[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-4968":3,"related-4968":82},{"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,"primary_topic":29,"secondary_topics":35,"classification":36,"search_phrases":66,"slug":69,"view_count":19,"doi":70,"paper":71,"created_at":81},4968,"A real-time deep learning method for detection and localization of non-emerged cabbage seedlings","https:\u002F\u002Fwww.frontiersin.org\u002Fjournals\u002Fplant-science\u002Farticles\u002F10.3389\u002Ffpls.2026.1984989\u002Ffull","Chen, Han, Liu, Kong, Maity, Peng, Di, Yu and Yu(《Frontiers in Plant Science》, 2026-10-14发表)提出基于无人机的精确、快速、实时未出苗甘蓝幼苗检测与定位方法。Enhanced YOLOv8n通过更高效骨干网络和双向特征融合颈部结构,达到81.4% mAP50-95,较基线提升8%,同时保持实时推理能力。整合Enhanced YOLOv8n、YOLO跟踪算法和图像处理算法,实现鲁棒的实时未出苗幼苗检测与定位方法,未出苗甘蓝幼苗识别准确率达96.9%,且可区分连续未出苗幼苗数量,人工成本降至原1\u002F5。",null,"Frontiers in Plant Science 2026 年 10 月 14 日","2026-10-14T00:00:00Z","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、数据扎实且落地价值明确，但属细分作物技术进展，未达重大突破层级。",[22],{"name":9,"url":6},[24,25,26,27,28],"智慧农业","农业人工智能","甘蓝","无人机遥感","幼苗检测",{"code":30,"slug":31,"name":32,"path":33,"parent_code":34},"I2","agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai","I",[],{"status":37,"locked":13,"version":38,"method":39,"evidence":40},"classified","2026-10-22.2","rules",{"candidates":41,"agricultural_evidence":44},[42,50,56,61],{"code":30,"score":43,"supported":44,"title":45,"abstract":47,"tags":49},12,true,[46],"deep learning",[48],"YOLO",[25],{"code":51,"score":52,"supported":44,"title":53,"abstract":54,"tags":55},"B1",4,[],[26],[26],{"code":57,"score":19,"supported":13,"title":58,"abstract":59,"tags":60},"I1",[],[],[24],{"code":62,"score":19,"supported":13,"title":63,"abstract":64,"tags":65},"I3",[],[],[27],[67,68],"甘蓝 未出苗 幼苗检测","无人机 甘蓝 实时检测","甘蓝未出苗幼苗检测-4968","10.3389\u002Ffpls.2026.1984989\u002Ffull",{"doi":70,"openalex_id":8,"authors":72,"venue":8,"cited_by_count":73,"oa_url":8,"card":74,"direction":78,"ingested_from":80},[],0,{"tldr":75,"method":76,"finding":77,"direction":78,"opportunity":79},"提出无人机实时检测定位未出苗甘蓝幼苗的深度学习方法，准确率96.9%。","改进YOLOv8n骨干与双向特征融合颈部，结合YOLO跟踪和图像处理。","Enhanced YOLOv8n的mAP50-95达81.4%，未出苗识别准确率96.9%，人工成本","农业遥感与作物表型","可迁移到其他作物出苗监测，并探索多光谱融合与边缘端轻量化实时部署。","agent","2026-10-15T00:06:08.716058Z",{"total":83,"page":19,"page_size":83,"items":84},6,[85,147,238,312,379,440],{"id":86,"title":87,"url":88,"summary":89,"summary_zh":8,"content":8,"source_name":90,"source_url":8,"published_at":91,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":92,"score_detail":93,"sources":99,"tags":101,"primary_topic":104,"secondary_topics":110,"classification":112,"search_phrases":135,"slug":138,"view_count":73,"doi":8,"paper":139,"created_at":146},5532,"精准棉花病害管理：融合无人机影像与定制YOLO深度学习框架","https:\u002F\u002Farxiv.org\u002Fabs\u002F2610.24182","研究提出单阶段全卷积检测器Cotton-Vision，融合幽灵瓶颈通道注意力与跨阶段部分融合三项轻量化设计，参数减少20%。在巴基斯坦旁遮普省7个棉田采集的2700余幅高分辨率无人机影像上测试，精确率97.7%、召回率77.7%、mAP@50为85.6%，较主流基线提升10至20个百分点。预印本，仅在巴基斯坦单一区域验证。","arXiv 预印本 2026-10-16","2026-10-16T00:00:00Z",69,{"impact":94,"substance":95,"depth":96,"authority":97,"freshness":83,"relevant":19,"comment":98},16,21,17,9,"方法轻量化与实测数据有增量，但仅单区域预印本验证，属细分技术进展，可入精选备选。",[100],{"name":90,"url":88},[24,25,102,27,103],"目标检测","棉花病害",{"code":105,"slug":106,"name":107,"path":108,"parent_code":109},"A6","cash-crops","棉麻糖与特色经作","\u002Ftopic\u002Ffield-crops\u002Fcash-crops","A",[111],{"code":30,"slug":31,"name":32,"path":33,"parent_code":34},{"status":37,"locked":13,"version":38,"method":39,"evidence":113},{"candidates":114,"agricultural_evidence":44},[115,120,127,131],{"code":30,"score":16,"supported":44,"title":116,"abstract":118,"tags":119},[117,48],"深度学习",[],[25,102],{"code":105,"score":121,"supported":44,"title":122,"abstract":124,"tags":126},11,[123],"棉花",[125],"cotton",[],{"code":57,"score":19,"supported":13,"title":128,"abstract":129,"tags":130},[],[],[24],{"code":62,"score":19,"supported":13,"title":132,"abstract":133,"tags":134},[],[],[27],[136,137],"Cotton-Vision 棉花病害","无人机影像 棉花 检测","Cotton-Vision棉花病害-5532",{"doi":8,"openalex_id":8,"authors":140,"venue":8,"cited_by_count":73,"oa_url":8,"card":141,"direction":78,"ingested_from":80},[],{"tldr":142,"method":143,"finding":144,"direction":78,"opportunity":145},"提出轻量化检测器Cotton-Vision，用无人机影像精准识别棉花病害。","基于YOLO改进，融合幽灵瓶颈通道注意力与跨阶段部分融合，用2700余幅无人机影","精确率97.7%、mAP@50达85.6%，较主流基线提升10至20个百分点。","仅在巴基斯坦单一区域验证，可扩展至多区域多品种，并探索边缘设备实时部署。","2026-10-22T00:07:09.816790Z",{"id":148,"title":149,"url":150,"summary":151,"summary_zh":152,"content":8,"source_name":153,"source_url":150,"published_at":154,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":155,"score_detail":156,"sources":159,"tags":161,"primary_topic":164,"secondary_topics":165,"classification":166,"search_phrases":195,"slug":198,"view_count":52,"doi":199,"paper":200,"created_at":237},4010,"Intelligent Automated Crop Monitoring System Based on Unmanned Aerial Vehicles and Deep Learning for Smart Agriculture","https:\u002F\u002Fdoi.org\u002F10.3390\u002Fapp16199747","The rapid development of precision agriculture technologies requires intelligent systems capable of automatically analyzing unmanned aerial vehicle (UAV) imagery and transforming image-processing results into structured information for crop-monitoring decision support. This study develops and computationally validates an integrated framework combining deep learning-based semantic segmentation, RGB-based spatial interpretation, image-space vectorization of candidate RGB heterogeneity zones, and monitoring-priority assessment. Experimental validation was performed using the open-access dataset “High-Resolution RGB Images and Corresponding Masks of Agricultural Fields”, comprising mixed agricultural vegetation scenes. Following data-integrity verification, 326 valid RGB image–reference mask pairs were retained and divided into 21,324 non-overlapping 512 × 512-pixel patches. U-Net-ResNet50, DeepLabV3-ResNet50, and FCN-ResNet50 were evaluated in three repeated source-image-grouped holdout experiments using the full dataset. U-Net-ResNet50 achieved the highest held-out performance, with mean Precision = 0.9946 ± 0.0008, Recall = 0.9931 ± 0.0021, vegetation IoU = 0.9877 ± 0.0020, and Dice = 0.9938 ± 0.0010, and was therefore selected for subsequent spatial analysis. Its segmentation outputs were combined with VARI, ExG, and GLI indices to identify candidate RGB heterogeneity zones, which were converted into image-space vector objects in source-image pixel coordinates. The revised full-data analysis generated 58,206 candidate RGB heterogeneity zones. These regions represent visible canopy heterogeneity rather than physiologically confirmed crop stress and are intended to support targeted field verification and monitoring prioritization. The results demonstrate the feasibility of an end-to-end UAV RGB processing workflow that transforms pixel-level segmentation outputs into structured spatial information for precision-agriculture monitoring and decision support.","精准农业技术的快速发展需要智能系统能够自动分析无人机（UAV）影像，并将图像处理结果转化为结构化信息，以支持作物监测决策。本研究开发并通过计算验证了一个集成框架，该框架结合了基于深度学习的语义分割、基于RGB的空间解译、候选RGB异质性区域的图像空间矢量化以及监测优先级评估。实验验证使用开放获取数据集“High-Resolution RGB Images and Corresponding Masks of Agricultural Fields”进行，该数据集包含混合农业植被场景。经数据完整性验证后，保留了326对有效RGB图像-参考掩膜对，并划分为21,324个不重叠的512 × 512像素图像块。使用完整数据集，在三次重复的以源图像分组的留出实验中评估了U-Net-ResNet50、DeepLabV3-ResNet50和FCN-ResNet50。U-Net-ResNet50取得了最高的留出性能，平均精确率 = 0.9946 ± 0.0008，召回率 = 0.9931 ± 0.0021，植被IoU = 0.9877 ± 0.0020，Dice = 0.9938 ± 0.0010，因此被选用于后续空间分析。其分割输出与VARI、ExG和GLI指数相结合，用于识别候选RGB异质性区域，并将其转换为源图像像素坐标下的图像空间矢量对象。修订后的全数据分析生成了58,206个候选RGB异质性区域。这些区域代表可见的冠层异质性，而非经生理学确认的作物胁迫，旨在支持有针对性的田间验证和监测优先级排序。结果表明，端到端无人机RGB处理工作流程具有可行性，该流程可将像素级分割输出转化为结构化空间信息，用于精准农业监测与决策支持。","Applied Sciences","2026-10-01T00:00:00Z",80,{"impact":16,"substance":17,"depth":16,"authority":157,"freshness":97,"relevant":19,"comment":158},13,"基于无人机RGB影像与深度学习语义分割的端到端作物监测框架，数据规模大、指标可靠，对精准农业决策支持有实质参考价值。",[160],{"name":153,"url":150},[24,25,27,162,163],"作物监测","图像分割",{"code":30,"slug":31,"name":32,"path":33,"parent_code":34},[],{"status":37,"locked":13,"version":38,"method":39,"evidence":167},{"candidates":168,"agricultural_evidence":44},[169,175,183,190],{"code":30,"score":170,"supported":44,"title":171,"abstract":172,"tags":174},19,[46],[117,46,173],"decision support",[25,163],{"code":57,"score":176,"supported":44,"title":177,"abstract":179,"tags":182},15,[178],"smart agriculture",[180,181],"精准农业","precision agriculture",[24],{"code":184,"score":83,"supported":44,"title":185,"abstract":186,"tags":189},"I4",[],[187,188],"空间分析","spatial analysis",[],{"code":62,"score":191,"supported":13,"title":192,"abstract":193,"tags":194},2,[],[],[162,27],[196,197],"UAV 作物监测 深度学习","U-Net 植被分割 无人机","UAV作物监测深度学习-4010","10.3390\u002Fapp16199747",{"doi":199,"openalex_id":201,"authors":202,"venue":153,"cited_by_count":73,"oa_url":150,"card":230,"direction":235,"ingested_from":236},"W7215074401",[203,206,209,212,214,216,219,221,224,227],{"name":204,"orcid":205},"Аnatoliy M. Тryhuba","https:\u002F\u002Forcid.org\u002F0000-0001-8014-5661",{"name":207,"orcid":208},"Paweł Kiełbasa","https:\u002F\u002Forcid.org\u002F0000-0003-0249-8626",{"name":210,"orcid":211},"Nazarii Ya. Koval","https:\u002F\u002Forcid.org\u002F0000-0001-7846-2924",{"name":213,"orcid":8},"Inna Tryhuba",{"name":215,"orcid":8},"Oleh Andrushkiv",{"name":217,"orcid":218},"Vitalij Grabovets","https:\u002F\u002Forcid.org\u002F0000-0002-0340-185X",{"name":220,"orcid":8},"Nataliia Kozak",{"name":222,"orcid":223},"Akinniyi Akinsunmade","https:\u002F\u002Forcid.org\u002F0000-0002-1602-1741",{"name":225,"orcid":226},"Anna Miernik","https:\u002F\u002Forcid.org\u002F0000-0003-0394-9185",{"name":228,"orcid":229},"Paweł Pysz","https:\u002F\u002Forcid.org\u002F0000-0001-7083-1078",{"tldr":231,"method":232,"finding":233,"direction":78,"opportunity":234},"构建无人机RGB影像深度学习分割与空间矢量化框架，实现作物监测优先级评估。","U-Net-ResNet50等模型在326张影像21,324个512×512块上","U-Net-ResNet50分割最优（IoU 0.9877），生成58,206个候选异质区用于田间验","可延伸研究：将RGB异质区与多光谱\u002F热红外数据融合，验证生理胁迫并构建实时决策模型。","智慧农业 \u002F 农业物联网","openalex","2026-10-02T23:30:11.747839Z",{"id":239,"title":240,"url":241,"summary":242,"summary_zh":243,"content":8,"source_name":244,"source_url":241,"published_at":245,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":246,"score_detail":247,"sources":251,"tags":253,"primary_topic":256,"secondary_topics":257,"classification":258,"search_phrases":288,"slug":291,"view_count":191,"doi":292,"paper":293,"created_at":311},3065,"A MULTIMODAL FRAMEWORK FOR LIVESTOCK DETECTION AND EPIZOOTIC MONITORING","https:\u002F\u002Fdoi.org\u002F10.67048\u002Fqxju2026ai128iss7m1055","This study proposes an automated computer vision framework to detect and monitor mixed herds of cattle and sheep. By integrating unmanned aerial vehicle (UAV) imagery with ground-level perspectives, a multimodal dataset was curated. Using this data, a lightweight, single-stage object detector (YOLOv8n), tailored for resource-constrained edge devices, was trained and optimised. The integration of computer vision and UAV technology offers agricultural agencies a scalable tool for proactive epizootic surveillance.","本研究提出了一种自动化计算机视觉框架，用于检测和监测牛与羊的混合畜群。通过将无人机（UAV）影像与地面视角相结合，构建了一个多模态数据集。利用该数据，训练并优化了一个轻量级单阶段目标检测器（YOLOv8n），专为资源受限的边缘设备而设计。计算机视觉与无人机技术的融合为农业机构提供了一种可扩展的工具，用于主动进行动物流行病监测。","Agro ilm","2026-09-18T00:00:00Z",71,{"impact":16,"substance":248,"depth":96,"authority":249,"freshness":249,"relevant":19,"comment":250},20,8,"多模态无人机视觉框架用于牛羊混群检测与疫病预警，方法具体、面向边缘设备，对智慧畜牧有实用参考价值。",[252],{"name":244,"url":241},[24,25,27,254,255],"畜牧养殖","疫病监测",{"code":30,"slug":31,"name":32,"path":33,"parent_code":34},[],{"status":37,"locked":13,"version":38,"method":39,"evidence":259},{"candidates":260,"agricultural_evidence":44},[261,267,273,280,284],{"code":30,"score":12,"supported":44,"title":262,"abstract":263,"tags":266},[],[102,264,265],"计算机视觉","computer vision",[25],{"code":268,"score":249,"supported":44,"title":269,"abstract":271,"tags":272},"C",[270],"livestock",[],[],{"code":274,"score":83,"supported":44,"title":275,"abstract":276,"tags":279},"C2",[],[277,278],"cattle","sheep",[],{"code":57,"score":19,"supported":13,"title":281,"abstract":282,"tags":283},[],[],[24],{"code":62,"score":19,"supported":13,"title":285,"abstract":286,"tags":287},[],[],[27],[289,290],"YOLOv8n 牲畜检测","无人机 畜牧 疫病监测","YOLOv8n牲畜检测-3065","10.67048\u002Fqxju2026ai128iss7m1055",{"doi":292,"openalex_id":294,"authors":295,"venue":244,"cited_by_count":73,"oa_url":305,"card":306,"direction":235,"ingested_from":236},"W7213668410",[296,299,302],{"name":297,"orcid":298},"R. N. Khadzhaev","https:\u002F\u002Forcid.org\u002F0009-0001-8124-2399",{"name":300,"orcid":301},"Sa’dulla Avezbayev","https:\u002F\u002Forcid.org\u002F0009-0008-8896-7078",{"name":303,"orcid":304},"Sayfuddin Sharipov","https:\u002F\u002Forcid.org\u002F0000-0002-4232-1369","https:\u002F\u002Fqxjurnal.uz\u002Findex.php\u002Fai\u002Farticle\u002Fdownload\u002F1055\u002F936",{"tldr":307,"method":308,"finding":309,"direction":235,"opportunity":310},"提出多模态计算机视觉框架，用无人机和地面图像检测牛羊并监测疫情。","融合无人机与地面视角构建多模态数据集，训练轻量级YOLOv8n边缘检测器。","轻量级模型可在资源受限设备上实现混合畜群检测，支持主动疫情监测。","可探索多物种、复杂地形下的实时边缘检测与疫情早期预警模型优化。","2026-09-21T23:30:13.271856Z",{"id":313,"title":314,"url":315,"summary":316,"summary_zh":8,"content":8,"source_name":317,"source_url":8,"published_at":318,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":319,"score_detail":320,"sources":322,"tags":324,"primary_topic":327,"secondary_topics":332,"classification":338,"search_phrases":367,"slug":370,"view_count":191,"doi":8,"paper":371,"created_at":378},2854,"面向冬小麦水分含量的无人机遥感自动机器学习预测——MDPI Remote Sensing","https:\u002F\u002Fwww.mdpi.com\u002F2072-4292\u002F18\u002F18\u002F3161","本研究探索了无人机遥感快速准确评估冬小麦水分含量的潜力。在开花期和灌浆期使用配备多光谱、RGB和热红外相机的无人机获取高分辨率冠层遥感图像。集成地面真值采样数据与无人机遥感数据,使用自动机器学习(AutoML)框架建立回归模型预测冬小麦水分含量(MC)。结果表明,MC预测在灌浆期表现最佳,TIR传感器精度最高(R²=0.812,MAE=0.0204,RMSE=0.0274)。多传感器融合相比单传感器方法进一步提升预测性能,MC预测的R²达0.876、MAE 0.0191、RMSE 0.0259。来自中国农业科学院农田灌溉研究所。","MDPI Remote Sensing","2026-09-15T00:00:00Z",74,{"impact":176,"substance":95,"depth":96,"authority":157,"freshness":249,"relevant":19,"comment":321},"中国农科院团队用AutoML融合多光谱、RGB与热红外无人机数据预测冬小麦水分含量，多传感器融合R²达0.876，方法新颖、结论可靠，对精准灌溉有实用价值，值得进入每日精选。",[323],{"name":317,"url":315},[24,25,27,325,326],"冬小麦","多传感器融合",{"code":328,"slug":329,"name":330,"path":331,"parent_code":109},"A2","wheat","小麦","\u002Ftopic\u002Ffield-crops\u002Fwheat",[333,337],{"code":62,"slug":334,"name":335,"path":336,"parent_code":34},"remote-sensing","遥感与农情监测","\u002Ftopic\u002Fdigital-agriculture\u002Fremote-sensing",{"code":30,"slug":31,"name":32,"path":33,"parent_code":34},{"status":37,"locked":13,"version":38,"method":39,"evidence":339},{"candidates":340,"agricultural_evidence":44},[341,347,351,356,363],{"code":62,"score":248,"supported":44,"title":342,"abstract":345,"tags":346},[343,344],"遥感","remote sensing",[343],[27],{"code":328,"score":43,"supported":44,"title":348,"abstract":349,"tags":350},[330],[330],[325],{"code":30,"score":43,"supported":44,"title":352,"abstract":354,"tags":355},[353],"机器学习",[353],[25],{"code":357,"score":358,"supported":44,"title":359,"abstract":360,"tags":362},"H3",3,[],[361],"灌溉",[],{"code":57,"score":19,"supported":13,"title":364,"abstract":365,"tags":366},[],[],[24],[368,369],"农业人工智能 多传感器融合 无人机遥感 智慧农业","农业人工智能 多传感器融合","农业人工智能多传感器融合无人机遥感智慧农业-2854",{"doi":8,"openalex_id":8,"authors":372,"venue":8,"cited_by_count":73,"oa_url":8,"card":373,"direction":78,"ingested_from":80},[],{"tldr":374,"method":375,"finding":376,"direction":78,"opportunity":377},"用无人机多传感器遥感结合AutoML预测冬小麦水分含量。","无人机多光谱、RGB、热红外图像+地面真值，AutoML回归建模。","灌浆期热红外精度最高，多传感器融合将R²提升至0.876。","可探索不同生育期与品种的泛化性，及将水分预测接入灌溉决策系统。","2026-09-18T00:03:30.758419Z",{"id":380,"title":381,"url":382,"summary":383,"summary_zh":384,"content":8,"source_name":385,"source_url":382,"published_at":386,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":92,"score_detail":387,"sources":389,"tags":391,"primary_topic":395,"secondary_topics":401,"classification":403,"search_phrases":419,"slug":422,"view_count":73,"doi":423,"paper":424,"created_at":439},5592,"Evaluation of Concept-Rank Collapse Audits for Computer Vision Agriculture Disease Classifiers Under Low-Label Regimes: An Interpretability Capacity Bound for Smallholder Fungicide Selectivity","https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23545193","Concept-based interpretability methods for plant-disease classifiers are typically evaluated by whether their concept activations correlate with human-annotated labels, but this criterion is silent on a prior question: whether the classifier's internal representation retains enough concept-rank structure for any audit to be informative. the present analysis formalises that prior question as an interpretability capacity bound. A classifier is modelled as a concept-rank operator whose singular values govern how much concept-level signal survives the projection from pixel space to class logits, and a collapse audit is defined as the ratio of retained concept-rank energy to total concept energy. Under a low-label regime, where concept annotations are scarce and the audit must be estimated from few labelled examples, the bound predicts that audit reliability degrades as a function of the concept-rank spectrum and the label budget. The analysis is a theoretical and methodological contribution: it derives the bound, states the conditions under which a collapse audit is identifiable, and characterises the operating characteristics of the audit as a decision procedure under a declared variance structure (between-cluster SD = 6.0, residual SD = 4.5, intraclass correlation = 0.64). Power was estimated across 2000 design cells for cluster counts of 8, 16, 24 and 32 and effect sizes of 3, 5 and 8 percentage points; 52 clusters were required for 5 percentage points at 80% power. The bound reframes concept audits as capacity tests rather than fidelity tests, with consequences for how smallholder fungicide-selectivity decisions should be supported by interpretable disease classifiers.","面向植物病害分类器的基于概念的可解释性方法，通常以其概念激活是否与人工标注标签相关作为评价标准，但该标准对一个更先前的问题保持沉默：分类器的内部表征是否保留了足够的概念秩结构，从而使任何审计具有信息量。本分析将该先前问题形式化为一个可解释性容量上界。分类器被建模为概念秩算子，其奇异值支配着从像素空间到类别logits的投影中概念级信号得以保留的程度，并将坍缩审计定义为保留的概念秩能量与总概念能量之比。在低标签情形下，概念标注稀缺且审计必须由少量标注样本估计，该上界预测审计可靠性会随概念秩谱和标签预算的函数而退化。本分析是一项理论和方法论贡献：推导出该上界，陈述坍缩审计可识别的条件，并在声明的方差结构（簇间SD = 6.0，残差SD = 4.5，组内相关系数 = 0.64）下刻画审计作为决策程序的操作特征。在2000个设计单元中，针对簇数为8、16、24和32以及效应量为3、5和8个百分点的情形估计了功效；在80%功效下达到5个百分点需要52个簇。该上界将概念审计重新框定为容量检验而非保真度检验，并对可解释病害分类器应如何支持小农杀菌剂选择性决策具有启示意义。","Zenodo (CERN European Organization for Nuclear Research)","2026-10-21T00:00:00Z",{"impact":249,"substance":95,"depth":16,"authority":43,"freshness":12,"relevant":19,"comment":388},"理论方法论文，提出可解释性容量上界并给出审计功效分析，对智慧农业病害识别可信度有参考价值，但属小众方法学研究、缺乏田间实证，暂不宜进入每日精选。",[390],{"name":385,"url":382},[24,25,392,393,394],"小农户","植物病害识别","可解释性",{"code":396,"slug":397,"name":398,"path":399,"parent_code":400},"G2","farm-services","社会化服务与规模经营","\u002Ftopic\u002Frural-services\u002Ffarm-services","G",[402],{"code":30,"slug":31,"name":32,"path":33,"parent_code":34},{"status":37,"locked":13,"version":38,"method":39,"evidence":404},{"candidates":405,"agricultural_evidence":44},[406,411,415],{"code":396,"score":43,"supported":44,"title":407,"abstract":409,"tags":410},[408],"smallholder",[408],[392],{"code":30,"score":12,"supported":44,"title":412,"abstract":413,"tags":414},[265],[],[25,393],{"code":57,"score":19,"supported":13,"title":416,"abstract":417,"tags":418},[],[],[24],[420,421],"农业人工智能 植物病害识别 可解释性 智慧农业","农业人工智能 植物病害识别","农业人工智能植物病害识别可解释性智慧农业-5592","10.5281\u002Fzenodo.23545193",{"doi":423,"openalex_id":425,"authors":426,"venue":385,"cited_by_count":73,"oa_url":382,"card":433,"direction":437,"ingested_from":236},"W7223137323",[427,429,431],{"name":428,"orcid":8},"Akshay Ramgoolam",{"name":430,"orcid":8},"Priya Bundhoo Appadoo",{"name":432,"orcid":8},"Jean-Noël LEspérance",{"tldr":434,"method":435,"finding":436,"direction":437,"opportunity":438},"提出可解释性容量上界，评估低标签下概念秩崩溃审计的可靠性。","将分类器建模为概念秩算子，推导容量上界并做2000格功效模拟。","审计可靠性随概念秩谱与标签预算下降；5个百分点效应需52个聚类达80%功效。","农业人工智能与决策模型","可将容量上界用于小农户杀菌剂选择性决策的可解释模型审计与标签预算设计。","2026-10-22T11:21:10.425290Z",{"id":441,"title":442,"url":443,"summary":444,"summary_zh":445,"content":8,"source_name":385,"source_url":443,"published_at":386,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":92,"score_detail":446,"sources":448,"tags":450,"primary_topic":454,"secondary_topics":459,"classification":461,"search_phrases":484,"slug":487,"view_count":73,"doi":488,"paper":489,"created_at":501},5591,"Hierarchical Conformal Risk Control for Artificial Intelligence Agriculture Irrigation Scheduling Under Unmodeled Infiltration: A Computer Safety-Guarantee Evaluation","https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23543665","Irrigation scheduling algorithms that rely on soil-moisture forecasts are vulnerable to unmodeled infiltration, the process by which water bypasses the root zone through preferential pathways that resistance-based soil models cannot represent. The framework is evaluated through a replication-based design study that characterises the operating characteristics of the before-after-control-impact estimator used to audit scheduler performance, under a declared variance structure with between-cluster standard deviation 6.0, residual standard deviation 4.5 and intraclass correlation 0.64. Power was estimated across 2000 design cells spanning cluster counts of 8, 16, 24 and 32 and effect sizes of 3, 5 and 8 percentage points. The analysis indicates that 54 clusters are required to detect a 5-percentage-point effect at 80% power, and that the minimum detectable effect declines steeply as cluster count rises from 8 to 32. The contribution is a formal safety guarantee for irrigation decision support that remains valid when the infiltration model is wrong, together with an audit design that makes the guarantee empirically checkable.","依赖土壤湿度预报的灌溉调度算法容易受到未建模入渗的影响，即水分通过优先路径绕过根区的过程，而基于阻力的土壤模型无法表征这一过程。该框架通过一项基于重复的设计研究进行评估，该研究刻画了用于审计调度器性能的前后对照影响估计量的运行特性，其所依据的声明方差结构为：组间标准差6.0、残差标准差4.5、组内相关系数0.64。功效估计覆盖2000个设计单元，涵盖聚类数8、16、24和32以及效应量3、5和8个百分点。分析表明，在80%功效下检测5个百分点效应需要54个聚类，且最小可检测效应随聚类数从8增至32而急剧下降。本文的贡献在于为灌溉决策支持提供了一种正式的安全保证，该保证在入渗模型错误时仍然有效，同时提出了一种使该保证可经实证检验的审计设计。",{"impact":249,"substance":17,"depth":16,"authority":43,"freshness":97,"relevant":19,"comment":447},"该研究提出在入渗模型失配条件下仍有效的灌溉决策安全保证框架，并给出可审计的试验设计，方法新颖、结论可靠，但属细分方法学研究，产业影响有限，适合作为专业精选而非头条。",[449],{"name":385,"url":443},[24,25,451,452,453],"灌溉决策","安全保证","试验设计",{"code":357,"slug":455,"name":456,"path":457,"parent_code":458},"irrigation","水资源与灌溉","\u002Ftopic\u002Fenvironment\u002Firrigation","H",[460],{"code":30,"slug":31,"name":32,"path":33,"parent_code":34},{"status":37,"locked":13,"version":38,"method":39,"evidence":462},{"candidates":463,"agricultural_evidence":44},[464,468,473,480],{"code":357,"score":176,"supported":44,"title":465,"abstract":466,"tags":467},[455],[361,455],[451],{"code":30,"score":43,"supported":44,"title":469,"abstract":471,"tags":472},[470],"artificial intelligence",[173],[25],{"code":474,"score":83,"supported":44,"title":475,"abstract":476,"tags":479},"H2",[],[477,478],"土壤","soil",[],{"code":57,"score":19,"supported":13,"title":481,"abstract":482,"tags":483},[],[],[24],[485,486],"灌溉调度 安全保证","农业人工智能 安全保证 智慧农业 灌溉决策","灌溉调度安全保证-5591","10.5281\u002Fzenodo.23543665",{"doi":488,"openalex_id":490,"authors":491,"venue":385,"cited_by_count":73,"oa_url":443,"card":496,"direction":437,"ingested_from":236},"W7223131205",[492,494],{"name":493,"orcid":8},"Nassima Benkhelifa",{"name":495,"orcid":8},"Yacine Aït Ahmed",{"tldr":497,"method":498,"finding":499,"direction":437,"opportunity":500},"提出分层保形风险控制，为未建模入渗下的灌溉调度提供安全保证。","分层保形风险控制与基于重复设计的BACI审计，2000个设计单元功效分析。","检测5个百分点效应需54个集群，最小可检测效应随集群数增加陡降。","可将保形风险控制扩展到其他未建模扰动下的农业决策安全保证与审计设计。","2026-10-22T11:21:10.283561Z"]