[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-1629":3,"related-1629":121},{"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":27,"primary_topic":33,"secondary_topics":39,"classification":40,"search_phrases":92,"slug":95,"view_count":96,"doi":97,"paper":98,"created_at":120},1629,"PlantC2USeg: Cross-Scale Consistent Pre-Training for Few-Shot Unified Plant Point Cloud Segmentation","https:\u002F\u002Fdoi.org\u002F10.48550\u002Farxiv.2609.02860","Modern crop breeding demands precise organ-level analysis for trait quantification, making plant point cloud segmentation (PPCS) increasingly important. However, conventional deep learning approaches rely heavily on densely annotated datasets that are labor-intensive to acquire. Unified PPCS adaptation from distribution-shifted examples with minimal additional training remains challenging. To address this, we propose PlantC2USeg, a deep transfer learning framework featuring cross-scale consistency learning to explicitly align features across spatial scales and an information-restricted decoding strategy that prevents reconstruction shortcuts and promotes robust adaptation. The resulting pre-training enables stable few-shot generalization across species and sensing conditions, while unified fine-tuning with inherited thresholds further reduces adaptation overhead. Under full supervision on Soybean3D, PlantC2USeg achieves the highest semantic IoU and instance mWCov among compared methods, at 91.91% and 94.62%. With 20 labeled samples, it leads both metrics at 89.78% and 90.27%; with only 10 samples, it retains the highest mWCov of 83.23% while achieving 83.19% IoU. Across HR3D, 10-shot transfer to tobacco, tomato, and sorghum averages 78.41% IoU and 79.42% mWCov, while 22-shot transfer to SYAU-Maize achieves the highest IoU and mRec at 92.75% and 93.51%. Furthermore, a leading category-averaged mIoU of 85.0% on ShapeNet Part demonstrates the framework's capability to handle diverse shape variations beyond agricultural domains. These results demonstrate that PlantC2USeg reduces overall adaptation effort under distribution shifts, enabling scalable plant phenotyping and transferable 3D representation learning beyond agriculture.","现代作物育种需要精确的器官级分析以实现性状量化，这使得植物点云分割（PPCS）日益重要。然而，传统深度学习方法严重依赖密集标注的数据集，而这类数据集的获取劳动强度大。在分布偏移的样本上进行统一PPCS适配且仅需最少额外训练，仍具挑战性。为此，我们提出PlantC2USeg，一种深度迁移学习框架，其特点在于跨尺度一致性学习以显式对齐不同空间尺度的特征，以及信息受限解码策略，该策略防止重建捷径并促进稳健适配。由此产生的预训练支持跨物种和传感条件下的稳定少样本泛化，而继承阈值的统一微调进一步降低了适配开销。在Soybean3D上的全监督条件下，PlantC2USeg在比较方法中实现了最高的语义IoU和实例mWCov，分别为91.91%和94.62%。使用20个标注样本时，其在两项指标上均领先，达到89.78%和90.27%；仅使用10个样本时，其仍保持最高的mWCov（83.23%），同时IoU达到83.19%。在HR3D上，对烟草、番茄和高粱的10样本迁移平均IoU为78.41%，mWCov为79.42%；对SYAU-Maize的22样本迁移则实现了最高的IoU和mRec，分别为92.75%和93.51%。此外，在ShapeNet Part上领先的类别平均mIoU达85.0%，表明该框架具备处理农业领域之外多样化形状变化的能力。这些结果表明，PlantC2USeg在分布偏移下减少了总体适配工作量，实现了可扩展的植物表型分析以及超越农业领域的可迁移3D表示学习。",null,"arXiv (Cornell University)","2026-09-02T00:00:00Z","论文",25,false,78,{"impact":17,"substance":18,"depth":17,"authority":19,"freshness":20,"relevant":21,"comment":22},18,22,13,7,1,"提出跨尺度一致预训练框架，显著提升植物点云分割的少样本泛化能力，对智慧农业表型分析有实质推进。",[24,25],{"name":10,"url":6},{"name":10,"url":26},"https:\u002F\u002Farxiv.org\u002Fabs\u002F2609.02860",[28,29,30,31,32],"智慧农业","农业人工智能","少样本学习","作物表型","点云分割",{"code":34,"slug":35,"name":36,"path":37,"parent_code":38},"I2","agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai","I",[],{"status":41,"locked":14,"version":42,"method":43,"evidence":44},"classified","2026-10-22.2","rules",{"candidates":45,"agricultural_evidence":47},[46,53,61,68,75,82,87],{"code":34,"score":20,"supported":47,"title":48,"abstract":49,"tags":52},true,[],[50,51],"深度学习","deep learning",[29],{"code":54,"score":55,"supported":47,"title":56,"abstract":57,"tags":60},"A5",6,[],[58,59],"高粱","sorghum",[],{"code":62,"score":55,"supported":47,"title":63,"abstract":64,"tags":67},"A6",[],[65,66],"烟草","tobacco",[],{"code":69,"score":55,"supported":47,"title":70,"abstract":71,"tags":74},"B1",[],[72,73],"番茄","tomato",[],{"code":76,"score":77,"supported":47,"title":78,"abstract":79,"tags":81},"A3",3,[],[80],"maize",[],{"code":83,"score":21,"supported":14,"title":84,"abstract":85,"tags":86},"I1",[],[],[28],{"code":88,"score":21,"supported":14,"title":89,"abstract":90,"tags":91},"I3",[],[],[31],[93,94],"农业人工智能 少样本学习 作物表型 智慧农业","农业人工智能 少样本学习","农业人工智能少样本学习作物表型智慧农业-1629",0,"10.48550\u002Farxiv.2609.02860",{"doi":97,"openalex_id":99,"authors":100,"venue":10,"cited_by_count":96,"oa_url":6,"card":112,"direction":118,"ingested_from":119},"W7207715140",[101,103,105,108,110],{"name":102,"orcid":9},"Yu Tian",{"name":104,"orcid":9},"Xintong Jiang",{"name":106,"orcid":107},"Jan Adamowski","https:\u002F\u002Forcid.org\u002F0000-0002-1242-3364",{"name":109,"orcid":9},"Shiv O. Prasher",{"name":111,"orcid":9},"Shangpeng Sun",{"tldr":113,"method":114,"finding":115,"direction":116,"opportunity":117},"提出PlantC2USeg，通过跨尺度一致性预训练实现少样本植物点云分割。","跨尺度一致性学习与信息受限解码策略，用于深度迁移学习预训练。","在多个数据集上少样本分割性能最优，且跨物种迁移有效。","农业人工智能与决策模型","可探索跨尺度一致性在更多农业点云任务中的应用，或结合自监督学习减少标注依赖。","农业遥感与作物表型","openalex","2026-09-04T23:30:23.311761Z",{"total":55,"page":21,"page_size":55,"items":122},[123,186,281,352,411,477],{"id":124,"title":125,"url":126,"summary":127,"summary_zh":128,"content":9,"source_name":129,"source_url":126,"published_at":130,"category":12,"cover_url":9,"hotness":131,"is_selected":14,"score":132,"score_detail":133,"sources":139,"tags":141,"primary_topic":143,"secondary_topics":144,"classification":145,"search_phrases":161,"slug":164,"view_count":77,"doi":165,"paper":166,"created_at":185},3963,"MSO-Det: A multi-level synergistic optimization framework for robust spike object detection in field scenes","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.compag.2026.112473","MSO-Det: A multi-level synergistic optimization framework for robust spike object detection in field scenes。Computers and Electronics in Agriculture","MSO-Det：一种用于田间场景中鲁棒脉冲目标检测的多级协同优化框架。《计算机与电子农业》","Computers and Electronics in Agriculture","2026-10-01T00:00:00Z",10,69,{"impact":134,"substance":17,"depth":135,"authority":136,"freshness":137,"relevant":21,"comment":138},12,16,14,9,"发表于农业信息领域核心期刊的穗部目标检测新方法，方法有创新但属细分技术进展，公共影响有限。",[140],{"name":129,"url":126},[28,29,142,31],"目标检测",{"code":34,"slug":35,"name":36,"path":37,"parent_code":38},[],{"status":41,"locked":14,"version":42,"method":43,"evidence":146},{"candidates":147,"agricultural_evidence":47},[148,153,157],{"code":34,"score":149,"supported":47,"title":150,"abstract":151,"tags":152},5,[],[142],[29,142],{"code":83,"score":21,"supported":14,"title":154,"abstract":155,"tags":156},[],[],[28],{"code":88,"score":21,"supported":14,"title":158,"abstract":159,"tags":160},[],[],[31],[162,163],"MSO-Det 穗部检测","田间场景 穗目标检测","MSO-Det穗部检测-3963","10.1016\u002Fj.compag.2026.112473",{"doi":165,"openalex_id":167,"authors":168,"venue":129,"cited_by_count":96,"oa_url":9,"card":180,"direction":116,"ingested_from":119},"W7215045657",[169,171,173,175,177],{"name":170,"orcid":9},"Yiqun Wang",{"name":172,"orcid":9},"Shuo Zhang",{"name":174,"orcid":9},"Xiaolong Liu",{"name":176,"orcid":9},"Ze Ji",{"name":178,"orcid":179},"Weikuan Jia","https:\u002F\u002Forcid.org\u002F0000-0001-6242-3269",{"tldr":181,"method":182,"finding":183,"direction":116,"opportunity":184},"提出MSO-Det多级协同优化框架，用于田间场景下鲁棒的穗部目标检测。","多级协同优化框架，结合田间场景图像进行穗目标检测。","该框架提升了田间穗检测的鲁棒性，但摘要未给出具体指标。","可探索多级优化在复杂田间光照与遮挡下的泛化能力及轻量化部署。","2026-10-02T23:30:01.492947Z",{"id":187,"title":188,"url":189,"summary":190,"summary_zh":191,"content":9,"source_name":192,"source_url":189,"published_at":193,"category":12,"cover_url":9,"hotness":131,"is_selected":14,"score":194,"score_detail":195,"sources":199,"tags":201,"primary_topic":203,"secondary_topics":209,"classification":211,"search_phrases":242,"slug":245,"view_count":149,"doi":246,"paper":247,"created_at":280},3165,"Explainable growth stage classification of cacao (Theobroma cacao L.) leaves and key feature visualization using vision transformer and transfer learning","https:\u002F\u002Fdoi.org\u002F10.3389\u002Ffpls.2026.1885906","Accurate and objective plant phenotyping is crucial for optimizing agricultural practices, understanding plant development, and enabling rapid responses to environmental changes. Traditional methods, often relying on visual observation, can be subjective, time-consuming, and may overlook subtle but important differences. This study demonstrates the power of combining digital imaging with deep learning to classify plant material with high accuracy, even when visual differences are minimal. We focused on differentiating between stage D (light green) and stage E (dark green) leaves of cacao ( Theobroma cacao L. ), which are visually very similar in size and overall structure. Using cleared and stained leaves of the SCA 6 genotype to highlight the venation network, we trained a Vision Transformer (ViT) model, a deep learning architecture, on image patches. At the patch level, the model achieved an overall accuracy of 97.03% on an independent test set, with a recall of 96.0% for stage D and 98.3% for stage E. At the whole-leaf level, majority voting correctly classified 14 of 15 independent test leaves (93.3%). Attention maps indicated that image regions containing the midrib and primary lateral veins contributed strongly to classification. These attention maps identify discriminative image regions, but they do not by themselves determine the biological mechanism underlying the signal. The major-vein signal may reflect developmental differences in vein-associated structure, stage-associated differences in Safranin O uptake or optical density, tissue thickness, or a combination of these factors. Because vascular anatomy, lignification, hydraulic conductance, phloem loading, and source–sink status were not directly measured, these mechanisms are treated as hypotheses requiring future anatomical, histochemical, and physiological validation. Thus, this study provides a proof-of-concept for interpretable image-based classification of stage D and stage E leaves within greenhouse-grown SCA 6 cacao. Extension to other cacao genotypes, field-grown plants, independent seasons, staining batches, stress detection, species identification, genotype discrimination, or precision-agriculture deployment will require external validation.","准确、客观的植物表型分析对于优化农业实践、理解植物发育以及快速响应环境变化至关重要。传统方法通常依赖视觉观察，可能具有主观性、耗时，并且可能忽略细微但重要的差异。本研究展示了将数字成像与深度学习相结合，即使在视觉差异极小的情况下，也能以高精度对植物材料进行分类。我们聚焦于区分可可（Theobroma cacao L.）的D期（浅绿色）和E期（深绿色）叶片，这些叶片在大小和整体结构上视觉上非常相似。利用SCA 6基因型的透明染色叶片以突出脉序网络，我们在图像块上训练了Vision Transformer（ViT）模型，一种深度学习架构。在图像块水平上，该模型在独立测试集上达到了97.03%的总体准确率，D期的召回率为96.0%，E期为98.3%。在整叶水平上，多数投票正确分类了15片独立测试叶片中的14片（93.3%）。注意力图表明，包含中脉和初级侧脉的图像区域对分类贡献显著。这些注意力图识别了具有判别力的图像区域，但它们本身并不能确定信号背后的生物学机制。主脉信号可能反映了脉相关结构的发育差异、番红O摄取或光密度的阶段相关差异、组织厚度，或这些因素的组合。由于未直接测量维管解剖结构、木质化、水力导度、韧皮部装载和源–库状态，这些机制被视为假设，需要未来的解剖学、组织化学和生理学验证。因此，本研究为温室种植的SCA 6可可中D期和E期叶片的可解释图像分类提供了概念验证。扩展到其他可可基因型、田间种植植株、独立季节、染色批次、胁迫检测、物种鉴定、基因型区分或精准农业部署将需要外部验证。","Frontiers in Plant Science","2026-09-21T00:00:00Z",71,{"impact":134,"substance":196,"depth":197,"authority":19,"freshness":137,"relevant":21,"comment":198},20,17,"方法新颖、数据可靠的可解释作物表型概念验证研究，但属实验室小样本，产业影响有限。",[200],{"name":192,"url":189},[28,29,50,202,31],"可可",{"code":204,"slug":205,"name":206,"path":207,"parent_code":208},"B3","tea-coffee","茶叶与咖啡","\u002Ftopic\u002Fhorticulture\u002Ftea-coffee","B",[210],{"code":34,"slug":35,"name":36,"path":37,"parent_code":38},{"status":41,"locked":14,"version":42,"method":43,"evidence":212},{"candidates":213,"agricultural_evidence":47},[214,220,225,232,238],{"code":204,"score":215,"supported":47,"title":216,"abstract":218,"tags":219},15,[217],"cacao",[202,217],[202],{"code":34,"score":221,"supported":47,"title":222,"abstract":223,"tags":224},8,[],[50,51],[29,50],{"code":226,"score":55,"supported":47,"title":227,"abstract":228,"tags":231},"B7",[],[229,230],"温室","greenhouse",[],{"code":83,"score":233,"supported":47,"title":234,"abstract":235,"tags":237},4,[],[236],"精准农业",[28],{"code":88,"score":21,"supported":14,"title":239,"abstract":240,"tags":241},[],[],[31],[243,244],"可可 叶片 生长阶段 分类","Vision Transformer 作物表型","可可叶片生长阶段分类-3165","10.3389\u002Ffpls.2026.1885906",{"doi":246,"openalex_id":248,"authors":249,"venue":192,"cited_by_count":96,"oa_url":189,"card":275,"direction":118,"ingested_from":119},"W7213901640",[250,252,255,257,259,261,264,267,270,272],{"name":251,"orcid":9},"Ezekiel Ahn",{"name":253,"orcid":254},"Eun-Sung Park","https:\u002F\u002Forcid.org\u002F0000-0001-6826-2865",{"name":256,"orcid":9},"Moon S. Kim",{"name":258,"orcid":9},"Hangi Kim",{"name":260,"orcid":9},"Lalit M. Kandpal",{"name":262,"orcid":263},"Sunchung Park","https:\u002F\u002Forcid.org\u002F0000-0002-7398-9476",{"name":265,"orcid":266},"Seunghyun Lim","https:\u002F\u002Forcid.org\u002F0000-0003-3023-4863",{"name":268,"orcid":269},"Lyndel W. Meinhardt","https:\u002F\u002Forcid.org\u002F0000-0001-8299-2629",{"name":271,"orcid":9},"Byoung-Kwan Cho",{"name":273,"orcid":274},"Insuck Baek","https:\u002F\u002Forcid.org\u002F0000-0003-1044-349X",{"tldr":276,"method":277,"finding":278,"direction":118,"opportunity":279},"用ViT和迁移学习对可可叶D、E期进行可解释分类，准确率达97%。","透明染色叶片图像块训练ViT，注意力图可视化关键区域。","模型准确区分D\u002FE期，中脉和主侧脉区域贡献最大。","可扩展到多基因型、田间、胁迫检测，并验证脉信号生物学机制。","2026-09-22T23:30:19.972949Z",{"id":282,"title":283,"url":284,"summary":285,"summary_zh":286,"content":9,"source_name":287,"source_url":284,"published_at":193,"category":12,"cover_url":9,"hotness":131,"is_selected":14,"score":288,"score_detail":289,"sources":291,"tags":293,"primary_topic":296,"secondary_topics":301,"classification":307,"search_phrases":329,"slug":332,"view_count":333,"doi":334,"paper":335,"created_at":351},3164,"Spatiotemporal Deep Learning for Rice Plant Height Estimation from Multi-Temporal UAV RGB Imagery","https:\u002F\u002Fdoi.org\u002F10.3390\u002Fagriculture16182034","Accurate plant height estimation is important for monitoring crop growth and supporting precision agricultural management. Manual measurements are labor-intensive, while LiDAR-based methods are expensive and require complex processing. UAV photogrammetry provides a lower-cost alternative but remains challenging in flooded rice paddies because of canopy deformation and difficulties in terrain extraction. This study proposes Rice-STNet, a spatiotemporal deep learning framework for end-to-end rice plant height estimation using multi-temporal UAV RGB imagery. Rice-STNet integrates a convolutional neural network for spatial feature extraction, Time2Vec for temporal encoding, and a gated recurrent unit network for modeling temporal dependencies across observation dates. The framework was evaluated using field data collected from rice paddies over two growing seasons. Rice-STNet achieved an R2 of 0.97, a root mean squared error of 1.97 cm, and a mean absolute error of 1.14 cm. It outperformed random forest, support vector regression, a CNN-only baseline, and a UAV photogrammetry-based point-cloud approach. In addition, the framework generated high-resolution plant height maps for field-scale analysis of spatial growth variability. These results underscore the importance of jointly modeling spatial and temporal characteristics for continuously evolving crop traits. The proposed framework offers an accurate, scalable, and non-destructive solution for large-scale crop phenotyping and precision agriculture.","准确的株高估算对于监测作物生长和支持精准农业管理具有重要意义。人工测量劳动强度大，而基于激光雷达（LiDAR）的方法成本高昂且需要复杂的处理。无人机摄影测量提供了一种成本较低的替代方案，但在淹水稻田中仍面临挑战，原因在于冠层变形和地形提取困难。本研究提出了Rice-STNet，一种时空深度学习框架，用于利用多时相无人机RGB影像进行端到端水稻株高估算。Rice-STNet集成了用于空间特征提取的卷积神经网络、用于时间编码的Time2Vec，以及用于建模观测日期之间时间依赖关系的门控循环单元网络。该框架利用两个生长季从稻田采集的田间数据进行了评估。Rice-STNet取得了R²为0.97、均方根误差为1.97 cm、平均绝对误差为1.14 cm的结果。其性能优于随机森林、支持向量回归、仅使用CNN的基线方法以及基于无人机摄影测量的点云方法。此外，该框架生成了高分辨率株高图，用于田块尺度空间生长变异性分析。这些结果凸显了联合建模空间与时间特征对于持续变化的作物性状的重要性。所提出的框架为大规模作物表型分析和精准农业提供了一种准确、可扩展且非破坏性的解决方案。","Agriculture",80,{"impact":17,"substance":18,"depth":17,"authority":19,"freshness":137,"relevant":21,"comment":290},"提出时空深度学习框架Rice-STNet，用多时相无人机RGB影像实现水稻株高高精度估算，方法新颖、数据跨两个生长季，对作物表型与精准农业有实用价值。",[292],{"name":287,"url":284},[28,29,294,295,31],"水稻","遥感",{"code":297,"slug":298,"name":294,"path":299,"parent_code":300},"A1","rice","\u002Ftopic\u002Ffield-crops\u002Frice","A",[302,303],{"code":34,"slug":35,"name":36,"path":37,"parent_code":38},{"code":88,"slug":304,"name":305,"path":306,"parent_code":38},"remote-sensing","遥感与农情监测","\u002Ftopic\u002Fdigital-agriculture\u002Fremote-sensing",{"status":41,"locked":14,"version":42,"method":43,"evidence":308},{"candidates":309,"agricultural_evidence":47},[310,315,319,324],{"code":297,"score":17,"supported":47,"title":311,"abstract":312,"tags":314},[298],[294,313,298],"稻田",[294],{"code":34,"score":215,"supported":47,"title":316,"abstract":317,"tags":318},[51],[50,51],[29],{"code":88,"score":221,"supported":47,"title":320,"abstract":321,"tags":323},[],[31,322],"crop phenotyping",[31,295],{"code":83,"score":20,"supported":47,"title":325,"abstract":326,"tags":328},[],[236,327],"precision agriculture",[28],[330,331],"无人机 RGB 水稻株高","Rice-STNet 水稻表型","无人机RGB水稻株高-3164",2,"10.3390\u002Fagriculture16182034",{"doi":334,"openalex_id":336,"authors":337,"venue":287,"cited_by_count":96,"oa_url":284,"card":346,"direction":118,"ingested_from":119},"W7213887432",[338,341,343],{"name":339,"orcid":340},"Weiguo Wang","https:\u002F\u002Forcid.org\u002F0009-0003-4028-9363",{"name":342,"orcid":9},"Noboru Noguchi",{"name":344,"orcid":345},"Liangliang Yang","https:\u002F\u002Forcid.org\u002F0000-0002-5055-3987",{"tldr":347,"method":348,"finding":349,"direction":118,"opportunity":350},"提出Rice-STNet时空深度学习框架，用多时相无人机RGB影像估算水稻株高。","CNN提取空间特征，Time2Vec编码时间，GRU建模时序依赖，两季稻田数据验","R²达0.97、RMSE 1.97cm，优于随机森林、SVR、纯CNN及点云方法。","可迁移至其他作物与多源遥感融合，探索轻量化模型及实时田间部署。","2026-09-22T23:30:18.545958Z",{"id":353,"title":354,"url":355,"summary":356,"summary_zh":357,"content":9,"source_name":358,"source_url":355,"published_at":359,"category":12,"cover_url":9,"hotness":131,"is_selected":14,"score":132,"score_detail":360,"sources":362,"tags":364,"primary_topic":367,"secondary_topics":368,"classification":369,"search_phrases":384,"slug":387,"view_count":333,"doi":388,"paper":389,"created_at":410},3067,"Automated detection of visible venation patterns in cowpea leaves under agricultural engineering","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.sasc.2026.200636","Accurate assessment of crop leaf characteristics is important for automated crop phenotyping and precision agriculture. However, manual assessment of leaf characteristics is time-consuming and may be affected by subjective interpretation. This study proposes an RGB-based convolutional neural network (CNN) framework for automated classification of surface-visible venation patterns in cowpea (Vigna unguiculata) leaves. A dataset of 2500 RGB cowpea leaf images, comprising Healthy and Structurally Stressed samples, was used to develop and evaluate the proposed framework. The images were preprocessed using standard image-processing operations and subsequently analyzed using a CNN for binary classification of the two visible-venation categories. Canny edge detection was additionally employed as a supplementary visualization technique to highlight surface-visible venation patterns and support qualitative interpretation; it was not used as the primary classification mechanism. The proposed CNN achieved 95.4% accuracy, 94.8% precision, 95.1% recall, and 95.0% F1-score on the evaluated test dataset. Comparative experiments with conventional feature-based approaches, including GLCM, SIFT, and HOG, showed lower observed classification performance. The results demonstrate the potential of RGB-based deep learning for automated analysis of surface-visible cowpea leaf venation patterns and provide a non-destructive image-based approach for crop phenotyping. The proposed framework can support scalable agricultural image analysis while avoiding claims of direct Surface-visible vein pattern anatomical characterization. The dataset is available at this link: “ https:\u002F\u002Fgithub.com\u002Fvijayachoudharyresearcher\u002FComputer-Vision–Cowpea-Leaves-Internal-Structure “.","准确评估作物叶片特征对于自动化作物表型分析和精准农业具有重要意义。然而，人工评估叶片特征耗时较长，且可能受到主观解释的影响。本研究提出了一种基于RGB的卷积神经网络（CNN）框架，用于对豇豆（Vigna unguiculata）叶片表面可见脉纹模式进行自动分类。研究使用包含健康和结构胁迫样本的2500张RGB豇豆叶片图像数据集来开发和评估所提出的框架。图像经标准图像处理操作预处理后，利用CNN对两个可见脉纹类别进行二分类分析。此外，研究还采用Canny边缘检测作为补充可视化技术，以突出表面可见脉纹模式并支持定性解释；其并未被用作主要分类机制。所提出的CNN在评估测试数据集上取得了95.4%的准确率、94.8%的精确率、95.1%的召回率和95.0%的F1分数。与传统基于特征的方法（包括GLCM、SIFT和HOG）的比较实验显示，其分类性能较低。结果表明，基于RGB的深度学习在自动化分析豇豆叶片表面可见脉纹模式方面具有潜力，并为作物表型分析提供了一种基于图像的非破坏性方法。所提出的框架可支持可扩展的农业图像分析，同时避免声称对表面可见脉纹模式进行直接的解剖学表征。数据集可通过以下链接获取：“https:\u002F\u002Fgithub.com\u002Fvijayachoudharyresearcher\u002FComputer-Vision–Cowpea-Leaves-Internal-Structure”。","Systems and Soft Computing","2026-09-18T00:00:00Z",{"impact":134,"substance":196,"depth":135,"authority":19,"freshness":221,"relevant":21,"comment":361},"基于RGB与CNN的豇豆叶脉自动识别研究，方法清晰、数据规模可观，对作物表型与精准农业有参考价值，但属细分领域技术进展，影响范围有限。",[363],{"name":358,"url":355},[28,29,31,365,366],"图像识别","豇豆",{"code":34,"slug":35,"name":36,"path":37,"parent_code":38},[],{"status":41,"locked":14,"version":42,"method":43,"evidence":370},{"candidates":371,"agricultural_evidence":47},[372,376,380],{"code":34,"score":221,"supported":47,"title":373,"abstract":374,"tags":375},[],[50,51],[29,365],{"code":83,"score":20,"supported":47,"title":377,"abstract":378,"tags":379},[],[236,327],[28],{"code":88,"score":20,"supported":47,"title":381,"abstract":382,"tags":383},[],[31,322],[31],[385,386],"豇豆叶片 叶脉 识别","RGB CNN 作物表型","豇豆叶片叶脉识别-3067","10.1016\u002Fj.sasc.2026.200636",{"doi":388,"openalex_id":390,"authors":391,"venue":358,"cited_by_count":96,"oa_url":355,"card":405,"direction":118,"ingested_from":119},"W7213555214",[392,395,398,400,402],{"name":393,"orcid":394},"Vijaya Choudhary","https:\u002F\u002Forcid.org\u002F0000-0003-3417-1943",{"name":396,"orcid":397},"Paramita Guha","https:\u002F\u002Forcid.org\u002F0000-0001-6696-4812",{"name":399,"orcid":9},"Jai Prakash Mishra",{"name":401,"orcid":9},"Himanshu Sharma",{"name":403,"orcid":404},"Munish Sabharwal","https:\u002F\u002Forcid.org\u002F0000-0002-7338-6982",{"tldr":406,"method":407,"finding":408,"direction":118,"opportunity":409},"用RGB卷积神经网络自动分类豇豆叶片可见脉纹，区分健康与结构胁迫样本。","2500张RGB豇豆叶图像，CNN二分类，Canny边缘检测辅助可视化，对比GL","CNN达95.4%准确率，优于传统特征方法，可实现无损作物表型分析。","可扩展至多作物、多胁迫类型及田间实时检测，并结合多光谱提升脉纹表型解析。","2026-09-21T23:30:18.575016Z",{"id":412,"title":413,"url":414,"summary":415,"summary_zh":416,"content":9,"source_name":417,"source_url":414,"published_at":418,"category":12,"cover_url":9,"hotness":131,"is_selected":14,"score":419,"score_detail":420,"sources":422,"tags":424,"primary_topic":426,"secondary_topics":427,"classification":428,"search_phrases":450,"slug":453,"view_count":333,"doi":454,"paper":455,"created_at":476},1429,"RGB-based phenotyping of grapevine leaves reveals stronger cultivar discrimination on the abaxial surface","https:\u002F\u002Fdoi.org\u002F10.1556\u002F446.2026.00335","Abstract Grapevine leaves have dorsiventral anatomy with distinct adaxial (upper) and abaxial (lower) surfaces. Although morphological descriptor lists and ampelographic literature provide information on both the upper and lower side characteristics, in practice, the color traits of the upper side have become the focus of scientific publications. This study introduces the practical application of the recently developed LeafLaminaMap software and the use of trichromatic color indices in grapevine characterization. We aimed to compare colorimetric information on the adaxial and abaxial leaf surfaces as well as to explore the potential of machine learning models in classification. Five statistical descriptors (mean, standard deviation, contrast, energy, and entropy) were calculated for 25 RGB-based color indices on both leaf surfaces of 120 samples collected from four grapevine cultivars (‘Chardonnay’, ‘Pinot noir’, ‘Sauvignon blanc’, and ‘Syrah’). Data was subjected to multivariate statistical analysis and machine learning classifiers. Results showed that the abaxial leaf surface had stronger cultivar-specific color signatures, supporting its suitability for cultivar discrimination. These findings suggest that RGB-based analysis of both adaxial and abaxial leaf surfaces has potential for grapevine cultivar discrimination, offering a new perspective for cost-efficient plant phenotyping.","葡萄叶片具有背腹型解剖结构，其近轴面（上表面）与远轴面（下表面）特征明显不同。尽管形态描述符列表和葡萄品种志文献提供了上下两面特性的信息，但在实际应用中，上表面颜色性状已成为科学研究关注的焦点。本研究介绍了新开发的LeafLaminaMap软件的实际应用以及三色颜色指数在葡萄品种鉴定中的使用。我们旨在比较叶片近轴面与远轴面的比色信息，并探索机器学习模型在分类中的潜力。对来自四个葡萄品种（‘霞多丽’、‘黑比诺’、‘长相思’和‘西拉’）的120个样本，计算了基于RGB的25个颜色指数在叶片两面的五个统计描述符（均值、标准差、对比度、能量和熵）。数据经过多元统计分析和机器学习分类器处理。结果表明，远轴面具有更强的品种特异性颜色特征，支持其适用于品种鉴别。这些发现表明，基于RGB的叶片近轴面和远轴面分析在葡萄品种鉴别方面具有潜力，为经济高效的植物表型分析提供了新视角。","Progress in Agricultural Engineering Sciences","2026-09-01T00:00:00Z",63,{"impact":134,"substance":17,"depth":135,"authority":131,"freshness":20,"relevant":21,"comment":421},"研究提出基于RGB图像的葡萄叶片双面表型分析方法，机器学习分类效果良好，为低成本作物表型鉴定提供新思路。",[423],{"name":417,"url":414},[28,29,425,31],"葡萄",{"code":34,"slug":35,"name":36,"path":37,"parent_code":38},[],{"status":41,"locked":14,"version":42,"method":43,"evidence":429},{"candidates":430,"agricultural_evidence":47},[431,437,442,446],{"code":34,"score":20,"supported":47,"title":432,"abstract":433,"tags":436},[],[434,435],"机器学习","machine learning",[29],{"code":438,"score":233,"supported":47,"title":439,"abstract":440,"tags":441},"B2",[],[425],[425],{"code":83,"score":21,"supported":14,"title":443,"abstract":444,"tags":445},[],[],[28],{"code":88,"score":21,"supported":14,"title":447,"abstract":448,"tags":449},[],[],[31],[451,452],"农业人工智能 作物表型 智慧农业 葡萄","农业人工智能 作物表型","农业人工智能作物表型智慧农业葡萄-1429","10.1556\u002F446.2026.00335",{"doi":454,"openalex_id":456,"authors":457,"venue":417,"cited_by_count":96,"oa_url":414,"card":471,"direction":118,"ingested_from":119},"W7204971124",[458,461,463,466,469],{"name":459,"orcid":460},"Péter Bodor-Pesti","https:\u002F\u002Forcid.org\u002F0000-0001-8346-4975",{"name":462,"orcid":9},"Gábor Vértes",{"name":464,"orcid":465},"Lien Le Phuong Nguyen","https:\u002F\u002Forcid.org\u002F0000-0002-5975-9906",{"name":467,"orcid":468},"László Baranyai","https:\u002F\u002Forcid.org\u002F0000-0001-6177-7364",{"name":470,"orcid":9},"Diána Ágnes Nyitrainé Sárdy",{"tldr":472,"method":473,"finding":474,"direction":118,"opportunity":475},"比较葡萄叶片上下表面颜色特征，发现下表面更能区分品种。","使用LeafLaminaMap软件和RGB颜色指数，结合统计描述符与机器学习分类","叶片下表面颜色特征品种特异性更强，更适合品种鉴别。","可探索其他作物叶片双面表型分析，或结合高光谱成像提升品种鉴别精度。","2026-09-02T23:30:47.440706Z",{"id":478,"title":479,"url":480,"summary":481,"summary_zh":482,"content":9,"source_name":483,"source_url":480,"published_at":484,"category":12,"cover_url":9,"hotness":131,"is_selected":14,"score":132,"score_detail":485,"sources":488,"tags":490,"primary_topic":494,"secondary_topics":500,"classification":502,"search_phrases":519,"slug":522,"view_count":96,"doi":523,"paper":524,"created_at":538},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":221,"substance":486,"depth":17,"authority":134,"freshness":131,"relevant":21,"comment":487},21,"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。",[489],{"name":483,"url":480},[28,29,491,492,493],"小农户","植物病害识别","可解释性",{"code":495,"slug":496,"name":497,"path":498,"parent_code":499},"G2","farm-services","社会化服务与规模经营","\u002Ftopic\u002Frural-services\u002Ffarm-services","G",[501],{"code":34,"slug":35,"name":36,"path":37,"parent_code":38},{"status":41,"locked":14,"version":42,"method":43,"evidence":503},{"candidates":504,"agricultural_evidence":47},[505,510,515],{"code":495,"score":134,"supported":47,"title":506,"abstract":508,"tags":509},[507],"smallholder",[507],[491],{"code":34,"score":131,"supported":47,"title":511,"abstract":513,"tags":514},[512],"computer vision",[],[29,492],{"code":83,"score":21,"supported":14,"title":516,"abstract":517,"tags":518},[],[],[28],[520,521],"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":523,"openalex_id":525,"authors":526,"venue":483,"cited_by_count":96,"oa_url":480,"card":533,"direction":116,"ingested_from":119},"W7223137323",[527,529,531],{"name":528,"orcid":9},"Akshay Ramgoolam",{"name":530,"orcid":9},"Priya Bundhoo Appadoo",{"name":532,"orcid":9},"Jean-Noël LEspérance",{"tldr":534,"method":535,"finding":536,"direction":116,"opportunity":537},"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。","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"]