[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-4701":3,"related-4701":75},{"id":4,"title":5,"url":6,"summary":7,"summary_zh":8,"content":9,"source_name":10,"source_url":8,"published_at":11,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":15,"score_detail":16,"sources":24,"tags":26,"primary_topic":32,"secondary_topics":38,"classification":39,"search_phrases":69,"slug":72,"view_count":73,"doi":8,"paper":8,"created_at":74},4701,"张柱会见阿塞拜疆农业部部长马马多夫","https:\u002F\u002Fwww.agri.cn\u002Fzx\u002Fnyyw\u002F202610\u002Ft20261010_8875675.htm","农业农村部部长张柱在京会见阿塞拜疆农业部部长马马多夫。双方就加强两国农业政策对话、拓展农业经贸合作、促进农业科技协作深入交流,会后签署谅解备忘录,深化智慧农业、葡萄酒、乡村振兴等领域合作",null,"本网讯　10月10日下午，农业农村部部长张柱在京会见阿塞拜疆农业部部长马马多夫，就加强两国农业农村领域合作进行深入交流。\n\n张柱表示，在两国元首战略引领下，中阿各领域合作发展迅速，农业务实合作取得明显进展。中阿农业互补性强，中方愿与阿方共同努力，贯彻落实好两国元首合作共识，用好农业工作组机制，加强农业政策对话、拓展农业经贸合作、促进农业科技协作，推动两国农业农村领域合作迈上新台阶，为中阿全面战略伙伴关系发展贡献力量。\n\n马马多夫表示，阿方高度重视对华农业合作，希望学习借鉴中国农业农村发展经验，吸引中国农业企业赴阿投资，扩大对华农产品贸易，在农业政策、智慧农业、葡萄酒、乡村振兴等领域深化交流。\n\n会后，双方共同签署了《中华人民共和国农业农村部与阿塞拜疆共和国农业部谅解备忘录》。","中国农业农村信息网 2026-10-10","2026-10-10T08:00:00Z","报道",10,false,75,{"impact":17,"substance":18,"depth":19,"authority":20,"freshness":21,"relevant":22,"comment":23},22,18,12,15,8,1,"中阿农业部长会见并签署谅解备忘录，属部级外事合作实质进展，但以通稿表述为主，专业深度有限。",[25],{"name":10,"url":6},[27,28,29,30,31],"智慧农业","乡村振兴","外事合作","农业经贸","谅解备忘录",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},"B2","fruit","果树","\u002Ftopic\u002Fhorticulture\u002Ffruit","B",[],{"status":40,"locked":14,"version":41,"method":42,"evidence":43},"classified","2026-10-22.2","rules",{"candidates":44,"agricultural_evidence":48},[45,52,57,63],{"code":46,"score":47,"supported":48,"title":49,"abstract":50,"tags":51},"G",4,true,[],[28],[28],{"code":53,"score":47,"supported":48,"title":54,"abstract":55,"tags":56},"I1",[],[27],[27],{"code":33,"score":58,"supported":48,"title":59,"abstract":60,"tags":62},3,[],[61],"葡萄",[],{"code":64,"score":58,"supported":48,"title":65,"abstract":66,"tags":68},"J1",[],[67],"农业政策",[],[70,71],"张柱 阿塞拜疆 农业合作","阿塞拜疆 谅解备忘录","张柱阿塞拜疆农业合作-4701",2,"2026-10-12T00:10:54.600939Z",{"total":76,"page":22,"page_size":76,"items":77},6,[78,122,202,245,293,343],{"id":79,"title":80,"url":81,"summary":82,"summary_zh":8,"content":83,"source_name":84,"source_url":8,"published_at":85,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":86,"score_detail":87,"sources":92,"tags":94,"primary_topic":98,"secondary_topics":99,"classification":100,"search_phrases":116,"slug":119,"view_count":120,"doi":8,"paper":8,"created_at":121},5516,"上海农科院葡萄团队赴银川参加第47届世界葡萄与葡萄酒大会","https:\u002F\u002Fwww.saas.sh.cn\u002Fhzjl\u002Fgjhz\u002Fcontent_57128","10月12日至16日第47届世界葡萄与葡萄酒大会在银川举办，智慧农业研究中心葡萄团队负责人奚晓军研究员作专题报告，阐明疏穗与环剥处理对果实糖、花青素及脱落酸积累的源库调控机制。团队还观摩酿酒葡萄AI综合作业机器人成果展示，与研发单位围绕葡萄智慧装备与果园智能化管理深入交流。","2026年10月12-16日，第47届世界葡萄与葡萄酒大会在银川举办，智慧农业研究中心葡萄团队负责人奚晓军研究员一行赴会交流。团队围绕“提升葡萄种植生产韧性与可持续发展能力”主题，向全球同行展示最新科研成果。\n\n会上，奚晓军作“Changes in sugar, anthocyanin and ABA accumulation in response to cluster thinning and girdling in ‘Jumeigui’grape”专题报告。报告围绕葡萄疏穗与环剥处理开展研究，阐明源库比对果实糖、花青素及脱落酸积累的调控机制，为葡萄优质栽培技术提供理论支撑，报告内容引发国内外园艺专家广泛研讨。参会期间，团队前往银川国际会展中心观摩酿酒葡萄 AI 综合作业机器人成果展示，详细了解该装备在葡萄园智能作业方面的技术特点与应用前景，并与研发单位围绕葡萄智慧装备研发、果园智能化管理等方向开展深入交流。\n\n本次参会是葡萄团队持续推进科研国际化、对标国际前沿开展研究的重要实践。","上海市农业科学院 2026-10-19","2026-10-19T00:00:00Z",65,{"impact":88,"substance":89,"depth":90,"authority":90,"freshness":76,"relevant":22,"comment":91},16,17,13,"国际学术会议交流并涉及葡萄AI作业机器人观摩，有实质科研内容与行业参考价值，但属团队参会动态，影响面有限。",[93],{"name":84,"url":81},[27,95,61,96,97],"智能农机","栽培技术","国际交流",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[],{"status":40,"locked":14,"version":41,"method":42,"evidence":101},{"candidates":102,"agricultural_evidence":48},[103,107,111],{"code":33,"score":19,"supported":48,"title":104,"abstract":105,"tags":106},[61],[61],[61],{"code":53,"score":47,"supported":48,"title":108,"abstract":109,"tags":110},[],[27],[27],{"code":112,"score":22,"supported":14,"title":113,"abstract":114,"tags":115},"F5",[],[],[95],[117,118],"上海农科院 葡萄 银川","世界葡萄与葡萄酒大会 奚晓军","上海农科院葡萄银川-5516",0,"2026-10-22T00:07:04.061911Z",{"id":123,"title":124,"url":125,"summary":126,"summary_zh":127,"content":8,"source_name":128,"source_url":125,"published_at":129,"category":130,"cover_url":8,"hotness":13,"is_selected":14,"score":131,"score_detail":132,"sources":135,"tags":137,"primary_topic":142,"secondary_topics":143,"classification":144,"search_phrases":173,"slug":176,"view_count":120,"doi":177,"paper":178,"created_at":201},5401,"A HIERARCHICAL DISTRIBUTED EDGE AI FRAMEWORK FOR REAL-TIME APPLE LEAF DISEASE CLASSIFICATION","https:\u002F\u002Fdoi.org\u002F10.22190\u002Ffuacr260506002g","The application of artificial intelligence (AI) in agriculture has transformed crop management, particularly in the detection of plant diseases. Apple trees, a key agricultural commodity, are vulnerable to various diseases, which can severely impact crop yields. Traditional disease detection methods, often reliant on manual inspections, are time-consuming and prone to error. In response, this study proposes a novel two-stage edge AI framework for apple leaf disease classification that integrates RGB imagery with distributed edge computing. The first stage performs binary classification to distinguish between healthy and diseased leaves, achieving an F1 score of 95.50%. This optimization is achieved by utilizing lightweight models on edge devices, thereby enhancing computational efficiency. The second stage employs a consensus-based multiclass classification approach, achieving a 96.44% F1 score in identifying specific diseases. This dual-stage system, supported by an edge AI architecture, ensures real-time disease detection while minimizing resource consumption and enhancing scalability in large-scale agricultural environments. The full system achieved a 95.50% accuracy and F1 score on the special dataset. The results demonstrate the system’s efficacy in controlled laboratory conditions, offering a promising solution for precise, real-time agricultural monitoring. This study underscores the potential of distributed edge AI systems for scalable and efficient disease detection in precision agriculture.","人工智能（AI）在农业中的应用已深刻改变了作物管理方式，尤其是在植物病害检测方面。苹果树作为重要的农业经济作物，易受多种病害侵袭，严重影响作物产量。传统的病害检测方法通常依赖人工检查，耗时且易出错。为此，本研究提出了一种新颖的两阶段边缘AI框架，用于苹果叶片病害分类，该框架将RGB图像与分布式边缘计算相结合。第一阶段执行二分类以区分健康叶片与患病叶片，F1分数达到95.50%。该优化通过在边缘设备上采用轻量级模型实现，从而提升了计算效率。第二阶段采用基于共识的多类分类方法，在识别具体病害方面F1分数达到96.44%。该双阶段系统在边缘AI架构的支持下，实现了实时病害检测，同时最大限度地降低了资源消耗，并增强了在大规模农业环境中的可扩展性。完整系统在专用数据集上达到了95.50%的准确率和F1分数。结果表明，该系统在受控实验室条件下具有有效性，为精准、实时的农业监测提供了一种有前景的解决方案。本研究凸显了分布式边缘AI系统在精准农业中实现可扩展且高效病害检测的潜力。","Facta Universitatis Series Automatic Control and Robotics","2026-10-18T00:00:00Z","论文",76,{"impact":18,"substance":133,"depth":89,"authority":19,"freshness":21,"relevant":22,"comment":134},21,"该研究提出两级边缘AI框架，在苹果叶片病害识别上取得95%以上F1值，方法新颖、数据可靠，对精准农业实时监测有参考价值，但尚处实验室验证阶段，产业影响有限。",[136],{"name":128,"url":125},[27,138,139,140,141],"农业人工智能","边缘计算","作物病害识别","苹果病害",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[],{"status":40,"locked":14,"version":41,"method":42,"evidence":145},{"candidates":146,"agricultural_evidence":48},[147,153,160,166],{"code":33,"score":18,"supported":48,"title":148,"abstract":150,"tags":152},[149],"apple",[35,151,149],"苹果",[141],{"code":53,"score":154,"supported":48,"title":155,"abstract":156,"tags":159},7,[],[157,158],"精准农业","precision agriculture",[27],{"code":161,"score":154,"supported":48,"title":162,"abstract":163,"tags":165},"I5",[],[139,164],"edge computing",[139],{"code":167,"score":168,"supported":48,"title":169,"abstract":170,"tags":172},"I2",5,[],[171],"artificial intelligence",[140,138],[174,175],"苹果叶片病害 边缘AI","分布式边缘计算 病害分类","苹果叶片病害边缘AI-5401","10.22190\u002Ffuacr260506002g",{"doi":177,"openalex_id":179,"authors":180,"venue":128,"cited_by_count":120,"oa_url":125,"card":193,"direction":199,"ingested_from":200},"W7222964752",[181,184,187,190],{"name":182,"orcid":183},"Milan Grujev","https:\u002F\u002Forcid.org\u002F0009-0002-3211-6844",{"name":185,"orcid":186},"Dragan H. Stojanovic","https:\u002F\u002Forcid.org\u002F0000-0003-4893-3814",{"name":188,"orcid":189},"Miloš Ilić","https:\u002F\u002Forcid.org\u002F0000-0002-9696-2874",{"name":191,"orcid":192},"Aleksandar Milosavljević","https:\u002F\u002Forcid.org\u002F0000-0003-3057-7727",{"tldr":194,"method":195,"finding":196,"direction":197,"opportunity":198},"提出两级边缘AI框架，用RGB图像实时分类苹果叶片病害。","分布式边缘计算+轻量模型，先二分类再共识多分类。","两级系统准确率与F1均达95.50%，多分类F1为96.44%。","智慧农业 \u002F 农业物联网","仅在受控实验室验证，可延伸至田间复杂光照与多作物泛化及边缘能耗优化。","农业人工智能与决策模型","openalex","2026-10-20T23:30:33.830748Z",{"id":203,"title":204,"url":205,"summary":206,"summary_zh":8,"content":8,"source_name":207,"source_url":8,"published_at":208,"category":130,"cover_url":8,"hotness":13,"is_selected":14,"score":209,"score_detail":210,"sources":212,"tags":214,"primary_topic":219,"secondary_topics":220,"classification":221,"search_phrases":232,"slug":235,"view_count":120,"doi":8,"paper":236,"created_at":244},5297,"高志永团队《基于特征优选与优化算法的干热河谷区柑橘蒸腾模拟》","https:\u002F\u002Fwww.sohu.com\u002Fa\u002F1087305323_122014422","《智慧农业(中英文)》2026年第8卷第4期刊载云南农业大学吴明清、高志永等团队论文,以10年生柑橘树为研究对象收集2020—2022年树干液流与环境因子数据;针对干热河谷加快建设智慧灌区背景下传统机理模型参数需求高、对部分参数高度敏感且误差来源复杂等难题,提出基于特征优选与优化算法的蒸腾模拟方法,可为优化灌区水分调控与配水策略提供决策支撑。","智慧农业(中英文)(2026,8(4))","2026-10-13T00:00:00Z",72,{"impact":88,"substance":133,"depth":89,"authority":90,"freshness":168,"relevant":22,"comment":211},"核心期刊论文，方法有创新且数据规模扎实，对干热河谷智慧灌区水分调控有决策参考价值，但属细分领域研究，时效性一般。",[213],{"name":207,"url":205},[27,215,216,217,218],"云南","柑橘","智慧灌区","蒸腾模拟",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[],{"status":40,"locked":14,"version":41,"method":42,"evidence":222},{"candidates":223,"agricultural_evidence":48},[224,228],{"code":33,"score":19,"supported":48,"title":225,"abstract":226,"tags":227},[216],[216],[216],{"code":53,"score":47,"supported":48,"title":229,"abstract":230,"tags":231},[],[27],[27],[233,234],"云南农业大学 柑橘 蒸腾模拟","干热河谷 智慧灌区","云南农业大学柑橘蒸腾模拟-5297",{"doi":8,"openalex_id":8,"authors":237,"venue":8,"cited_by_count":120,"oa_url":8,"card":238,"direction":197,"ingested_from":243},[],{"tldr":239,"method":240,"finding":241,"direction":197,"opportunity":242},"用特征优选与优化算法模拟干热河谷区柑橘蒸腾。","基于2020—2022年树干液流与环境因子数据构建特征优选与优化算法模型。","该方法可降低机理模型参数依赖与敏感性，提升蒸腾模拟精度。","可探索将蒸腾模拟嵌入智慧灌区实时配水决策，研究多树种迁移与边缘部署。","agent","2026-10-19T00:07:48.135945Z",{"id":246,"title":247,"url":248,"summary":249,"summary_zh":8,"content":8,"source_name":250,"source_url":8,"published_at":251,"category":130,"cover_url":8,"hotness":13,"is_selected":14,"score":15,"score_detail":252,"sources":255,"tags":257,"primary_topic":260,"secondary_topics":261,"classification":262,"search_phrases":281,"slug":284,"view_count":120,"doi":8,"paper":285,"created_at":292},5130,"DefoliationCAM:面向葡萄藤食叶严重度实时评估的边缘AI系统 [预印本]","https:\u002F\u002Farxiv.org\u002Fabs\u002F2610.20879","Fuchen Li等提出DefoliationCAM(D-CAM),一种太阳能供电的边缘AI系统,用于估算葡萄藤树冠食叶程度并通过LoRa将结果传输至Web仪表盘。流水线使用YOLOX-Nano在4K图像中定位葡萄藤主干,并使用基于MobileNet的轻量级通道注意力分类器对320×320像素的斑块进行分类。29,674个田间采集斑块被标注为背景或四种损伤等级之一;合并中度和重度类别将类别不平衡比从约40:1降至8.8:1。合成增强使macro-F1从0.768提升至0.838。分类器以更少的参数量(1.08M vs 23.52M)在macro-F1上优于ResNet-50。利用TensorRT FP16在NVIDIA Jetson Orin Nano上每张4K图像处理时间约1.5秒,完整运行周期57-60秒。两台设备为期五周的田间部署验证了自主太阳能运行、LoRa通信和远程食叶监测的可行性。","arXiv 2610.20879 2026 年 10 月 14 日","2026-10-14T00:00:00Z",{"impact":88,"substance":17,"depth":18,"authority":253,"freshness":21,"relevant":22,"comment":254},11,"预印本提出太阳能边缘AI系统，实现葡萄藤食叶严重度实时评估与LoRa远程监测，方法新颖、数据规模大、田间验证充分，具智慧农业应用价值。",[256],{"name":250,"url":248},[27,138,139,258,259],"葡萄藤","病害评估",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[],{"status":40,"locked":14,"version":41,"method":42,"evidence":263},{"candidates":264,"agricultural_evidence":48},[265,269,273,277],{"code":33,"score":253,"supported":48,"title":266,"abstract":267,"tags":268},[61],[61],[],{"code":53,"score":22,"supported":14,"title":270,"abstract":271,"tags":272},[],[],[27],{"code":167,"score":22,"supported":14,"title":274,"abstract":275,"tags":276},[],[],[138],{"code":161,"score":22,"supported":14,"title":278,"abstract":279,"tags":280},[],[],[139],[282,283],"DefoliationCAM 葡萄藤 食叶","Jetson Orin Nano 边缘AI 葡萄","DefoliationCAM葡萄藤食叶-5130",{"doi":8,"openalex_id":8,"authors":286,"venue":8,"cited_by_count":120,"oa_url":8,"card":287,"direction":197,"ingested_from":243},[],{"tldr":288,"method":289,"finding":290,"direction":197,"opportunity":291},"提出太阳能边缘AI系统DefoliationCAM，实时评估葡萄藤食叶严重度并远程传输。","YOLOX-Nano定位主干，MobileNet通道注意力分类，TensorRT","轻量分类器macro-F1达0.838，优于ResNet-50，田间部署验证可行。","可探索多作物迁移、损伤等级细化及边缘模型自适应更新以提升泛化性。","2026-10-17T00:13:44.932340Z",{"id":294,"title":295,"url":296,"summary":297,"summary_zh":8,"content":298,"source_name":299,"source_url":8,"published_at":300,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":301,"score_detail":302,"sources":306,"tags":308,"primary_topic":313,"secondary_topics":314,"classification":315,"search_phrases":338,"slug":341,"view_count":22,"doi":8,"paper":8,"created_at":342},5026,"深耕'土特产'挑稳'金扁担'——习近平总书记指引特色产业发展纪实","https:\u002F\u002Fwww.agri.cn\u002Fzx\u002Fnyyw\u002F202610\u002Ft20261015_8876548.htm","10月12日至16日,第47届世界葡萄与葡萄酒大会在宁夏银川举办,创办百余年来首次落地亚洲,宁夏葡萄酒产业综合效益超450亿元。文章系统梳理习近平总书记关于特色产业发展的重要论述:2020年6月宁夏考察为葡萄酒产业指明路径,2022年10月陕西延安考察指出陕北'非常适宜发展苹果种植',2024年6月宁夏考察关切闽宁镇近况、勉励'一定不要有浮躁心理,脚踏实地去积累,酒好不怕巷子深'。'十五五'规划纲要将'发展县域富民产业'单列一节,国务院相关文件将'培育壮大乡村特色产业'置于拓宽农民增收致富渠道首位,推动乡村产业蓬勃发展。","10月12日至16日，第47届世界葡萄与葡萄酒大会在宁夏银川举办。创办百余年来，这一业界顶级盛会首次落地亚洲。\n\n“随着人民生活水平不断提高，葡萄酒产业大有前景。”2020年6月，习近平总书记在宁夏考察时，为当地葡萄酒产业指明“提高技术水平，增加文化内涵，加强宣传推介，打造自己的知名品牌，提高附加值和综合效益”的发展路径。\n\n如今，宁夏葡萄酒产业综合效益超450亿元，成为走向世界的一张名片。\n\n发展特色产业是实现乡村振兴的重要途径，关乎农民增收致富、农业强国建设。\n\n2024年6月，习近平总书记在宁夏考察时关切询问闽宁镇近况。这里是总书记在福建工作时亲自提议福建和宁夏共建的生态移民点。\n\n“听说现在建设得很好，也有了特色产业。过去这么艰苦的一个地方，变成了出葡萄美酒的地方，令人感慨和欣慰啊！”习近平总书记动情地说。\n\n从脱贫攻坚到乡村振兴，习近平总书记一次次深入田间地头，为地方特色产业“把脉开方”。这背后饱含着对农民群众的殷殷牵挂，蕴含着对推进农业农村现代化的深邃思考。","新华社 中国农业农村信息网 2026 年 10 月 15 日","2026-10-15T00:00:00Z",81,{"impact":303,"substance":88,"depth":20,"authority":20,"freshness":304,"relevant":22,"comment":305},26,9,"央媒综述，紧扣特色产业与乡村振兴主线，宁夏葡萄酒450亿数据与大会首次落地亚洲具新闻价值，但属宣传性纪实报道，信息增量有限。",[307],{"name":299,"url":296},[28,309,310,311,312],"特色产业","土特产","宁夏葡萄酒","县域富民产业",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[],{"status":40,"locked":14,"version":41,"method":42,"evidence":316},{"candidates":317,"agricultural_evidence":48},[318,322,328,334],{"code":33,"score":76,"supported":48,"title":319,"abstract":320,"tags":321},[],[151,61],[],{"code":323,"score":168,"supported":48,"title":324,"abstract":325,"tags":327},"G4",[],[326],"乡村产业",[310,309],{"code":329,"score":58,"supported":48,"title":330,"abstract":331,"tags":333},"G6",[],[332],"农民增收",[],{"code":46,"score":22,"supported":14,"title":335,"abstract":336,"tags":337},[],[],[28],[339,340],"宁夏 葡萄酒 世界葡萄与葡萄酒大会","宁夏葡萄酒 综合效益 450亿","宁夏葡萄酒世界葡萄与葡萄酒大会-5026","2026-10-16T00:13:02.635607Z",{"id":344,"title":345,"url":346,"summary":347,"summary_zh":348,"content":8,"source_name":349,"source_url":346,"published_at":251,"category":130,"cover_url":8,"hotness":13,"is_selected":14,"score":301,"score_detail":350,"sources":353,"tags":355,"primary_topic":359,"secondary_topics":364,"classification":371,"search_phrases":423,"slug":426,"view_count":22,"doi":427,"paper":428,"created_at":449},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",{"impact":18,"substance":17,"depth":18,"authority":351,"freshness":304,"relevant":22,"comment":352},14,"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。",[354],{"name":349,"url":346},[27,138,356,357,358],"产量预测","多模态融合","遥感",{"code":167,"slug":360,"name":361,"path":362,"parent_code":363},"agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai","I",[365,366],{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},{"code":367,"slug":368,"name":369,"path":370,"parent_code":363},"I3","remote-sensing","遥感与农情监测","\u002Ftopic\u002Fdigital-agriculture\u002Fremote-sensing",{"status":40,"locked":14,"version":41,"method":42,"evidence":372},{"candidates":373,"agricultural_evidence":48},[374,382,388,394,398,405,411,417],{"code":167,"score":253,"supported":48,"title":375,"abstract":376,"tags":381},[],[377,378,379,380],"深度学习","机器学习","machine learning","deep learning",[138,357],{"code":33,"score":304,"supported":48,"title":383,"abstract":384,"tags":387},[],[151,385,386,149],"芒果","orchard",[],{"code":367,"score":21,"supported":48,"title":389,"abstract":390,"tags":393},[],[391,392],"无人机多光谱","multispectral",[356,358],{"code":53,"score":154,"supported":48,"title":395,"abstract":396,"tags":397},[],[157,158],[27],{"code":399,"score":76,"supported":48,"title":400,"abstract":401,"tags":404},"H2",[],[402,403],"土壤","soil",[],{"code":161,"score":76,"supported":48,"title":406,"abstract":407,"tags":410},[],[408,409],"物联网","IoT",[],{"code":412,"score":58,"supported":48,"title":413,"abstract":414,"tags":416},"H6",[],[415],"农业生态",[],{"code":418,"score":58,"supported":48,"title":419,"abstract":420,"tags":422},"H7",[],[421],"气候适应",[],[424,425],"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":427,"openalex_id":429,"authors":430,"venue":349,"cited_by_count":120,"oa_url":346,"card":444,"direction":197,"ingested_from":200},"W7221795850",[431,434,436,438,440,442],{"name":432,"orcid":433},"Shabir Ahmad Mir","https:\u002F\u002Forcid.org\u002F0000-0003-2311-218X",{"name":435,"orcid":8},"Umarfarooque Momin",{"name":437,"orcid":8},"Sudeep Kumar E",{"name":439,"orcid":8},"Itfaq Ahmad Mir",{"name":441,"orcid":8},"Ab Rouf Khan",{"name":443,"orcid":8},"Ghulam Mohammad Mir",{"tldr":445,"method":446,"finding":447,"direction":199,"opportunity":448},"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"]