[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-3277":3,"related-3277":60},{"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":23,"tags":25,"search_phrases":31,"slug":34,"view_count":35,"doi":36,"paper":37,"created_at":59},3277,"Downscaling of SMAP Soil Moisture Based on the Transformer Algorithm in Anhui Province","https:\u002F\u002Fdoi.org\u002F10.3390\u002Frs18193272","Soil moisture (SM) is critical for climate, water, and agriculture, but Soil Moisture Active Passive (SMAP) passive microwave products have coarse resolution, limiting regional applications. This study develops an SM downscaling framework based on Transformer and its variants (PatchTST and iTransformer), integrating multi-source satellite and groundwater data to generate 1 km daily SM products (2015–2022). Compared with Random Forest (RF), Long Short-Term Memory (LSTM), and Convolutional Neural Network–LSTM (CNN-LSTM), Transformer and its variants achieve superior accuracy and generalization. Validated against in situ measurements and SMCI1.0, the Transformer-downscaled SM product achieved the best accuracy with ubRMSE = 0.0372 m3\u002Fm3 and RMSE = 0.0591 m3\u002Fm3. The downscaled SM dataset not only captured finer spatial details but also preserved the spatial patterns and seasonal dynamics of the original SMAP product and showed good responsiveness to precipitation events. Feature importance analysis revealed that, aside from precipitation, the diurnal land surface temperature difference had a greater impact on SM than individual daytime or nighttime land surface temperature, ranking just below vegetation indices and soil texture factors, while groundwater level showed higher importance than elevation and surface temperature. This study confirms the effectiveness of Transformer-based models for SM spatial downscaling, providing a novel framework integrating remote sensing and deep hydrological information to generate accurate 1 km SM products.","土壤水分（SM）对气候、水资源和农业至关重要，但土壤水分主动被动（SMAP）被动微波产品分辨率较粗，限制了区域应用。本研究构建了一个基于Transformer及其变体（PatchTST和iTransformer）的土壤水分降尺度框架，融合多源卫星和地下水数据，生成1 km日尺度土壤水分产品（2015—2022年）。与随机森林（RF）、长短期记忆网络（LSTM）和卷积神经网络—长短期记忆网络（CNN-LSTM）相比，Transformer及其变体取得了更高的精度和泛化能力。利用站点实测数据和SMCI1.0进行验证，Transformer降尺度土壤水分产品精度最优，ubRMSE = 0.0372 m³\u002Fm³，RMSE = 0.0591 m³\u002Fm³。降尺度土壤水分数据集不仅捕捉到了更精细的空间细节，还保留了原始SMAP产品的空间格局和季节动态，并对降水事件表现出良好的响应。特征重要性分析表明，除降水外，昼夜地表温差对土壤水分的影响大于单独的白天或夜间地表温度，其重要性仅次于植被指数和土壤质地因子，而地下水埋深的重要性高于高程和地表温度。本研究证实了基于Transformer的模型在土壤水分空间降尺度中的有效性，为融合遥感和深层水文信息生成准确的1 km土壤水分产品提供了一种新框架。",null,"Remote Sensing","2026-09-22T00:00:00Z","论文",10,false,81,{"impact":17,"substance":18,"depth":17,"authority":19,"freshness":20,"relevant":21,"comment":22},18,22,14,9,1,"基于Transformer的SMAP土壤水分1km降尺度研究，方法新颖、验证充分，对区域农业旱情监测有实用价值。",[24],{"name":10,"url":6},[26,27,28,29,30],"智慧农业","农业人工智能","深度学习","遥感","土壤墒情",[32,33],"SMAP 土壤水分 降尺度","Transformer 土壤水分 安徽","SMAP土壤水分降尺度-3277",0,"10.3390\u002Frs18193272",{"doi":36,"openalex_id":38,"authors":39,"venue":10,"cited_by_count":35,"oa_url":6,"card":52,"direction":56,"ingested_from":58},"W7208807695",[40,42,44,46,48,50],{"name":41,"orcid":9},"Yuyang Fan",{"name":43,"orcid":9},"Jianwei Ma",{"name":45,"orcid":9},"Mengmeng Li",{"name":47,"orcid":9},"Changqing Ke",{"name":49,"orcid":9},"Bin Cheng",{"name":51,"orcid":9},"Zheng Duan",{"tldr":53,"method":54,"finding":55,"direction":56,"opportunity":57},"基于Transformer及变体融合多源卫星与地下水数据，将SMAP土壤湿度降尺度至1km日尺度。","Transformer、PatchTST、iTransformer，融合多源卫星","Transformer降尺度产品精度最优（ubRMSE=0.0372），保留原产品时空格局并响应降水","农业遥感与作物表型","可探索Transformer降尺度产品在区域干旱监测、灌溉决策及作物估产中的耦合应用。","openalex","2026-09-23T23:30:19.132307Z",{"total":61,"page":21,"page_size":61,"items":62},6,[63,104,134,161,201,229],{"id":64,"title":65,"url":66,"summary":67,"summary_zh":68,"content":9,"source_name":69,"source_url":66,"published_at":70,"category":12,"cover_url":9,"hotness":71,"is_selected":14,"score":72,"score_detail":73,"sources":78,"tags":82,"search_phrases":84,"slug":87,"view_count":35,"doi":88,"paper":89,"created_at":103},1534,"Deep Learning Based Semantic Segmentation of Satellite Imagery for Land Cover Classification","https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.22251467","Land cover classification from remote sensing imagery underpins agricultural monitoring, urban planning, disaster response, and environmental assessment, and deep learning has become the dominant paradigm for extracting land cover information from satellite scenes at pixel-level granularity. This survey reviews recent advances in deep learning-based land cover classification and semantic segmentation, organizing the literature into convolutional neural network (CNN)-based encoder–decoder architectures, attention and self-attention mechanisms, transformer and hybrid CNN–Transformer architectures, and optimization-augmented and foundation-model-adapted approaches. We summarize the datasets, evaluation metrics, and comparative performance reported across the reviewed studies, and use a representative hybrid EfficientNet–Vision Transformer framework as a case study to illustrate current design trends. Persistent challenges are identified, including class imbalance for narrow structures such as roads, the computational cost of hybrid transformer-based models, limited cross-dataset generalization, and the scarcity of densely annotated data. The survey concludes by outlining promising future directions, including lightweight hybrid architectures, self-supervised and weakly supervised learning, adaptation of large vision foundation models, multi-sensor data fusion, and explainable land cover mapping.","基于遥感影像的土地覆盖分类支撑着农业监测、城市规划、灾害响应和环境评估，而深度学习已成为从卫星场景中提取像素级土地覆盖信息的主导范式。本综述回顾了基于深度学习的土地覆盖分类与语义分割的最新进展，将相关文献组织为基于卷积神经网络（CNN）的编码器-解码器架构、注意力与自注意力机制、Transformer及CNN-Transformer混合架构，以及优化增强和基础模型适配方法等类别。我们总结了所评述研究中报告的数据集、评估指标和对比性能，并以一个具有代表性的混合EfficientNet-Vision Transformer框架作为案例研究，以说明当前的设计趋势。本文指出了持续存在的挑战，包括道路等狭窄结构的类别不平衡问题、基于混合Transformer模型的计算成本、跨数据集泛化能力有限，以及密集标注数据的稀缺性。综述最后概述了有前景的未来方向，包括轻量级混合架构、自监督和弱监督学习、大型视觉基础模型的适配、多传感器数据融合以及可解释的土地覆盖制图。","Zenodo (CERN European Organization for Nuclear Research)","2026-09-02T00:00:00Z",25,67,{"impact":74,"substance":75,"depth":17,"authority":13,"freshness":76,"relevant":21,"comment":77},12,20,7,"综述深度学习用于卫星影像土地覆盖分类，对农业监测有参考价值，但偏学术，影响有限。",[79,80],{"name":69,"url":66},{"name":69,"url":81},"https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.22251468",[26,27,28,29,83],"土地覆盖分类",[85,86],"农业人工智能 土地覆盖分类 智慧农业 深度学习","农业人工智能 土地覆盖分类","农业人工智能土地覆盖分类智慧农业深度学习-1534","10.5281\u002Fzenodo.22251467",{"doi":88,"openalex_id":90,"authors":91,"venue":69,"cited_by_count":35,"oa_url":66,"card":98,"direction":56,"ingested_from":58},"W7206172334",[92,94,96],{"name":93,"orcid":9},"Antony Daniel Rex. J",{"name":95,"orcid":9},"R. Vidya",{"name":97,"orcid":9},"A. Martin",{"tldr":99,"method":100,"finding":101,"direction":56,"opportunity":102},"综述深度学习用于卫星影像土地覆盖分类的语义分割方法，并分析挑战与未来方向。","综述CNN、Transformer及混合架构，案例为EfficientNet-V","混合架构性能优，但存在类别不平衡、计算成本高、泛化差等问题。","可探索轻量级混合架构或自监督学习，以提升农业土地覆盖分类的精度与泛化能力。","2026-09-03T23:30:30.230250Z",{"id":105,"title":106,"url":107,"summary":108,"summary_zh":9,"content":9,"source_name":109,"source_url":9,"published_at":110,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":111,"score_detail":112,"sources":115,"tags":117,"search_phrases":119,"slug":122,"view_count":123,"doi":9,"paper":124,"created_at":133},1371,"Diag-STFN：用于全球收获前作物产量预测的诊断时空多模态融合网络","https:\u002F\u002Flinkinghub.elsevier.com\u002Fretrieve\u002Fpii\u002FS1574954126002669","针对全球 38 国玉米和 29 国小麦在早、中、晚三个收获前时段的产量预测，提出 Diag-STFN 诊断时空多模态融合网络；根据数据集特征自动选择时空趋势耦合与空间模块激活；在 CY-Bench 基准下，玉米和小麦所有时段均取得最低池化 NRMSE，在 MAPE 和 KGE 上亦保持领先；消融研究表明诊断模块选择是性能提升的首要因素，融合策略是次要因素；方差分解显示国别间差异大于模型间差异。","Ecological Informatics · Elsevier · 2026-08-31","2026-08-30T16:00:00Z",75,{"impact":17,"substance":18,"depth":17,"authority":19,"freshness":113,"relevant":21,"comment":114},3,"全球尺度的作物产量预测研究，方法新颖且数据规模大，对农业信息化有参考价值。",[116],{"name":109,"url":107},[26,27,118,28,29],"产量预测",[120,121],"农业人工智能 产量预测 智慧农业 深度学习","农业人工智能 产量预测","农业人工智能产量预测智慧农业深度学习-1371",2,{"doi":9,"openalex_id":9,"authors":125,"venue":9,"cited_by_count":35,"oa_url":9,"card":126,"direction":130,"ingested_from":132},[],{"tldr":127,"method":128,"finding":129,"direction":130,"opportunity":131},"提出Diag-STFN网络，用于全球收获前玉米和小麦产量预测，在CY-Bench基准上取得最优性能。","诊断时空多模态融合网络，自动选择时空趋势耦合与空间模块激活，基于CY-Bench","Diag-STFN在玉米和小麦所有时段均取得最低NRMSE，诊断模块选择是性能提升首要因素。","农业人工智能与决策模型","国别间差异大于模型间差异，可探索区域自适应或跨域迁移的产量预测模型，以提升全球适用性。","agent","2026-09-02T00:05:07.231943Z",{"id":135,"title":136,"url":137,"summary":138,"summary_zh":9,"content":9,"source_name":139,"source_url":9,"published_at":140,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":141,"score_detail":142,"sources":144,"tags":146,"search_phrases":149,"slug":152,"view_count":35,"doi":9,"paper":153,"created_at":160},3324,"Diag-STFN：全球收获前作物产量预测的诊断时空多模态融合网络——覆盖38国玉米29国小麦（Ecological Informatics 2026）","https:\u002F\u002Fm2.mtmt.hu\u002Fapi\u002Fpublication\u002F37354974?&&labelLang=hun","《Ecological Informatics》2026年第96期：Zhuang等提出Diag-STFN——一种诊断时空多模态融合网络，用于全球收获前作物产量预测。该网络基于数据集特征选择模型结构，以确定是否需要时间趋势耦合和空间模块激活。在三种前置期（早、中、晚季）下，基于覆盖38国玉米和29国小麦的CY-Bench基准数据集进行评估。结果表明，所提方法在所有前置期均实现了两种作物的最低汇总NRMSE，并在MAPE和KGE等补充指标上保持领先。消融研究表明诊断模块选择提供了主要的性能提升；方差分解显示性能差异在国家之间大于模型之间。","《Ecological Informatics》96 (2026) 103860","2026-09-17T00:00:00Z",78,{"impact":17,"substance":18,"depth":17,"authority":19,"freshness":61,"relevant":21,"comment":143},"方法新颖、覆盖38国玉米与29国小麦的全球收获前产量预测研究，学术价值突出但产业落地尚早，适合作为前沿技术资讯收录。",[145],{"name":139,"url":137},[26,27,118,147,148,29],"小麦","玉米",[150,151],"Diag-STFN 作物产量预测","CY-Bench 玉米 小麦","Diag-STFN作物产量预测-3324",{"doi":9,"openalex_id":9,"authors":154,"venue":9,"cited_by_count":35,"oa_url":9,"card":155,"direction":130,"ingested_from":132},[],{"tldr":156,"method":157,"finding":158,"direction":130,"opportunity":159},"提出诊断式时空多模态融合网络Diag-STFN，实现全球收获前玉米小麦产量预测。","基于CY-Bench基准，按数据特征诊断选择时间趋势与空间模块，覆盖38国玉米2","各前置期均取得最低NRMSE，诊断模块选择贡献最大，国家间差异大于模型间差异。","可探索自适应诊断机制迁移至其他作物，并针对国家间差异开展区域化建模与不确定性量化。","2026-09-24T00:04:02.684732Z",{"id":162,"title":163,"url":164,"summary":165,"summary_zh":166,"content":9,"source_name":167,"source_url":164,"published_at":11,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":168,"score_detail":169,"sources":172,"tags":174,"search_phrases":177,"slug":180,"view_count":35,"doi":181,"paper":182,"created_at":200},3256,"Leakage-aware, calibrated, and explainable deep learning for robust almond disease classification","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.compag.2026.112464","Leakage-aware, calibrated, and explainable deep learning for robust almond disease classification。Computers and Electronics in Agriculture","面向稳健杏仁病害分类的泄漏感知、校准且可解释的深度学习。《农业计算机与电子》","Computers and Electronics in Agriculture",72,{"impact":74,"substance":75,"depth":170,"authority":19,"freshness":20,"relevant":21,"comment":171},17,"核心期刊论文，方法上有防泄漏、校准与可解释性创新，但作物小众、属细分技术进展，未达每日精选门槛。",[173],{"name":167,"url":164},[26,27,28,175,176],"病害识别","巴旦木",[178,179],"巴旦木 病害 深度学习","农业人工智能 智慧农业 深度学习 病害识别","巴旦木病害深度学习-3256","10.1016\u002Fj.compag.2026.112464",{"doi":181,"openalex_id":183,"authors":184,"venue":167,"cited_by_count":35,"oa_url":9,"card":195,"direction":130,"ingested_from":58},"W7213988471",[185,188,191,193],{"name":186,"orcid":187},"Abebaw Degu Workneh","https:\u002F\u002Forcid.org\u002F0000-0001-7694-1577",{"name":189,"orcid":190},"Badr Elkari","https:\u002F\u002Forcid.org\u002F0000-0002-0893-783X",{"name":192,"orcid":9},"Meryam El Mouhtadi",{"name":194,"orcid":9},"Mohammad Furqan Ali",{"tldr":196,"method":197,"finding":198,"direction":130,"opportunity":199},"提出防泄漏、校准且可解释的深度学习框架，用于稳健的杏仁病害分类。","采用防数据泄漏的深度学习训练、概率校准与可解释性分析。","该框架能提升杏仁病害分类的稳健性、可信度与可解释性。","可探索防泄漏与校准机制在其他作物病害识别中的泛化及田间部署。","2026-09-23T23:30:01.628054Z",{"id":202,"title":203,"url":204,"summary":205,"summary_zh":9,"content":9,"source_name":206,"source_url":9,"published_at":11,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":207,"score_detail":208,"sources":212,"tags":214,"search_phrases":217,"slug":220,"view_count":35,"doi":9,"paper":221,"created_at":228},3250,"农机化研究2026(11)：基于多源感知与语义分割的果树行识别与路径跟踪控制——郑州工业应用技术学院许洋洋等 DeepLabv3+ F1达0.94","http:\u002F\u002Fwww.qikanvip.com\u002Fqkml\u002F166268.html","农机化研究2026年第11期发表郑州工业应用技术学院许洋洋、柴亚珂、王莹、牛瑞利团队论文：针对果园中裸露土壤、遮蔽物干扰和冠层阴影等因素导致导航路径难以稳定提取的问题，提出融合多源感知与语义分割的自走式植保机果树识别与路径控制方法。首先利用无人机获取多光谱影像并生成正射影像（DOM）与数字表面模型（DSM），计算归一化差异绿度指数（NDGI），增强果树冠层与背景的可分性；然后采用U-Net\u002FDeepLabv3+语义分割网络在保留空间上下文信息的同时实现果树行\u002F背景端到端分割。田间对比试验结果表明，DeepLabv3+整体分割F1可达0.94，每样区生成导航线断裂1.2次，且转向工况下横向误差RMSE为0.11 m，显著优于传统阈值分割和SVM方法。","农机化研究",79,{"impact":209,"substance":18,"depth":17,"authority":210,"freshness":13,"relevant":21,"comment":211},16,13,"多源遥感与语义分割结合的果园导航研究，F1达0.94且横向误差0.11米，方法新颖、数据扎实，具备行业参考价值。",[213],{"name":206,"url":204},[26,27,215,29,216],"农机导航","果园植保",[218,219],"郑州工业应用技术学院 果树行识别","DeepLabv3 果园导航 路径跟踪","郑州工业应用技术学院果树行识别-3250",{"doi":9,"openalex_id":9,"authors":222,"venue":9,"cited_by_count":35,"oa_url":9,"card":223,"direction":56,"ingested_from":132},[],{"tldr":224,"method":225,"finding":226,"direction":56,"opportunity":227},"提出融合无人机多光谱与语义分割的果树行识别与路径跟踪方法，实现植保机自主导航。","无人机多光谱DOM\u002FDSM与NDGI，U-Net\u002FDeepLabv3+语义分割，","DeepLabv3+分割F1达0.94，导航线断裂1.2次，转向横向误差RMSE 0.11 m，优于","可探索多光谱与DSM特征融合的轻量化分割模型，并迁移至多树种、多季节果园导航。","2026-09-23T00:04:33.496233Z",{"id":230,"title":231,"url":232,"summary":233,"summary_zh":9,"content":9,"source_name":234,"source_url":9,"published_at":11,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":235,"score_detail":236,"sources":238,"tags":240,"search_phrases":243,"slug":246,"view_count":35,"doi":9,"paper":247,"created_at":254},3246,"Geospatial Artificial Intelligence in Precision Agriculture: A Systematic Review（精准农业中的地理空间人工智能：系统综述）","https:\u002F\u002Fwww.mdpi.com\u002F3043-1204\u002F1\u002F1\u002F4","MDPI AI Precis. Agric. 1(1) 4 发表系统综述：综述2019-2026年Scopus、Web of Science及补充检索文献，考察精准农业中的应用、多模态数据集成与决策支持。综述表明研究范式从孤立的制图与预测转向集成工作流，整合卫星与无人机影像、田间传感器、气象、土壤与管理记录。报告效益包括产量预测改善、压力与病害更早检测、土壤制图更精细、灌溉与投入品更精准。然而，产量、环境和经济效益仍依赖当地条件和运行验证。核心发现是：地理参考数据本身不能确保真正的地理空间AI；可靠工作流必须解决空间依赖性、尺度不匹配、可迁移性和不确定性。可解释性和融入农场运营对可执行推荐同样重要。采用仍受数据互操作性、连接性、可负担性、隐私和技术能力限制。","MDPI AI in Precision Agriculture",82,{"impact":17,"substance":18,"depth":17,"authority":19,"freshness":13,"relevant":21,"comment":237},"系统综述梳理2019-2026年GeoAI在精准农业的集成工作流与落地瓶颈，结论扎实、时效性强，对智慧农业技术路线有参考价值。",[239],{"name":234,"url":232},[26,27,29,241,242],"决策支持","多模态数据",[244,245],"GeoAI 精准农业 系统综述","地理空间人工智能 精准农业","GeoAI精准农业系统综述-3246",{"doi":9,"openalex_id":9,"authors":248,"venue":9,"cited_by_count":35,"oa_url":9,"card":249,"direction":56,"ingested_from":132},[],{"tldr":250,"method":251,"finding":252,"direction":56,"opportunity":253},"系统综述2019-2026年地理空间AI在精准农业的应用、多模态数据集成与决策支持。","系统综述Scopus、Web of Science等文献，分析多模态数据集成与决","地理参考数据不等于地理空间AI，可靠工作流须解决空间依赖、尺度、可迁移性与不确定性。","可研究跨尺度空间依赖建模与可迁移性评估，提升模型在不同农场条件下的泛化能力。","2026-09-23T00:04:33.172821Z"]