[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-2903":3,"related-2903":45},{"id":4,"title":5,"url":6,"summary":7,"summary_zh":8,"content":8,"source_name":9,"source_url":8,"published_at":10,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":14,"score_detail":15,"sources":21,"tags":23,"search_phrases":29,"slug":32,"view_count":33,"doi":34,"paper":35,"created_at":44},2903,"基于QYmax叶绿素荧光成像和改进LCRNet模型的小麦白粉病智能识别与诊断,登Frontiers in Plant Science","https:\u002F\u002Fwww.frontiersin.org\u002Fjournals\u002Fplant-science\u002Farticles\u002F10.3389\u002Ffpls.2026.1837843\u002Ffull","本研究整合叶绿素荧光成像与深度学习算法,刻画小麦白粉病发展过程中QYmax等关键荧光参数的时序动态;构建基于QYmax的小麦白粉病叶绿素荧光图像数据集;提出LCRNet智能识别模型,通过LSAF大型选择性自适应融合和CA坐标注意力模块的协同处理机制,实现高效病害识别。在独立测试集上模型准确率达98.0%、F1-score达98.3%,显著优于对比模型,为作物病害预防控制系统的智能化与精准化绿色转型提供高精度方案。",null,"Frontiers in Plant Science","2026-09-18T00:00:00Z","论文",10,false,82,{"impact":16,"substance":17,"depth":16,"authority":18,"freshness":12,"relevant":19,"comment":20},18,22,14,1,"方法新颖、数据扎实且时效性强，但属细分领域论文，产业影响有限，适合进入每日精选。",[22],{"name":9,"url":6},[24,25,26,27,28],"智慧农业","农业人工智能","作物病害识别","小麦白粉病","叶绿素荧光成像",[30,31],"小麦白粉病 叶绿素荧光成像 LCRNet","QYmax 小麦白粉病 智能识别","小麦白粉病叶绿素荧光成像LCRNet-2903",0,"10.3389\u002Ffpls.2026.1837843\u002Ffull",{"doi":34,"openalex_id":8,"authors":36,"venue":8,"cited_by_count":33,"oa_url":8,"card":37,"direction":41,"ingested_from":43},[],{"tldr":38,"method":39,"finding":40,"direction":41,"opportunity":42},"用QYmax叶绿素荧光成像与改进LCRNet实现小麦白粉病高精度识别。","构建QYmax荧光图像数据集，提出含LSAF与CA模块的LCRNet模型。","独立测试集准确率98.0%、F1-score 98.3%，显著优于对比模型。","农业遥感与作物表型","可探索多病害、多生育期与田间自然光下的荧光成像泛化及轻量化部署。","agent","2026-09-19T00:06:08.843770Z",{"total":46,"page":19,"page_size":46,"items":47},6,[48,91,119,147,174,201],{"id":49,"title":50,"url":51,"summary":52,"summary_zh":53,"content":8,"source_name":54,"source_url":51,"published_at":55,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":56,"score_detail":57,"sources":62,"tags":64,"search_phrases":67,"slug":70,"view_count":33,"doi":71,"paper":72,"created_at":90},2541,"Evidence integrity and review utility in crop-image decision support: an audit-to-review framework for precision agriculture","https:\u002F\u002Fdoi.org\u002F10.1186\u002Fs42269-026-01492-x","Abstract Background Agricultural artificial intelligence benchmarks can overstate decision reliability when multiple files represent the same biological evidence or when split provenance is unclear. We evaluated a classifier-agnostic audit-to-review framework that fixes evidence identity and role provenance before model comparison, declares aggregation estimands, and links predictive uncertainty to review workload. A public wheat-leaf corpus of 7,595 files was hashed to 5,118 unique contents; eligibility criteria defined a 915-content five-class task. Archived file-level benchmarks were separated from three frozen near-duplicate-grouped P2 partition configurations, and semantic, texture and fusion representations were evaluated under partition-weighted and one-content-one-weight estimands. Split conformal prediction was assessed by empirical coverage, review rate, error capture, review yield and dimensionless review utility. Results Macro-F1 was 0.962 under the legacy P0 benchmark, 0.805 ± 0.030 for semantic-only P2 and 0.819 ± 0.029 for direct fusion. The descriptive P0-P2 workflow-sensitivity gap was approximately 15.7% points, versus a 1.4-point partition-weighted fusion increment (95% bootstrap interval − 0.006 to 0.036). The leading fusion-related variants were numerically close, so no inferential ranking is claimed. At α = 0.10, global split conformal prediction achieved 0.896 empirical coverage while reviewing 20.8% of images and capturing 51.4% of top-1 errors; class-conditional Mondrian calibration achieved 0.923 coverage while reviewing 29.7% and capturing 64.8% of errors. Conclusions Defining the evidence unit is part of the statistical specification of a crop-image benchmark, not merely a data-cleaning step. In this corpus, benchmark interpretation was much more sensitive to evidence-workflow specification than to the evaluated representation refinement. The framework supports auditable internal evaluation and review triage but does not establish field, farm-level or intervention performance.","摘要 背景 当多个文件代表同一生物学证据或划分来源不明时，农业人工智能基准可能高估决策可靠性。我们评估了一个与分类器无关的“审计到复核”框架，该框架在模型比较前固定证据身份和角色来源，声明聚合估计目标，并将预测不确定性关联到复核工作量。一个包含7,595个文件的公开小麦叶片语料库经哈希处理得到5,118个唯一内容；资格标准定义了一个915个内容的五分类任务。将归档的文件级基准与三种冻结的近重复分组P2划分配置分离，并在划分加权和“一内容一权重”估计目标下评估了语义、纹理和融合表示。通过经验覆盖率、复核率、错误捕获率、复核产出率和无量纲复核效用评估分裂保形预测。结果 在遗留P0基准下，宏F1为0.962；仅语义P2为0.805 ± 0.030；直接融合为0.819 ± 0.029。描述性的P0-P2工作流敏感性差距约为15.7个百分点，而划分加权融合增量为1.4个百分点（95%自助法区间−0.006至0.036）。领先的融合相关变体在数值上接近，因此不声称推断性排序。在α = 0.10时，全局分裂保形预测实现了0.896的经验覆盖率，同时复核了20.8%的图像并捕获了51.4%的top-1错误；类条件Mondrian校准实现了0.923的覆盖率，同时复核了29.7%并捕获了64.8%的错误。结论 定义证据单元是作物图像基准统计规范的一部分，而不仅仅是数据清理步骤。在该语料库中，基准解释对证据-工作流规范的敏感性远高于对所评估表示改进的敏感性。该框架支持可审计的内部评估和复核分诊，但不能确立田间、农场级或干预性能。","Bulletin of the National Research Centre\u002FBulletin of the National Research Center","2026-09-14T00:00:00Z",77,{"impact":58,"substance":17,"depth":16,"authority":59,"freshness":60,"relevant":19,"comment":61},15,13,9,"该论文提出面向作物图像决策支持的证据完整性审计框架，揭示基准测试中证据单元定义对性能评估的显著影响，方法新颖、数据规模较大，对农业AI模型评估具有参考价值，但属细分领域方法学研究，产业影响有限。",[63],{"name":54,"url":51},[24,25,65,26,66],"遥感监测","模型评估",[68,69],"作物病害识别 农业人工智能 智慧农业 模型评估","作物病害识别 农业人工智能","作物病害识别农业人工智能智慧农业模型评估-2541","10.1186\u002Fs42269-026-01492-x",{"doi":71,"openalex_id":73,"authors":74,"venue":54,"cited_by_count":33,"oa_url":51,"card":83,"direction":87,"ingested_from":89},"W7212800882",[75,78,80],{"name":76,"orcid":77},"Xin Li","https:\u002F\u002Forcid.org\u002F0009-0005-0670-5006",{"name":79,"orcid":8},"Bojian Guo",{"name":81,"orcid":82},"A.Dzh. Kartanova","https:\u002F\u002Forcid.org\u002F0000-0003-1479-0747",{"tldr":84,"method":85,"finding":86,"direction":87,"opportunity":88},"提出审计到评审框架，先固定证据身份再比较作物图像分类器，并关联预测不确定性与评审工作量。","对7595个小麦叶片文件哈希去重，构建915内容五类任务，用分裂保形预测评估。","基准解释对证据工作流规范远比对表示优化敏感，P0与P2宏F1差约15.7个百分点。","农业人工智能与决策模型","可将证据单元定义与评审效用纳入农业AI基准标准，并探索田间级验证与主动学习结合。","openalex","2026-09-15T23:30:34.690905Z",{"id":92,"title":93,"url":94,"summary":95,"summary_zh":8,"content":8,"source_name":96,"source_url":8,"published_at":97,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":98,"score_detail":99,"sources":103,"tags":105,"search_phrases":108,"slug":110,"view_count":33,"doi":8,"paper":111,"created_at":118},2498,"Bangladesh AI Model AgroVisNet Spots Crop Blight on Phones(轻量级 99.52% 准确率 0.46 MB 量化部署)","https:\u002F\u002Fagritechinsights.com\u002Findex.php\u002F2026\u002F09\u002F10\u002Fbangladesh-ai-model-spots-crop-blight-on-phones","孟加拉国研究团队发表《AgroVisNet: A lightweight Convolutional Network and the BD-PlantDX Expert-Validated Benchmark for Radish, Potato and Pointed Gourd Disease Classification》。BD-PlantDX 基准数据集包含 12432 张高分辨率田间图像,涵盖 12 类(萝卜、马铃薯、尖瓜的健康与患病状态)。AgroVisNet 采用分组瓶颈残差块配以序列通道和空间注意力机制,在 BD-PlantDX 上达到 99.52% 测试准确率和相同加权 F1 分数。仅 290572 可训练参数,比所评估的 ImageNet 预训练轻量级骨干少 8.7-16.8 倍参数量。量化部署仅 0.46 MB,单 CPU 8.40 毫秒即可分类一张图像。","Agritech Insights 2026-09-10","2026-09-09T16:00:00Z",78,{"impact":16,"substance":17,"depth":16,"authority":100,"freshness":101,"relevant":19,"comment":102},12,8,"孟加拉国团队提出轻量级CNN模型AgroVisNet并发布万张级田间病害基准数据集，99.52%准确率、0.46MB量化模型可在手机端8.4毫秒推理，对发展中国家小农户病害识别具有实质参考价值。",[104],{"name":96,"url":94},[24,25,106,107,26],"边缘计算","小农农业",[109,69],"作物病害识别 农业人工智能 小农农业 智慧农业","作物病害识别农业人工智能小农农业智慧农业-2498",{"doi":8,"openalex_id":8,"authors":112,"venue":8,"cited_by_count":33,"oa_url":8,"card":113,"direction":87,"ingested_from":43},[],{"tldr":114,"method":115,"finding":116,"direction":87,"opportunity":117},"提出轻量网络AgroVisNet与BD-PlantDX数据集，实现手机端作物病害高精度识别。","构建12432张12类田间图像基准，用分组瓶颈残差块与注意力机制训练。","测试准确率99.52%，仅29万参数，量化后0.46MB，CPU单张8.4毫秒。","可探索跨作物跨区域泛化、田间复杂光照鲁棒性及边缘设备实时多病害检测。","2026-09-15T00:04:27.679059Z",{"id":120,"title":121,"url":122,"summary":123,"summary_zh":8,"content":8,"source_name":124,"source_url":8,"published_at":125,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":98,"score_detail":126,"sources":129,"tags":131,"search_phrases":135,"slug":138,"view_count":33,"doi":8,"paper":139,"created_at":146},2904,"Decoupled Foundation Models:基于YOLO26m+SAM2+DINOv2的湿度诱导番茄叶坏死实例分割与检测,登MDPI Agriculture 16(18)1997","https:\u002F\u002Fwww.mdpi.com\u002F2077-0472\u002F16\u002F18\u002F1997","本研究针对温室番茄相对湿度过高引发的非生物胁迫(生理性叶坏死,与生物感染症状相似),提出多步AI管道自动化分割与分类坏死叶斑。采集218张RGB图像、3218个标注(棕色坏死斑\u002F黄色坏死斑\u002F无坏死),系统评估6种端到端实例分割管道(YOLO26m检测+SAM2零样本分割+微调DINOv2或EfficientNet-B3分类);微调DINOv2宏F1达0.926,优于EfficientNet-B3、ResNet-50、Swin-Small基线(0.886-0.901);最佳配置mAP@50=0.828,较YOLO26m单模型提升约8%。","MDPI Agriculture","2026-09-17T00:00:00Z",{"impact":127,"substance":17,"depth":16,"authority":59,"freshness":60,"relevant":19,"comment":128},16,"方法组合新颖、数据规模与对比基线扎实，对温室番茄生理性叶坏死自动识别有实用价值，值得进入每日精选。",[130],{"name":124,"url":122},[24,25,132,133,134],"设施农业","番茄","病害识别",[136,137],"番茄叶坏死 实例分割","农业人工智能 智慧农业 病害识别 设施农业","番茄叶坏死实例分割-2904",{"doi":8,"openalex_id":8,"authors":140,"venue":8,"cited_by_count":33,"oa_url":8,"card":141,"direction":87,"ingested_from":43},[],{"tldr":142,"method":143,"finding":144,"direction":87,"opportunity":145},"用YOLO26m+SAM2+DINOv2多步管道分割并分类高湿诱导的番茄叶坏死斑。","218张RGB图像、3218个标注，评估6种实例分割管道并微调DINOv2分类。","微调DINOv2宏F1达0.926，最佳配置mAP@50=0.828，较单模型提升约8%。","可探索零样本基础模型在多种非生物胁迫症状上的泛化与轻量化温室部署。","2026-09-19T00:06:09.021594Z",{"id":148,"title":149,"url":150,"summary":151,"summary_zh":8,"content":8,"source_name":152,"source_url":8,"published_at":55,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":98,"score_detail":153,"sources":156,"tags":158,"search_phrases":162,"slug":165,"view_count":33,"doi":8,"paper":166,"created_at":173},2902,"TSAFI-DT:可持续性感知花生产量预测数字孪生框架,登MDPI AI 7(9)364","https:\u002F\u002Fwww.mdpi.com\u002F2673-2688\u002F7\u002F9\u002F364","本研究提出TSAFI-DT可回顾验证的、数据驱动的Digital Twin原型,集成时空数据重建、可持续状态表征、分层产量预测、反事实分析与情景模拟。基于1997-2023年印度地区级花生数据,采用贝叶斯优化的XGBoost模型进行一步前瞻产量预测,RMSE=0.171 t\u002Fha,显著优于基线;结合固定效应与合成控制分析,eRAI扩展再生农业指数整合作物多样性、生产力稳定性、土地利用效率和产量趋势,预测产量在可持续性扰动下可提升12.4%。","MDPI AI",{"impact":16,"substance":17,"depth":154,"authority":59,"freshness":46,"relevant":19,"comment":155},19,"方法新颖、数据规模扎实的农业数字孪生研究，对智慧农业与产量预测领域有参考价值，但属细分学术进展，非产业级事件。",[157],{"name":152,"url":150},[24,25,159,160,161],"数字孪生","可持续农业","花生产量预测",[163,164],"TSAFI-DT 花生 数字孪生","印度 花生 产量预测","TSAFI-DT花生数字孪生-2902",{"doi":8,"openalex_id":8,"authors":167,"venue":8,"cited_by_count":33,"oa_url":8,"card":168,"direction":87,"ingested_from":43},[],{"tldr":169,"method":170,"finding":171,"direction":87,"opportunity":172},"提出可持续性感知数字孪生框架TSAFI-DT，用于印度花生产量预测与情景模拟。","基于1997-2023年印度地区级数据，用贝叶斯优化XGBoost和合成控制分析","XGBoost预测RMSE为0.171 t\u002Fha，可持续性扰动下产量可提升12.4%。","可探索将数字孪生与实时物联网数据结合，实现动态可持续性评估与决策支持。","2026-09-19T00:06:08.754978Z",{"id":175,"title":176,"url":177,"summary":178,"summary_zh":8,"content":8,"source_name":179,"source_url":8,"published_at":125,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":180,"score_detail":181,"sources":183,"tags":185,"search_phrases":189,"slug":192,"view_count":33,"doi":8,"paper":193,"created_at":200},2901,"AgriScope:面向农业图像的像素级多模态理解统一框架,arXiv 2609.20325(预印本)","https:\u002F\u002Farxiv.org\u002Fabs\u002F2609.20325","Boudiaf、Alanssari、Hussain、Javed提出AgriScope,一个统一的像素级多模态农业图像理解框架,联合支持图像级、区域级、像素级理解,可实现接地描述生成、指代表达分割、多轮多模态交互等任务。集成生物专用语义表征、密集空间表征与像素解码;引入大规模像素级农业多模态指令调优数据集AgriGround,包含50万+图像和1100万+指令跟随样本,涵盖植物病害分析、作物与杂草识别、昆虫识别、细粒度植物理解。实验表明AgriScope在多项农业视觉语言任务上有效。","arXiv (preprint)",75,{"impact":16,"substance":17,"depth":16,"authority":101,"freshness":60,"relevant":19,"comment":182},"提出统一像素级农业多模态理解框架并开源50万图像、1100万指令样本的大规模数据集，方法新颖、数据规模突出，但为arXiv预印本、未经同行评审，权威性有限，值得作为前沿技术动态精选。",[184],{"name":179,"url":177},[24,25,186,187,188],"农业遥感","植物病害识别","多模态大模型",[190,191],"AgriScope 农业图像 多模态","AgriGround 像素级 农业数据集","AgriScope农业图像多模态-2901",{"doi":8,"openalex_id":8,"authors":194,"venue":8,"cited_by_count":33,"oa_url":8,"card":195,"direction":87,"ingested_from":43},[],{"tldr":196,"method":197,"finding":198,"direction":87,"opportunity":199},"提出AgriScope统一框架，实现农业图像像素级多模态理解与多任务交互。","构建AgriGround数据集（50万+图像、1100万+指令样本），融合语义与","AgriScope在接地描述、指代分割、多轮交互等农业视觉语言任务上有效。","可探索像素级多模态模型在田间实时病害诊断与精准施药决策中的落地与轻量化。","2026-09-19T00:06:08.678379Z",{"id":202,"title":203,"url":204,"summary":205,"summary_zh":8,"content":8,"source_name":206,"source_url":8,"published_at":55,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":207,"score_detail":208,"sources":211,"tags":213,"search_phrases":218,"slug":221,"view_count":33,"doi":8,"paper":222,"created_at":229},2900,"Crop-GPA 2.0:安徽农业大学团队发布跨物种基因-表型预测深度学习工具,登The Crop Journal","https:\u002F\u002Fwww.global-agriculture.com\u002Fag-tech-research-news\u002Fchinese-researchers-launch-ai-tool-that-predicts-crop-gene-trait-links-across-species\u002F","安徽农业大学岳振宇教授团队(第一作者高玉佳)开发的Crop-GPA 2.0跨物种基因-表型预测深度学习工具于2026年9月10日发表在The Crop Journal。该工具通过分层基因组表征、跨物种预训练、性状感知学习三种联动技术,在水稻、玉米、小麦测试中跨性状(产量、抗病性、抗逆性、籽粒品质)均优于现有方法,且能在新物种或训练数据有限时保持较好性能,在线平台为育种者提供免费SNP排名服务,极大降低标记辅助选择门槛。","Global Agriculture \u002F The Crop Journal",83,{"impact":17,"substance":209,"depth":16,"authority":18,"freshness":46,"relevant":19,"comment":210},23,"跨物种基因-表型预测工具登核心期刊，方法新颖、结论可靠，对智能育种有实质推动，值得进入每日精选。",[212],{"name":206,"url":204},[24,25,214,215,216,217],"水稻","智能育种","玉米","基因-表型预测",[219,220],"安徽农业大学 Crop-GPA 2.0","Crop-GPA 2.0 基因-表型预测","安徽农业大学Crop-GPA2.0-2900",{"doi":8,"openalex_id":8,"authors":223,"venue":8,"cited_by_count":33,"oa_url":8,"card":224,"direction":87,"ingested_from":43},[],{"tldr":225,"method":226,"finding":227,"direction":87,"opportunity":228},"安徽农业大学团队发布跨物种基因-表型预测深度学习工具Crop-GPA 2.0，并在水稻、玉米、小麦中","分层基因组表征、跨物种预训练、性状感知学习，基于多物种SNP与表型数据。","跨性状预测均优于现有方法，新物种或小样本下仍表现良好，并提供免费在线SNP排名服务。","可探索将跨物种预训练模型与田间表型组、环境数据融合，提升复杂性状预测与育种决策的可解释性。","2026-09-19T00:06:08.595080Z"]