",[],[254],"农产品加工",[],{"code":72,"score":22,"supported":14,"title":257,"abstract":258,"tags":259},[],[],[27],[261,262],"RGB成像 多光谱 食品质量","14通道 多光谱 扫描仪","RGB成像多光谱食品质量-5582","10.22214\u002Fijraset.2026.85255",{"doi":264,"openalex_id":266,"authors":267,"venue":197,"cited_by_count":96,"oa_url":194,"card":280,"direction":286,"ingested_from":103},"W7223139905",[268,270,272,274,276,278],{"name":269,"orcid":9},"Eshwari Take",{"name":271,"orcid":9},"Avanish Raut",{"name":273,"orcid":9},"Vedant Joshi",{"name":275,"orcid":9},"Shravani Akolkar",{"name":277,"orcid":9},"R. M. Gharat",{"name":279,"orcid":9},"Shruti Dive",{"tldr":281,"method":282,"finding":283,"direction":284,"opportunity":285},"综述RGB与多光谱成像结合AI的食品品质无损检测，并提出一款集成14通道多光谱传感器的紧凑型扫描仪方","综述RGB成像、多光谱传感与机器学习\u002F深度学习在食品检测中的应用，并提出传感器集","RGB与多光谱融合在番茄成熟度、苹果缺陷、稻米分类和蜂蜜掺假检测中表现良好。","农业人工智能与决策模型","可研究多源光谱特征融合与轻量化模型，构建标准化数据集和便携设备独立验证体系。","数字乡村与农业信息化","2026-10-22T11:20:48.684398Z",{"id":289,"title":290,"url":291,"summary":292,"summary_zh":293,"content":9,"source_name":294,"source_url":291,"published_at":198,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":295,"score_detail":296,"sources":299,"tags":301,"primary_topic":305,"secondary_topics":306,"classification":308,"search_phrases":323,"slug":326,"view_count":96,"doi":327,"paper":328,"created_at":338},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)",67,{"impact":17,"substance":297,"depth":139,"authority":153,"freshness":13,"relevant":22,"comment":298},20,"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。",[300],{"name":294,"url":291},[27,28,302,303,304],"产量预测","遥感","可解释性",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[307],{"code":40,"slug":41,"name":42,"path":43,"parent_code":37},{"status":45,"locked":14,"version":46,"method":47,"evidence":309},{"candidates":310,"agricultural_evidence":52},[311,315,319],{"code":33,"score":153,"supported":52,"title":312,"abstract":313,"tags":314},[223],[221],[28],{"code":40,"score":17,"supported":52,"title":316,"abstract":317,"tags":318},[],[303,60],[302,303],{"code":72,"score":22,"supported":14,"title":320,"abstract":321,"tags":322},[],[],[27],[324,325],"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":327,"openalex_id":329,"authors":330,"venue":294,"cited_by_count":96,"oa_url":291,"card":333,"direction":101,"ingested_from":103},"W7223114002",[331],{"name":332,"orcid":9},"Kadiatou Barry",{"tldr":334,"method":335,"finding":336,"direction":101,"opportunity":337},"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":340,"title":341,"url":342,"summary":343,"summary_zh":344,"content":9,"source_name":345,"source_url":342,"published_at":346,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":347,"score_detail":348,"sources":350,"tags":352,"primary_topic":355,"secondary_topics":356,"classification":364,"search_phrases":389,"slug":392,"view_count":96,"doi":393,"paper":394,"created_at":414},5006,"Mapping Soil Salinity in Arid Agricultural Regions Using Multi-Source Remote Sensing and a Long-Range Dependency-Enhanced Deep Learning Framework","https:\u002F\u002Fdoi.org\u002F10.3390\u002Fland15101966","Soil salinity estimation across arid agricultural regions is challenging due to pronounced environmental heterogeneity and complex dependencies among multi-source variables. This study proposes a Long-Range Dependency-Enhanced Network (LRD-Net) for regional-scale soil salinity estimation in multiple agricultural oasis regions using multi-source remote sensing and environmental covariates. A dataset comprising 180 field samples from three agricultural oasis regions in Xinjiang, China, was constructed with 25 covariates derived from spectral, topographic, climatic, and soil property data. LRD-Net consists of three stages: a multi-scale feature fusion backbone based on depthwise separable convolutions and cross-scale interaction; a Long-Range Dependency Enhancement Module (LDEM) with an asymmetric dual-branch architecture that captures forward and reverse-aware dependencies through reverse alignment and cross-directional fusion; and multi-scale attention refinement with a regression head for continuous salinity estimation. Under the region-stratified random-split evaluation, LRD-Net achieved an R2 of 0.905, an RMSE of 1.207 g·kg−1, and an MAE of 0.868 g·kg−1. Performance decreased under more spatially independent validation, with an R2 of 0.874 under spatial cross-validation and LORO test R2 values ranging from 0.739 to 0.801, indicating that prediction across spatially and geographically independent samples is more challenging than region-stratified random-split evaluation. The proposed framework effectively integrates multi-source information and captures complementary long-range dependencies, providing a potential approach for regional-scale soil salinity estimation within the studied arid agricultural oasis environments and supporting soil salinization monitoring and sustainable land management in the study region.","干旱农业区域土壤盐分估算面临显著的环境异质性和多源变量间复杂依赖关系的挑战。本研究提出了一种长程依赖增强网络（LRD-Net），利用多源遥感与环境协变量对多个农业绿洲区域进行区域尺度土壤盐分估算。研究构建了包含中国新疆三个农业绿洲区域180个野外样本的数据集，并衍生出25个协变量，涵盖光谱、地形、气候和土壤属性数据。LRD-Net由三个阶段组成：基于深度可分离卷积和跨尺度交互的多尺度特征融合骨干网络；具有非对称双分支架构的长程依赖增强模块（LDEM），通过反向对齐和跨方向融合捕获前向与反向感知依赖关系；以及多尺度注意力细化与回归头，用于连续盐分估算。在区域分层随机划分评估下，LRD-Net取得了R²为0.905、RMSE为1.207 g·kg⁻¹、MAE为0.868 g·kg⁻¹的性能。在空间独立性更强的验证下性能有所下降，空间交叉验证R²为0.874，留一区域测试R²范围为0.739至0.801，表明跨空间和地理独立样本的预测比区域分层随机划分评估更具挑战性。所提出的框架有效整合了多源信息并捕获了互补的长程依赖关系，为研究区干旱农业绿洲环境下的区域尺度土壤盐分估算提供了潜在方法，并支持研究区土壤盐渍化监测与可持续土地管理。","Land","2026-10-14T00:00:00Z",72,{"impact":153,"substance":119,"depth":139,"authority":20,"freshness":21,"relevant":22,"comment":349},"方法新颖、数据扎实的遥感土壤盐渍化估算研究，对干旱区盐碱地监测与治理有参考价值，但属细分领域学术进展，公共影响有限。",[351],{"name":345,"url":342},[27,28,303,353,354],"土壤盐渍化","新疆农业",{"code":40,"slug":41,"name":42,"path":43,"parent_code":37},[357,363],{"code":358,"slug":359,"name":360,"path":361,"parent_code":362},"H2","soil","土壤健康与改良","\u002Ftopic\u002Fenvironment\u002Fsoil","H",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},{"status":45,"locked":14,"version":46,"method":47,"evidence":365},{"candidates":366,"agricultural_evidence":52},[367,371,377,381,385],{"code":40,"score":51,"supported":52,"title":368,"abstract":369,"tags":370},[60],[303,60],[303],{"code":358,"score":372,"supported":52,"title":373,"abstract":374,"tags":376},14,[359],[375,359],"土壤",[],{"code":33,"score":21,"supported":52,"title":378,"abstract":379,"tags":380},[148],[],[28],{"code":64,"score":80,"supported":52,"title":382,"abstract":383,"tags":384},[],[68],[],{"code":72,"score":22,"supported":14,"title":386,"abstract":387,"tags":388},[],[],[27],[390,391],"新疆 农业绿洲 土壤盐渍化 遥感","LRD-Net 土壤盐分 估算","新疆农业绿洲土壤盐渍化遥感-5006","10.3390\u002Fland15101966",{"doi":393,"openalex_id":395,"authors":396,"venue":345,"cited_by_count":96,"oa_url":342,"card":409,"direction":101,"ingested_from":103},"W7221816183",[397,399,402,404,406],{"name":398,"orcid":9},"Mingjie Qian",{"name":400,"orcid":401},"Liangjun Wang","https:\u002F\u002Forcid.org\u002F0009-0006-8718-8326",{"name":403,"orcid":9},"Peng Wang",{"name":405,"orcid":9},"Hangyuan Liu",{"name":407,"orcid":408},"Weitao Chen","https:\u002F\u002Forcid.org\u002F0000-0002-6272-1618",{"tldr":410,"method":411,"finding":412,"direction":101,"opportunity":413},"提出LRD-Net深度学习框架，融合多源遥感与环境协变量，估算新疆干旱绿洲区土壤盐分。","构建180个样点、25个协变量的数据集，设计含长程依赖增强模块的多尺度网络。","区域分层随机划分下R²达0.905，但空间独立验证R²降至0.739-0.874，跨区域预测更难。","可探索跨区域迁移学习与不确定性量化，提升空间独立场景下盐分制图泛化能力。","2026-10-15T23:30:29.667342Z",{"id":416,"title":417,"url":418,"summary":419,"summary_zh":420,"content":9,"source_name":421,"source_url":418,"published_at":346,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":422,"score_detail":423,"sources":425,"tags":427,"primary_topic":429,"secondary_topics":430,"classification":433,"search_phrases":478,"slug":481,"view_count":22,"doi":482,"paper":483,"created_at":504},4988,"Hierarchical cross-modal spatiotemporal attention network for crop yield prediction in complex Himalayan agroecosystems","https:\u002F\u002Fdoi.org\u002F10.3389\u002Ffpls.2026.1933033","Introduction Accurate crop yield prediction is essential for precision agriculture, climate-resilient farm management, and timely decision-making. However, yield prediction remains challenging because of complex, nonlinear, and multi-scale interactions among crop physiology, soil conditions, weather variability, and management practices. Conventional statistical, machine learning, and deep learning approaches, including convolutional neural networks (CNNs), recurrent neural networks (RNNs), and long short-term memory (LSTM) networks, often have limitations in capturing long-range temporal dependencies, hierarchical spatial relationships, heterogeneous data modalities, and predictive uncertainty. Methods This study proposes a Hierarchical Cross-Modal Spatio-Temporal Attention Network (HCM-STAN) for crop yield prediction using a unified attention-based framework for integrating heterogeneous agro-environmental data. The architecture incorporates four principal components: (i) hierarchical spatio-temporal attention operating across tree, block, orchard, and regional scales; (ii) a cross-modal attention module for adaptive fusion of IoT-based soil sensor observations, weather data, plant physiological measurements, and drone-based multispectral imagery; (iii) a soil-aware attention head that explicitly models soil memory, hysteresis effects, and root-zone dynamics; and (iv) an uncertainty-aware prediction module combining Monte Carlo dropout and quantile regression to generate calibrated prediction intervals. The framework was evaluated using a multi-source dataset collected from Himalayan apple orchards during 2023–2026, comprising 1,247 monitored trees across three commercial orchards. Results The proposed HCM-STAN achieved an R 2 of 0.89 for four-week-ahead crop yield prediction, representing a 15.6% improvement over the CNN-LSTM baseline. Furthermore, zero-shot transfer to mango orchards achieved an R 2 of 0.67, corresponding to approximately 75% of the performance obtained under full training. Discussion The results demonstrate that integrating hierarchical spatio-temporal representation learning, cross-modal attention, soil-aware modelling, and uncertainty estimation can effectively address the heterogeneous and dynamic nature of crop yield prediction in complex mountain agroecosystems. The observed zero-shot transfer performance further indicates the potential of HCM-STAN to generalize across crop systems without crop-specific retraining. Overall, the proposed framework provides a promising foundation for reliable, uncertainty-aware yield forecasting and can support precision agriculture, climate-smart farm management, and data-driven decision-making in heterogeneous agroecosystems.","引言 准确的作物产量预测对于精准农业、气候适应性农场管理以及及时决策至关重要。然而，由于作物生理、土壤条件、天气变异和管理措施之间复杂、非线性且多尺度的相互作用，产量预测仍然具有挑战性。传统的统计方法、机器学习以及深度学习方法，包括卷积神经网络（CNN）、循环神经网络（RNN）和长短期记忆（LSTM）网络，在捕捉长程时间依赖关系、层次化空间关系、异构数据模态以及预测不确定性方面往往存在局限。方法 本研究提出了一种用于作物产量预测的层次化跨模态时空注意力网络（HCM-STAN），采用统一的基于注意力的框架来整合异构农业环境数据。该架构包含四个主要组成部分：（i）在树、地块、果园和区域尺度上运行的层次化时空注意力；（ii）用于自适应融合基于物联网的土壤传感器观测、天气数据、植物生理测量和无人机多光谱影像的跨模态注意力模块；（iii）显式建模土壤记忆、滞后效应和根区动态的土壤感知注意力头；（iv）结合蒙特卡洛 dropout 和分位数回归以生成校准预测区间的不确定性感知预测模块。该框架使用2023—2026年在喜马拉雅苹果园采集的多源数据集进行了评估，该数据集包含三个商业果园中1，247棵监测树木。结果 所提出的 HCM-STAN 在提前四周作物产量预测中达到了0.89的 R²，相较于 CNN-LSTM 基线提升了15.6%。此外，零样本迁移至芒果园达到了0.67的 R²，相当于完整训练下所获性能的约75%。讨论 结果表明，整合层次化时空表征学习、跨模态注意力、土壤感知建模和不确定性估计，能够有效应对复杂山地农业生态系统中作物产量预测的异构性和动态性。观察到的零样本迁移性能进一步表明，HCM-STAN 具有在无需作物特定再训练的情况下跨作物系统泛化的潜力。总体而言，该 p","Frontiers in Plant Science",81,{"impact":19,"substance":18,"depth":19,"authority":372,"freshness":21,"relevant":22,"comment":424},"提出分层跨模态时空注意力网络，融合物联网土壤传感、气象、生理与无人机多光谱数据，苹果园四周产量预测R²达0.89并实现芒果园零样本迁移，方法新颖、数据扎实，对智慧农业产量预报有参考价值。",[426],{"name":421,"url":418},[27,28,302,428,303],"多模态融合",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[431,432],{"code":127,"slug":128,"name":129,"path":130,"parent_code":131},{"code":40,"slug":41,"name":42,"path":43,"parent_code":37},{"status":45,"locked":14,"version":46,"method":47,"evidence":434},{"candidates":435,"agricultural_evidence":52},[436,441,447,452,456,460,467,473],{"code":33,"score":437,"supported":52,"title":438,"abstract":439,"tags":440},11,[],[150,56,54,148],[28,428],{"code":127,"score":21,"supported":52,"title":442,"abstract":443,"tags":446},[],[143,444,445,141],"芒果","orchard",[],{"code":40,"score":17,"supported":52,"title":448,"abstract":449,"tags":451},[],[450,155],"无人机多光谱",[302,303],{"code":72,"score":65,"supported":52,"title":453,"abstract":454,"tags":455},[],[161,162],[27],{"code":358,"score":106,"supported":52,"title":457,"abstract":458,"tags":459},[],[375,359],[],{"code":461,"score":106,"supported":52,"title":462,"abstract":463,"tags":466},"I5",[],[464,465],"物联网","IoT",[],{"code":468,"score":80,"supported":52,"title":469,"abstract":470,"tags":472},"H6",[],[471],"农业生态",[],{"code":64,"score":80,"supported":52,"title":474,"abstract":475,"tags":477},[],[476],"气候适应",[],[479,480],"喜马拉雅 苹果园 产量预测","HCM-STAN 作物产量预测","喜马拉雅苹果园产量预测-4988","10.3389\u002Ffpls.2026.1933033",{"doi":482,"openalex_id":484,"authors":485,"venue":421,"cited_by_count":96,"oa_url":418,"card":499,"direction":189,"ingested_from":103},"W7221795850",[486,489,491,493,495,497],{"name":487,"orcid":488},"Shabir Ahmad Mir","https:\u002F\u002Forcid.org\u002F0000-0003-2311-218X",{"name":490,"orcid":9},"Umarfarooque Momin",{"name":492,"orcid":9},"Sudeep Kumar E",{"name":494,"orcid":9},"Itfaq Ahmad Mir",{"name":496,"orcid":9},"Ab Rouf Khan",{"name":498,"orcid":9},"Ghulam Mohammad Mir",{"tldr":500,"method":501,"finding":502,"direction":284,"opportunity":503},"提出分层跨模态时空注意力网络HCM-STAN，融合多源农业数据预测喜马拉雅山区苹果产量。","用IoT土壤、气象、生理和无人机多光谱数据，结合分层时空与跨模态注意力及不确定性","四周前产量预测R²达0.89，比CNN-LSTM提升15.6%，零样本迁移芒果R²为0.67。","可探索跨作物零样本迁移的领域自适应机制，并验证模型在更多山地农业生态系统的泛化能力。","2026-10-15T23:30:08.750140Z",{"id":506,"title":507,"url":508,"summary":509,"summary_zh":510,"content":9,"source_name":511,"source_url":508,"published_at":512,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":422,"score_detail":513,"sources":515,"tags":517,"primary_topic":521,"secondary_topics":522,"classification":524,"search_phrases":535,"slug":538,"view_count":96,"doi":539,"paper":540,"created_at":560},4897,"Wild boar collision data and satellite computer vision refine habitat suitability mapping across France","https:\u002F\u002Fdoi.org\u002F10.1007\u002Fs10344-026-02160-1","Abstract Wild boar (Sus scrofa) populations have expanded rapidly across Europe, leading to escalating human-wildlife conflicts (HWCs), notably wildlife-vehicle collisions (WVCs), increased agricultural damage and disease transmission. In continental France these issues are compounded by the species’ ecological adaptability exhibiting increasing overlaps with urbanization and transportation networks. In this study, we leverage time-series environmental data, computer vision and Species Distribution Modeling (SDM) to predict wild boar habitat suitability and investigate its spatiotemporal drivers. We use presence-only data from WVC reports on the national railway and road networks, along with publicly available GBIF observation collections, as complementary occurrence streams in our SDMs, while addressing inherent sampling biases of these datasets with tailored corrections. A key innovation of this study is the integration of national scale, annually updated high-resolution land cover predictors explicitly focused on wild boar resource preference. By fine-tuning a multitemporal Vision Transformer foundational AI model on multispectral satellite remote sensing imagery we capture subtle seasonal phenological differences. Our results highlight clear spatial, seasonal and annual variations in wild boar habitat suitability. The multitemporal SDM pipeline yields meaningful predictions when extrapolating to novel environmental scenarios such as climate extremes. Adding rail and road WVC records to GBIF-based SDMs retained high performance metrics while altering predicted spatial patterns and feature importance distribution. The methodological and ecological insights gained through this study provide additional knowledge for French transportation planning, agriculture and wildlife management. Identifying regions with high seasonal habitat suitability can inform targeted interventions. More broadly, our results demonstrate that advanced, data-driven methods are becoming indispensable for proactively and sustainably addressing HWCs in an increasingly anthropogenic world.","摘要 野猪（Sus scrofa）种群在欧洲迅速扩张，导致人兽冲突（HWCs）不断升级，尤其是野生动物车辆碰撞（WVCs）、农业损害加剧和疾病传播。在法国大陆，该物种的生态适应性使其与城市化和交通网络的重叠日益增加，进一步加剧了这些问题。本研究利用时间序列环境数据、计算机视觉和物种分布模型（SDM）预测野猪栖息地适宜性，并探究其时空驱动因素。我们使用国家铁路和公路网络中WVC报告的存在-only数据，以及公开可用的GBIF观测集合，作为SDM中的互补出现数据流，同时通过针对性校正处理这些数据集固有的采样偏差。本研究的一个关键创新是整合了国家尺度、年度更新的高分辨率土地覆盖预测因子，明确聚焦于野猪的资源偏好。通过在多光谱卫星遥感影像上微调多时相视觉Transformer基础AI模型，我们捕捉到了细微的季节物候差异。我们的结果突出了野猪栖息地适宜性在空间、季节和年际上的明显变化。多时相SDM流程在推断至气候极端等新型环境情景时产生了有意义的预测。将铁路和公路WVC记录添加到基于GBIF的SDM中，在保持高性能指标的同时改变了预测的空间格局和特征重要性分布。本研究获得的方法学和生态学见解为法国交通规划、农业和野生动物管理提供了额外知识。识别具有高季节性栖息地适宜性的区域可为针对性干预提供依据。更广泛地说，我们的结果表明，在日益受到人类影响的世界中，先进的数据驱动方法正成为前瞻性和可持续地应对人兽冲突不可或缺的手段。","European Journal of Wildlife Research","2026-10-13T00:00:00Z",{"impact":19,"substance":18,"depth":19,"authority":372,"freshness":21,"relevant":22,"comment":514},"该研究将卫星遥感与计算机视觉用于野猪栖息地建模，方法新颖、数据规模大，对交通规划与农业野生动物管理有参考价值，但属法国区域研究，国内产业影响有限。",[516],{"name":511,"url":508},[28,29,518,519,520],"栖息地评估","野生动物管理","人兽冲突",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[523],{"code":40,"slug":41,"name":42,"path":43,"parent_code":37},{"status":45,"locked":14,"version":46,"method":47,"evidence":525},{"candidates":526,"agricultural_evidence":52},[527,531],{"code":33,"score":51,"supported":52,"title":528,"abstract":529,"tags":530},[223],[221,223],[28],{"code":40,"score":13,"supported":52,"title":532,"abstract":533,"tags":534},[],[303,60,155],[29],[536,537],"法国 野猪 栖息地 遥感","野猪 车辆碰撞 卫星影像","法国野猪栖息地遥感-4897","10.1007\u002Fs10344-026-02160-1",{"doi":539,"openalex_id":541,"authors":542,"venue":511,"cited_by_count":96,"oa_url":508,"card":555,"direction":101,"ingested_from":103},"W7221702437",[543,546,548,551,553],{"name":544,"orcid":545},"Adrian Meyer","https:\u002F\u002Forcid.org\u002F0000-0002-8652-1477",{"name":547,"orcid":9},"Théo Reibel",{"name":549,"orcid":550},"Кevin Мorelle","https:\u002F\u002Forcid.org\u002F0000-0001-9041-6541",{"name":552,"orcid":9},"Mathias Kneubühler",{"name":554,"orcid":9},"Denis Jordan",{"tldr":556,"method":557,"finding":558,"direction":101,"opportunity":559},"融合野猪碰撞数据与卫星视觉模型，预测法国野猪栖息地适宜性及其时空驱动因素。","时间序列环境数据、Vision Transformer卫星影像、物种分布模型与G","栖息地适宜性呈明显时空变化，加入碰撞记录改变预测格局与特征重要性。","可将碰撞数据与遥感表型结合，拓展至其他野生动物-农业冲突的时空预警与决策。","2026-10-14T23:30:27.244673Z"]