[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-2907":3,"related-2907":56},{"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":55},2907,"MMTA-ACDD: Multi-teacher online adaptive learning and multimodal augmentation synergy for robust crop disease detection","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.compag.2026.112429","The diagnosis and control of crop diseases are crucial for ensuring agricultural productivity. Single-stage real-time efficient detection models have been widely applied to crop disease detection tasks, but typically perform well only when the training and test sets have consistent distributions. However, in practical crop cultivation environments, real-time detection models not only need to perform excellently on source domain data but also must adapt to target domain data with different distributions. To address this challenge, this paper proposes a primary–auxiliary teacher collaborative enhanced supervision test-time adaptation framework to improve the robustness of real-time detection models in dynamic crop cultivation environments. Specifically, first, we propose a primary teacher supervision and auxiliary teacher guidance strategy, which combines weak augmentation and multimodal background strong augmentation techniques, respectively, to generate higher-quality pseudo-labels for the student model, thereby alleviating the problem of error accumulation in adaptive training. Second, we propose a maximum gradient dynamic adaptive recovery strategy(MGDA), which retains more effective knowledge from the source model by setting a maximum recovery trigger threshold and dynamic adaptive mechanism, preventing the student model from forgetting original knowledge when adapting to new domains. Finally, to enhance the robustness of the multi-teacher architecture under target distribution shift, we introduce a target class-aware contrastive learning strategy(OACL) to more effectively utilize pseudo-labels for feature learning in crop disease target detection tasks. Experimental results demonstrate that our proposed framework significantly improves the performance of real-time detection models in dynamic environments, providing an effective solution for crop disease detection.","作物病害的诊断与控制对保障农业生产力至关重要。单阶段实时高效检测模型已广泛应用于作物病害检测任务，但通常仅在训练集与测试集分布一致时表现良好。然而，在实际作物种植环境中，实时检测模型不仅需要在源域数据上表现优异，还必须适应具有不同分布的目标域数据。为解决这一挑战，本文提出了一种主辅教师协同增强监督的测试时自适应框架，以提升实时检测模型在动态作物种植环境中的鲁棒性。具体而言，首先，我们提出了主教师监督与辅助教师引导策略，分别结合弱增强和多模态背景强增强技术，为学生模型生成更高质量的伪标签，从而缓解自适应训练中的误差累积问题。其次，我们提出了最大梯度动态自适应恢复策略（MGDA），通过设置最大恢复触发阈值和动态自适应机制，保留源模型中更有效的知识，防止学生模型在适应新域时遗忘原有知识。最后，为增强多教师架构在目标分布偏移下的鲁棒性，我们引入了目标类别感知对比学习策略（OACL），以更有效地利用伪标签进行作物病害目标检测任务中的特征学习。实验结果表明，本文提出的框架显著提升了实时检测模型在动态环境中的性能，为作物病害检测提供了有效的解决方案。",null,"Computers and Electronics in Agriculture","2026-09-18T00:00:00Z","论文",10,false,81,{"impact":17,"substance":18,"depth":17,"authority":19,"freshness":20,"relevant":21,"comment":22},18,22,14,9,1,"提出主辅教师协同与多模态增强的测试时自适应框架，显著提升动态农田环境下实时病害检测鲁棒性，方法新颖且实验充分，值得精选。",[24],{"name":10,"url":6},[26,27,28,29,30],"智慧农业","农业人工智能","作物病害检测","域自适应","多模态增强",[32,33],"作物病害检测 域自适应 多教师","作物病害检测 农业人工智能 多模态增强 域自适应","作物病害检测域自适应多教师-2907",0,"10.1016\u002Fj.compag.2026.112429",{"doi":36,"openalex_id":38,"authors":39,"venue":10,"cited_by_count":35,"oa_url":6,"card":48,"direction":52,"ingested_from":54},"W7213546000",[40,42,44,46],{"name":41,"orcid":9},"Hao Sun",{"name":43,"orcid":9},"Shiyu Wang",{"name":45,"orcid":9},"Zhenqi Cheng",{"name":47,"orcid":9},"Rui Fu",{"tldr":49,"method":50,"finding":51,"direction":52,"opportunity":53},"提出多教师在线自适应与多模态增强框架，提升作物病害实时检测模型在动态环境下的鲁棒性。","主辅教师协同监督测试时自适应，弱增强与多模态背景强增强生成伪标签，MGDA与OA","框架显著提升实时检测模型在分布偏移下的性能，缓解伪标签误差累积与源知识遗忘。","农业人工智能与决策模型","可探索多模态增强与测试时自适应在田间边缘设备上的轻量化部署及跨作物泛化。","openalex","2026-09-19T23:30:01.851144Z",{"total":57,"page":21,"page_size":57,"items":58},6,[59,93,123,152,180,207],{"id":60,"title":61,"url":62,"summary":63,"summary_zh":9,"content":9,"source_name":64,"source_url":9,"published_at":65,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":66,"score_detail":67,"sources":73,"tags":75,"search_phrases":78,"slug":81,"view_count":35,"doi":82,"paper":83,"created_at":92},2399,"Lightweight architecture optimization of YOLOv12n for improved cotton verticillium wilt detection","https:\u002F\u002Fwww.frontiersin.org\u002Farticles\u002F10.3389\u002Ffpls.2026.1822081\u002Ffull","Frontiers in Plant Science 17:1822081（2026）。Ye Zhuang等基于YOLOv12n框架提出轻量精准检测模型YOLO-SCOD。引入StarNet架构作为骨干网；颈网络C3k模块集成通道聚合块；检测头用全维动态卷积替代深度卷积。精度和召回率分别提升至0.960和0.911，mAP50-95提升6.436%；参数量、FLOPs、模型大小分别减少13.728%、20.635%、12.727%，推理速度提升4.167%。","Frontiers in Plant Science | 2026-09","2026-09-08T00:00:00Z",76,{"impact":68,"substance":69,"depth":17,"authority":70,"freshness":71,"relevant":21,"comment":72},16,21,13,8,"基于YOLOv12n的轻量化检测模型在棉花黄萎病识别上兼顾精度与效率，方法新颖、数据扎实，对智慧植保具参考价值。",[74],{"name":64,"url":62},[26,27,76,77,28],"棉花黄萎病","轻量化模型",[79,80],"作物病害检测 农业人工智能 棉花黄萎病 轻量化模型","作物病害检测 农业人工智能","作物病害检测农业人工智能棉花黄萎病轻量化模型-2399","10.3389\u002Ffpls.2026.1822081\u002Ffull",{"doi":82,"openalex_id":9,"authors":84,"venue":9,"cited_by_count":35,"oa_url":9,"card":85,"direction":89,"ingested_from":91},[],{"tldr":86,"method":87,"finding":88,"direction":89,"opportunity":90},"提出轻量模型YOLO-SCOD，实现棉花黄萎病精准检测。","基于YOLOv12n，引入StarNet骨干、C3k模块与全维动态卷积。","精度召回率达0.960和0.911，mAP提升6.436%，模型更轻更快。","农业遥感与作物表型","可探索轻量模型在移动端或无人机实时病害检测中的部署与泛化。","agent","2026-09-14T00:06:33.263987Z",{"id":94,"title":95,"url":96,"summary":97,"summary_zh":9,"content":9,"source_name":98,"source_url":9,"published_at":99,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":100,"score_detail":101,"sources":105,"tags":107,"search_phrases":111,"slug":114,"view_count":35,"doi":9,"paper":115,"created_at":122},3002,"改进生物神经网络的农业播种机全覆盖路径规划","https:\u002F\u002Fwww.mdpi.com\u002F2077-0472\u002F16\u002F18\u002F1968","江苏大学魏军等提出一种考虑播种与非播种运动模式切换机制的改进生物神经网络（BNN）方法，基于周围环境条件将下一节点状态分类为播种、封闭或转移节点。在BNN景观引导下机器沿平行直线路径继续播种操作；检测到封闭节点时切换至非播种模式并使用深度优先搜索算法搜索潜在封闭区域；检测到转移节点时同样切换非播种模式搜索合理的新目标节点。仿真表明该方法实现播种操作的完全覆盖同时避免重复遍历已播种区域。","MDPI Agriculture 16(18):1968","2026-09-14T00:00:00Z",69,{"impact":102,"substance":69,"depth":103,"authority":70,"freshness":57,"relevant":21,"comment":104},12,17,"提出改进生物神经网络的全覆盖路径规划方法，方法新颖、结论可靠，但属细分领域学术进展，公共影响有限。",[106],{"name":98,"url":96},[26,27,108,109,110],"智能农机","路径规划","播种机",[112,113],"江苏大学 播种机 全覆盖路径规划","生物神经网络 播种机 路径规划","江苏大学播种机全覆盖路径规划-3002",{"doi":9,"openalex_id":9,"authors":116,"venue":9,"cited_by_count":35,"oa_url":9,"card":117,"direction":52,"ingested_from":91},[],{"tldr":118,"method":119,"finding":120,"direction":52,"opportunity":121},"提出改进生物神经网络，实现农业播种机全覆盖路径规划并避免重复播种。","改进BNN结合节点分类与深度优先搜索，仿真验证。","方法实现播种完全覆盖，同时避免重复遍历已播种区域。","可结合真实农田地形与多机协同，验证动态环境下的路径规划鲁棒性。","2026-09-20T00:03:08.288198Z",{"id":124,"title":125,"url":126,"summary":127,"summary_zh":9,"content":9,"source_name":128,"source_url":9,"published_at":129,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":130,"score_detail":131,"sources":134,"tags":136,"search_phrases":140,"slug":143,"view_count":35,"doi":9,"paper":144,"created_at":151},3000,"Plant-GeoAT：几何感知3D植物点云器官身份解析用于器官级表型分析","https:\u002F\u002Fwww.mdpi.com\u002F2073-4395\u002F16\u002F18\u002F1809","四川农业大学吴俊杰等提出Plant-GeoAT几何感知解析网络，在RGB融合前编码局部相对-XYZ邻域并将空间邻域与特征空间关系耦合用于密集点预测。在自建油菜数据集、基于图像的大豆数据集、激光扫描Pheno4D玉米和番茄数据集上评估，5个随机种子下mIoU分别达92.26±0.28%、82.50±0.24%、99.74±0.05%、94.75±0.15%，玉米Stem IoU达99.57±0.09%。","MDPI Agronomy 16(18):1809","2026-09-15T00:00:00Z",78,{"impact":17,"substance":132,"depth":17,"authority":70,"freshness":57,"relevant":21,"comment":133},23,"方法新颖、多作物多数据集验证且精度数据扎实，属器官级表型分析细分领域的重要技术进展，值得进入每日精选。",[135],{"name":128,"url":126},[26,27,137,138,139],"油菜","高通量表型","三维点云",[141,142],"四川农业大学 植物点云 器官识别","Plant-GeoAT 表型分析","四川农业大学植物点云器官识别-3000",{"doi":9,"openalex_id":9,"authors":145,"venue":9,"cited_by_count":35,"oa_url":9,"card":146,"direction":89,"ingested_from":91},[],{"tldr":147,"method":148,"finding":149,"direction":89,"opportunity":150},"提出几何感知网络Plant-GeoAT，实现3D植物点云器官身份解析与器官级表型分析。","编码局部相对XYZ邻域并耦合空间与特征空间关系，在油菜、大豆、玉米、番茄点云数据","五个数据集mIoU最高达99.74%，玉米茎IoU达99.57%，验证了几何感知对器官分割的有效性。","可探索跨物种、跨传感器的轻量化几何感知模型，并推动器官级表型与基因型关联分析。","2026-09-20T00:03:08.094512Z",{"id":153,"title":154,"url":155,"summary":156,"summary_zh":9,"content":9,"source_name":157,"source_url":9,"published_at":158,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":159,"score_detail":160,"sources":162,"tags":164,"search_phrases":168,"slug":171,"view_count":35,"doi":9,"paper":172,"created_at":179},2999,"基于改进DeepLabv3+的高标准农田田间道路提取与结构指标量化框架","https:\u002F\u002Fwww.mdpi.com\u002F2077-0472\u002F16\u002F18\u002F1986","沈阳农业大学刘永生等开发了基于MobileNetV2改进DeepLabv3+的高标准农田田间道路提取轻量化框架，集成Normalization-based Attention Module与Content-Aware ReAssembly of FEatures。三次随机种子训练平均mIoU 93.34%、mPA 96.75%、精度98.90%，模型参数6.14M、推理速度17.04 FPS；沥青、混凝土、砾石道路宽度预测R²分别为0.650、0.486、0.662，宽度MAE 0.130\u002F0.140\u002F0.100 m。第二验证区域连通性指数从0.4682提升至0.4795，支持高标准农田田间道路高效、可量化、可追溯的验收检查。","MDPI Agriculture 16(18):1986","2026-09-16T00:00:00Z",77,{"impact":68,"substance":18,"depth":17,"authority":70,"freshness":71,"relevant":21,"comment":161},"方法有创新、指标详实，对高标准农田道路验收有实用价值，但属细分技术论文，影响面有限。",[163],{"name":157,"url":155},[26,27,165,166,167],"高标准农田","遥感","田间道路",[169,170],"沈阳农业大学 高标准农田 道路提取","DeepLabv3 田间道路 遥感","沈阳农业大学高标准农田道路提取-2999",{"doi":9,"openalex_id":9,"authors":173,"venue":9,"cited_by_count":35,"oa_url":9,"card":174,"direction":89,"ingested_from":91},[],{"tldr":175,"method":176,"finding":177,"direction":89,"opportunity":178},"提出改进DeepLabv3+轻量框架，提取高标准农田田间道路并量化结构指标。","MobileNetV2+NAM+CARAFE改进DeepLabv3+，多区域遥感","mIoU 93.34%，道路宽度预测R²最高0.662，连通性指数提升至0.4795。","可拓展至多作物、多地形道路提取，并结合时序遥感实现道路损毁动态监测。","2026-09-20T00:03:08.023498Z",{"id":181,"title":182,"url":183,"summary":184,"summary_zh":9,"content":9,"source_name":185,"source_url":9,"published_at":11,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":186,"score_detail":187,"sources":189,"tags":191,"search_phrases":195,"slug":198,"view_count":35,"doi":9,"paper":199,"created_at":206},2998,"物候阶段渐进式多源数据融合的县域冬小麦产量预测","https:\u002F\u002Ffinance.sina.com.cn\u002Froll\u002F2026-09-18\u002Fdoc-inisheny8322115.shtml","西安财经大学王毅等联合江苏大学张立元副教授、西安理工大学西北旱区生态水利国家重点实验室、中国科学院重庆绿色智能技术研究院团队，以河南省100个冬小麦主产县为研究区融合2013—2022年遥感变量、气象变量和日光诱导叶绿素荧光光合变量，构建覆盖冬小麦分蘖期至成熟期的多源时序特征集。提出物候阶段渐进式多源数据融合方法明确不同物候阶段信息累积对县域冬小麦估产性能的影响。","智慧农业(中英文)2026,8(4):70-84",80,{"impact":17,"substance":18,"depth":17,"authority":19,"freshness":71,"relevant":21,"comment":188},"以河南100个主产县2013—2022年遥感、气象与SIF数据构建物候阶段渐进式融合估产方法，数据规模与方法新颖性突出，对县域粮食产量预测有参考价值。",[190],{"name":185,"url":183},[26,27,192,193,194],"遥感估产","冬小麦","多源数据融合",[196,197],"河南 冬小麦 遥感估产","物候阶段 多源数据融合 产量预测","河南冬小麦遥感估产-2998",{"doi":9,"openalex_id":9,"authors":200,"venue":9,"cited_by_count":35,"oa_url":9,"card":201,"direction":89,"ingested_from":91},[],{"tldr":202,"method":203,"finding":204,"direction":89,"opportunity":205},"融合遥感、气象与SIF数据，提出物候阶段渐进式融合方法预测河南县域冬小麦产量。","2013—2022年河南100县遥感、气象、SIF多源时序特征，按物候阶段渐进融","明确不同物候阶段信息累积对县域冬小麦估产性能的影响，提升预测精度。","可探索物候自适应加权与深度学习融合，并迁移至其他作物及极端气候情景下的县域估产。","2026-09-20T00:03:07.932461Z",{"id":208,"title":209,"url":210,"summary":211,"summary_zh":9,"content":9,"source_name":212,"source_url":9,"published_at":213,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":130,"score_detail":214,"sources":216,"tags":218,"search_phrases":222,"slug":225,"view_count":35,"doi":9,"paper":226,"created_at":233},2997,"基于无人机多光谱图像和VGG21模型的小麦渍害调控效果识别方法","https:\u002F\u002Fwww.toutiao.com\u002Farticle\u002F7685741381196825088","江苏省农业科学院农业信息研究所梁万杰等联合中国农科院农业环境与可持续发展研究所、湖北粮作所、扬州大学等团队，针对小麦渍害防控提出基于无人机多光谱图像和VGG21模型的快速无损识别方法。在小麦拔节-抽穗和抽穗-灌浆两个阶段开展对照、渍水胁迫、硅肥调控和氨基酸调控4个类别数据集，大疆精灵4多光谱无人机采集小麦冠层多光谱图像，测产评估调控效果。","智慧农业(中英文)2026,8(4):60-69","2026-09-15T12:42:00Z",{"impact":17,"substance":18,"depth":17,"authority":19,"freshness":57,"relevant":21,"comment":215},"多机构协作提出无人机多光谱结合VGG21的小麦渍害无损识别方法，方法新颖、数据扎实，对智慧农业植保监测有参考价值。",[217],{"name":212,"url":210},[26,27,219,220,221],"无人机遥感","小麦渍害","多光谱成像",[223,224],"江苏省农科院 小麦渍害 无人机多光谱","VGG21 小麦 渍害识别","江苏省农科院小麦渍害无人机多光谱-2997",{"doi":9,"openalex_id":9,"authors":227,"venue":9,"cited_by_count":35,"oa_url":9,"card":228,"direction":89,"ingested_from":91},[],{"tldr":229,"method":230,"finding":231,"direction":89,"opportunity":232},"用无人机多光谱图像和VGG21模型识别小麦渍害调控效果。","大疆精灵4多光谱无人机采集冠层图像，构建VGG21分类模型。","该方法可快速无损识别渍害及硅肥、氨基酸调控效果。","可探索多光谱与深度学习结合评估其他逆境调控措施，并迁移至多作物场景。","2026-09-20T00:03:07.858899Z"]