[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-3007":3,"related-3007":104},{"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":24,"tags":26,"primary_topic":32,"secondary_topics":38,"classification":39,"search_phrases":64,"slug":67,"view_count":68,"doi":69,"paper":70,"created_at":103},3007,"Estimation of coffee leaf physiological indicators based on VAG-ViT: A lightweight framework fusing multi-view RGB images and dual-task learning","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.compag.2026.112452","Estimation of coffee leaf physiological indicators based on VAG-ViT: A lightweight framework fusing multi-view RGB images and dual-task learning。Computers and Electronics in Agriculture","基于VAG-ViT的咖啡叶片生理指标估测：一种融合多视角RGB图像与双任务学习的轻量级框架。",null,"Computers and Electronics in Agriculture","2026-09-19T00:00:00Z","论文",10,false,77,{"impact":17,"substance":18,"depth":19,"authority":20,"freshness":21,"relevant":22,"comment":23},16,21,17,14,9,1,"提出融合多视图RGB与双任务学习的轻量化ViT框架，用于咖啡叶片生理指标估算，方法新颖且面向经济作物精准管理，具备行业参考价值。",[25],{"name":10,"url":6},[27,28,29,30,31],"智慧农业","农业人工智能","轻量化模型","咖啡种植","多视图图像",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},"B3","tea-coffee","茶叶与咖啡","\u002Ftopic\u002Fhorticulture\u002Ftea-coffee","B",[],{"status":40,"locked":14,"version":41,"method":42,"evidence":43},"classified","2026-10-22.2","rules",{"candidates":44,"agricultural_evidence":47},[45,53,59],{"code":33,"score":46,"supported":47,"title":48,"abstract":50,"tags":52},15,true,[49],"coffee",[51,49],"咖啡",[30],{"code":54,"score":55,"supported":14,"title":56,"abstract":57,"tags":58},"I2",2,[],[],[28,29],{"code":60,"score":22,"supported":14,"title":61,"abstract":62,"tags":63},"I1",[],[],[27],[65,66],"咖啡叶片 生理指标 多视图RGB","VAG-ViT 双任务学习","咖啡叶片生理指标多视图RGB-3007",3,"10.1016\u002Fj.compag.2026.112452",{"doi":69,"openalex_id":71,"authors":72,"venue":10,"cited_by_count":95,"oa_url":6,"card":96,"direction":100,"ingested_from":102},"W7213674051",[73,75,77,80,82,84,86,88,90,92],{"name":74,"orcid":9},"Zongyuan Lv",{"name":76,"orcid":9},"Jianping Yang",{"name":78,"orcid":79},"Li Chen","https:\u002F\u002Forcid.org\u002F0000-0002-7006-4443",{"name":81,"orcid":9},"Rongbiao Ji",{"name":83,"orcid":9},"Yadong Li",{"name":85,"orcid":9},"Mengyao Wu",{"name":87,"orcid":9},"Songling Huang",{"name":89,"orcid":9},"Jingyu Yuan",{"name":91,"orcid":9},"Zihao Jiang",{"name":93,"orcid":94},"Tong Li","https:\u002F\u002Forcid.org\u002F0000-0002-3257-213X",0,{"tldr":97,"method":98,"finding":99,"direction":100,"opportunity":101},"提出VAG-ViT轻量框架，融合多视角RGB图像与双任务学习，估算咖啡叶片生理指标。","多视角RGB图像融合、双任务学习、轻量级ViT框架。","VAG-ViT能有效估算咖啡叶片生理指标，兼顾精度与轻量化。","农业遥感与作物表型","可探索多视角RGB与双任务学习在更多作物生理指标估算中的泛化能力及田间部署。","openalex","2026-09-20T23:30:01.826242Z",{"total":105,"page":22,"page_size":105,"items":106},6,[107,201,250,320,388,474],{"id":108,"title":109,"url":110,"summary":111,"summary_zh":112,"content":9,"source_name":113,"source_url":110,"published_at":114,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":115,"score_detail":116,"sources":121,"tags":123,"primary_topic":126,"secondary_topics":127,"classification":133,"search_phrases":165,"slug":168,"view_count":22,"doi":169,"paper":170,"created_at":200},4081,"Aerial Phenotyping of Coffee Flowering Using Deep Learning for Early Yield Prediction","https:\u002F\u002Fdoi.org\u002F10.3390\u002Fplants15193019","Early yield estimation is important for improving decision-making and management practices in coffee production systems. However, approaches that use aerial imagery to characterize flowering and relate it to subsequent yield remain limited. This study evaluated the potential of RGB imagery acquired by remotely piloted aircraft (RPA) combined with Deep Learning techniques to classify coffee plants according to three levels of a flowering proxy and to investigate its relationship with yield. A dataset comprising 1593 segmented RGB images was classified into Low, Medium, and High flowering levels using a custom convolutional neural network and three transfer learning architectures: VGG16, ResNet50V2, and MobileNetV2. ResNet50V2 was selected for final evaluation based on its validation performance and achieved 92.12% accuracy, 92.40% macro-precision, 91.81% macro-recall, and 92.06% macro-F1 on the test dataset. The flowering proxy showed a positive association with field-observed yield, with Spearman’s ρ = 0.701 (p \u003C 0.001) and R2 = 0.528 for the combined dataset of Arara and Mundo Novo. In consolidated out-of-sample predictions, the model achieved R2 = 0.489, MAE = 1.03 kg plant−1, and RMSE = 1.18 kg plant−1. Performance differed between cultivars, with consolidated R2 values of 0.567 for Arara and 0.329 for Mundo Novo. Spatial analysis also showed correspondence between the distribution of flowering levels derived from RGB imagery and the subsequent spatial distribution of yield. These findings indicate that RPA-based RGB imagery and Deep Learning can provide spatialized information related to coffee flowering and its productive performance, with potential applications in precision agriculture. Further evaluation across production cycles, cultivars, and environmental conditions is needed to assess the robustness and generalization of the approach.","早期产量估测对于改善咖啡生产系统中的决策与管理实践具有重要意义。然而，利用航空影像表征开花并将其与后续产量相关联的方法仍然有限。本研究评估了由遥控驾驶航空器（RPA）获取的RGB影像结合深度学习技术，根据开花代理指标的三个水平对咖啡植株进行分类，并探讨其与产量之间关系的潜力。利用自定义卷积神经网络和三种迁移学习架构——VGG16、ResNet50V2和MobileNetV2，将包含1593张分割RGB图像的数据集分类为低、中、高开花水平。基于验证性能，选择ResNet50V2进行最终评估，其在测试数据集上达到了92.12%的准确率、92.40%的宏精确率、91.81%的宏召回率和92.06%的宏F1值。开花代理指标与田间观测产量呈正相关，Arara和Mundo Novo合并数据集的Spearman's ρ = 0.701（p \u003C 0.001），R² = 0.528。在合并的样本外预测中，模型达到了R² = 0.489、MAE = 1.03 kg plant⁻¹和RMSE = 1.18 kg plant⁻¹。不同品种间的表现存在差异，Arara和Mundo Novo的合并R²值分别为0.567和0.329。空间分析还表明，由RGB影像得出的开花水平分布与后续产量的空间分布之间存在对应关系。这些发现表明，基于RPA的RGB影像和深度学习能够提供与咖啡开花及其生产性能相关的空间化信息，在精准农业中具有应用潜力。需要在不同生产周期、品种和环境条件下进行进一步评估，以检验该方法的稳健性和泛化能力。","Plants","2026-10-02T00:00:00Z",78,{"impact":17,"substance":117,"depth":118,"authority":119,"freshness":21,"relevant":22,"comment":120},22,18,13,"该研究将无人机RGB影像与深度学习结合用于咖啡花期分类和早期产量预测，方法新颖、数据扎实、结论可靠，对智慧农业和精准农业有参考价值，但属于细分作物研究，影响范围有限。",[122],{"name":113,"url":110},[27,28,124,125,30],"产量预测","遥感",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[128],{"code":54,"slug":129,"name":130,"path":131,"parent_code":132},"agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai","I",{"status":40,"locked":14,"version":41,"method":42,"evidence":134},{"candidates":135,"agricultural_evidence":47},[136,140,146,153,160],{"code":33,"score":46,"supported":47,"title":137,"abstract":138,"tags":139},[49],[51,49],[30],{"code":54,"score":46,"supported":47,"title":141,"abstract":143,"tags":145},[142],"deep learning",[144,142],"深度学习",[28],{"code":60,"score":147,"supported":47,"title":148,"abstract":149,"tags":152},7,[],[150,151],"精准农业","precision agriculture",[27],{"code":154,"score":105,"supported":47,"title":155,"abstract":156,"tags":159},"I4",[],[157,158],"空间分析","spatial analysis",[],{"code":161,"score":55,"supported":14,"title":162,"abstract":163,"tags":164},"I3",[],[],[124,125],[166,167],"无人机 RGB 咖啡 花期 产量预测","ResNet50V2 咖啡 深度学习","无人机RGB咖啡花期产量预测-4081","10.3390\u002Fplants15193019",{"doi":169,"openalex_id":171,"authors":172,"venue":113,"cited_by_count":95,"oa_url":110,"card":195,"direction":100,"ingested_from":102},"W7216244707",[173,175,178,181,183,186,189,192],{"name":174,"orcid":9},"Elias Horácio Zavala",{"name":176,"orcid":177},"Gabriel Araújo e Silva Ferraz","https:\u002F\u002Forcid.org\u002F0000-0001-6403-2210",{"name":179,"orcid":180},"Felipe Andrés Obando Vega","https:\u002F\u002Forcid.org\u002F0000-0001-7241-7065",{"name":182,"orcid":9},"Mirian de Lourdes Oliveira",{"name":184,"orcid":185},"Milene Alves de Figueiredo Carvalho","https:\u002F\u002Forcid.org\u002F0000-0001-6138-8182",{"name":187,"orcid":188},"Samuel Henrique Braga da Cunha","https:\u002F\u002Forcid.org\u002F0000-0002-8783-0378",{"name":190,"orcid":191},"Andrea Pagliai","https:\u002F\u002Forcid.org\u002F0000-0002-0455-7426",{"name":193,"orcid":194},"Daniele Sarri","https:\u002F\u002Forcid.org\u002F0000-0001-6990-7573",{"tldr":196,"method":197,"finding":198,"direction":100,"opportunity":199},"用无人机RGB影像和深度学习对咖啡开花分级，实现早期产量预测。","1593张分割RGB图，自定义CNN与VGG16\u002FResNet50V2\u002FMobi","ResNet50V2开花分级准确率92.12%，开花等级与产量正相关（ρ=0.701）。","可跨生产周期、品种和环境验证模型泛化性，并融合多源遥感提升早期产量预测稳健性。","2026-10-03T23:30:27.930711Z",{"id":202,"title":203,"url":204,"summary":205,"summary_zh":9,"content":206,"source_name":207,"source_url":9,"published_at":208,"category":209,"cover_url":9,"hotness":13,"is_selected":14,"score":210,"score_detail":211,"sources":215,"tags":217,"primary_topic":221,"secondary_topics":222,"classification":223,"search_phrases":245,"slug":248,"view_count":55,"doi":9,"paper":9,"created_at":249},4278,"云南保山小粒咖啡数中求精 云智农链平台综合成本降83%亩均增收1500元","https:\u002F\u002Fnews.qq.com\u002Frain\u002Fa\u002F20261004A03PSF00","中国移动云南公司保山分公司联合保山学院浙江理工大学西北工业大学等高校科研团队打造云智农链平台，集农事管理水肥智能控制病虫害预警产品溯源于一体，辛家山咖啡种植基地应用后综合成本下降超83%、鲜果亩均增产超21%、瑕疵率降至3.34%、亩均增收近1500元。2025年保山精品咖啡种植面积18.06万亩、产量2.68万吨、全产业链产值223.28亿元。","辛家山咖啡种植基地负责人杨江春介绍，基地2020年建立之初，便瞄准精品化赛道，与云南省农业科学院热带亚热带经济作物研究所等科研团队合作筛选品种，并在咖啡地里间种香蕉、木豆等，用于遮阴、固氮等。\n\n咖啡树铺满山坡后，人工管护逐渐力不从心：不同坡向、地块的水肥需求不一，何时灌溉、施肥，多靠经验判断；夜间值守时，疲劳还易带来误判。“最棘手的是病虫害。过去，一些病害往往要等到巡园时才能发现，容易错过最佳防治期，影响咖啡品质。”杨江春说。\n\n2024年，庄园接入“云智农链”平台，管护开始有了“数字帮手”。该平台除后台远程调度外，土壤墒情低于预设阈值时，系统还可自动启动灌溉施肥；AI巡检无人机在咖啡林间往返，通过图像识别发现锈病等早期病害，并生成植保建议。“两年多来，庄园综合成本下降超83%，鲜果亩均增产超21%，瑕疵率降至3.34%，亩均增收近1500元。”杨江春说。","腾讯新闻 中新网 2026-10-04","2026-10-04T00:00:00Z","报道",67,{"impact":17,"substance":212,"depth":46,"authority":213,"freshness":213,"relevant":22,"comment":214},20,8,"智慧农业平台在咖啡种植中的落地案例，成本降幅与增收数据具体，具备行业参考价值。",[216],{"name":207,"url":204},[218,27,219,30,220],"数字乡村","水肥一体化","AI巡检",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[],{"status":40,"locked":14,"version":41,"method":42,"evidence":224},{"candidates":225,"agricultural_evidence":47},[226,231,236,240],{"code":33,"score":227,"supported":47,"title":228,"abstract":229,"tags":230},12,[51],[51],[30],{"code":232,"score":22,"supported":14,"title":233,"abstract":234,"tags":235},"F3",[],[],[219],{"code":60,"score":22,"supported":14,"title":237,"abstract":238,"tags":239},[],[],[27],{"code":241,"score":22,"supported":14,"title":242,"abstract":243,"tags":244},"I6",[],[],[218],[246,247],"云智农链 保山小粒咖啡","保山 咖啡 智慧种植","云智农链保山小粒咖啡-4278","2026-10-06T00:05:03.287378Z",{"id":251,"title":252,"url":253,"summary":254,"summary_zh":255,"content":9,"source_name":256,"source_url":253,"published_at":257,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":258,"score_detail":259,"sources":261,"tags":263,"primary_topic":265,"secondary_topics":266,"classification":271,"search_phrases":294,"slug":297,"view_count":22,"doi":298,"paper":299,"created_at":319},3893,"Modeling the productive potential of coffee trees in the Cerrado of Minas Gerais through hierarchical landscape analysis and remote sensing data","https:\u002F\u002Fdoi.org\u002F10.25186\u002F.v21i.2516","Brazil stands out as the world’s largest producer and exporter of Arabica coffee, one of the world’s most important commodities, which plays a crucial role in the country’s economy. In this context, this study develops a model to classify the productive potential of coffee trees based on hierarchical landscape analysis and coffee plant metrics, applied to different farms in the Córrego Feio watershed, Patrocínio, Minas Gerais. We employed a model based on landscape metrics (slope, water flow, vegetation, and slope shape), inspired by Ross’s (1994) proposal for Environmental Fragility. Each variable was reclassified according to hierarchical levels from the literature and empirical observations of the area. The data were integrated to generate two products: Factor T, which represents the potential of the area based on static soil and relief variables, and Calculated Potential, which combines Factor T with the NDVI vegetation index for a dynamic assessment. The application of the model demonstrated a strong capacity to interpret the qualitative hierarchy of productive potential, supporting the delineation of differentiated management zones. The results indicated that productive potential is more strongly associated with soil and topographic factors than with the vegetation index. It is concluded that the proposed method is a valuable tool for precision agriculture, allowing the delineation of management zones for the application of inputs in a more efficient and sustainable manner, contributing to resource conservation and environmental protection. Key words: Arabica coffee; environmental fragility; hierarchical analysis; vegetation index; watershed.","巴西是全球最大的阿拉比卡咖啡（Arabica coffee）生产国和出口国，阿拉比卡咖啡作为全球最重要的商品之一，在该国经济中发挥着至关重要的作用。在此背景下，本研究开发了一个基于层次景观分析和咖啡植株指标的咖啡树生产潜力分类模型，并将其应用于米纳斯吉拉斯州帕特罗西尼奥市Córrego Feio流域的不同农场。我们采用了一个基于景观指标（坡度、水流、植被和坡形）的模型，其灵感来源于Ross（1994）提出的环境脆弱性方案。每个变量根据文献和该区域经验观察所得的层次等级进行重新分类。数据经过整合生成了两个产品：因子T，代表基于静态土壤和地形变量的区域潜力；以及计算潜力，将因子T与NDVI植被指数相结合以进行动态评估。模型的应用展现出强大的能力，能够解释生产潜力的定性层次，支持差异化经营区的划分。结果表明，生产潜力与土壤和地形因子的关联性强于与植被指数的关联性。研究认为，所提出的方法是一种有价值的精准农业工具，能够以更高效和可持续的方式划定管理区以施用投入品，有助于资源保护和环境保护。关键词：阿拉比卡咖啡；环境脆弱性；层次分析；植被指数；流域。","Coffee Science","2026-09-30T00:00:00Z",70,{"impact":227,"substance":212,"depth":19,"authority":119,"freshness":213,"relevant":22,"comment":260},"基于景观分析与NDVI的咖啡生产潜力分级模型，方法新颖、结论可靠，对精准农业管理分区有实用价值，但属区域性研究，影响范围有限。",[262],{"name":256,"url":253},[27,150,125,264,30],"管理分区",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[267],{"code":161,"slug":268,"name":269,"path":270,"parent_code":132},"remote-sensing","遥感与农情监测","\u002Ftopic\u002Fdigital-agriculture\u002Fremote-sensing",{"status":40,"locked":14,"version":41,"method":42,"evidence":272},{"candidates":273,"agricultural_evidence":47},[274,278,283,287],{"code":33,"score":46,"supported":47,"title":275,"abstract":276,"tags":277},[49],[51,49],[30],{"code":161,"score":21,"supported":47,"title":279,"abstract":281,"tags":282},[280],"remote sensing",[],[125],{"code":60,"score":213,"supported":47,"title":284,"abstract":285,"tags":286},[],[150,151],[27,150],{"code":288,"score":105,"supported":47,"title":289,"abstract":290,"tags":293},"H2",[],[291,292],"土壤","soil",[],[295,296],"Minas Gerais 咖啡 遥感","Córrego Feio 流域 咖啡 生产潜力","MinasGerais咖啡遥感-3893","10.25186\u002F.v21i.2516",{"doi":298,"openalex_id":300,"authors":301,"venue":256,"cited_by_count":95,"oa_url":253,"card":314,"direction":100,"ingested_from":102},"W7214888407",[302,305,308,311],{"name":303,"orcid":304},"Jorge Eduardo Ferreira Cunha","https:\u002F\u002Forcid.org\u002F0000-0002-3679-6376",{"name":306,"orcid":307},"Mateus Duarte Segismundo","https:\u002F\u002Forcid.org\u002F0009-0001-1289-406X",{"name":309,"orcid":310},"George Deroco Martins","https:\u002F\u002Forcid.org\u002F0000-0001-9738-7325",{"name":312,"orcid":313},"João Vítor Meza Bravo","https:\u002F\u002Forcid.org\u002F0000-0002-5457-3192",{"tldr":315,"method":316,"finding":317,"direction":100,"opportunity":318},"基于景观指标与NDVI构建模型，对巴西塞拉多咖啡树生产潜力进行分级分类。","层次化景观分析（坡度、水流、植被、坡形）结合NDVI，生成因子T与计算潜力。","生产潜力与土壤和地形因素关联强于植被指数，可划分差异化精准管理区。","可引入多时相遥感与机器学习，动态预测咖啡生产潜力并优化管理区划分。","2026-10-01T23:30:28.485534Z",{"id":321,"title":322,"url":323,"summary":324,"summary_zh":9,"content":9,"source_name":325,"source_url":323,"published_at":326,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":327,"score_detail":328,"sources":330,"tags":332,"primary_topic":335,"secondary_topics":336,"classification":338,"search_phrases":354,"slug":357,"view_count":358,"doi":359,"paper":360,"created_at":387},3619,"Comparison of NDVI obtained from an active proximal sensor and UAV multispectral imagery in coffee","https:\u002F\u002Fdoi.org\u002F10.21203\u002Frs.3.rs-10767247\u002Fv1","Comparison of NDVI obtained from an active proximal sensor and UAV multispectral imagery in coffee。Research Square","Research Square","2026-09-24T00:00:00Z",51,{"impact":213,"substance":20,"depth":46,"authority":105,"freshness":213,"relevant":22,"comment":329},"咖啡园NDVI主动传感器与无人机多光谱对比研究，方法有参考价值但属细分作物技术验证，影响面有限。",[331],{"name":325,"url":323},[27,333,125,334,30],"无人机","NDVI",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[337],{"code":161,"slug":268,"name":269,"path":270,"parent_code":132},{"status":40,"locked":14,"version":41,"method":42,"evidence":339},{"candidates":340,"agricultural_evidence":47},[341,346,350],{"code":161,"score":119,"supported":47,"title":342,"abstract":344,"tags":345},[343],"multispectral",[343],[334,125],{"code":33,"score":227,"supported":47,"title":347,"abstract":348,"tags":349},[49],[49],[30],{"code":60,"score":22,"supported":14,"title":351,"abstract":352,"tags":353},[],[],[27],[355,356],"咖啡 NDVI 无人机 多光谱","Research Square 咖啡 遥感","咖啡NDVI无人机多光谱-3619",5,"10.21203\u002Frs.3.rs-10767247\u002Fv1",{"doi":359,"openalex_id":361,"authors":362,"venue":325,"cited_by_count":95,"oa_url":386,"card":9,"direction":100,"ingested_from":102},"W7214220732",[363,365,367,369,371,374,376,378,381,384],{"name":364,"orcid":9},"Gabriel de Morais Campos",{"name":366,"orcid":9},"Aline Bhering Silva",{"name":368,"orcid":9},"Cileimar Aparecida da Silva",{"name":370,"orcid":9},"Vanda Maria Salles Andrade",{"name":372,"orcid":373},"Thaline M. Pimenta","https:\u002F\u002Forcid.org\u002F0000-0002-5002-177X",{"name":375,"orcid":9},"Andressa Barcellos Silva",{"name":377,"orcid":9},"Marco Thúlio Gonçalves Vieira",{"name":379,"orcid":380},"Daniel Marçal de Queiroz","https:\u002F\u002Forcid.org\u002F0000-0003-0987-3855",{"name":382,"orcid":383},"Fábio Daniel Tancredi","https:\u002F\u002Forcid.org\u002F0000-0002-7619-2200",{"name":385,"orcid":9},"Flora Maria Melo Villar","https:\u002F\u002Fwww.researchsquare.com\u002Farticle\u002Frs-10767247\u002Flatest.pdf","2026-09-27T23:30:34.698079Z",{"id":389,"title":390,"url":391,"summary":392,"summary_zh":393,"content":9,"source_name":394,"source_url":391,"published_at":395,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":396,"score_detail":397,"sources":399,"tags":401,"primary_topic":404,"secondary_topics":405,"classification":407,"search_phrases":435,"slug":438,"view_count":358,"doi":439,"paper":440,"created_at":473},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":227,"substance":212,"depth":19,"authority":119,"freshness":21,"relevant":22,"comment":398},"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。",[400],{"name":394,"url":391},[27,28,144,402,403],"可可","作物表型",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[406],{"code":54,"slug":129,"name":130,"path":131,"parent_code":132},{"status":40,"locked":14,"version":41,"method":42,"evidence":408},{"candidates":409,"agricultural_evidence":47},[410,415,419,426,431],{"code":33,"score":46,"supported":47,"title":411,"abstract":413,"tags":414},[412],"cacao",[402,412],[402],{"code":54,"score":213,"supported":47,"title":416,"abstract":417,"tags":418},[],[144,142],[28,144],{"code":420,"score":105,"supported":47,"title":421,"abstract":422,"tags":425},"B7",[],[423,424],"温室","greenhouse",[],{"code":60,"score":427,"supported":47,"title":428,"abstract":429,"tags":430},4,[],[150],[27],{"code":161,"score":22,"supported":14,"title":432,"abstract":433,"tags":434},[],[],[403],[436,437],"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":439,"openalex_id":441,"authors":442,"venue":394,"cited_by_count":95,"oa_url":391,"card":468,"direction":100,"ingested_from":102},"W7213901640",[443,445,448,450,452,454,457,460,463,465],{"name":444,"orcid":9},"Ezekiel Ahn",{"name":446,"orcid":447},"Eun-Sung Park","https:\u002F\u002Forcid.org\u002F0000-0001-6826-2865",{"name":449,"orcid":9},"Moon S. Kim",{"name":451,"orcid":9},"Hangi Kim",{"name":453,"orcid":9},"Lalit M. Kandpal",{"name":455,"orcid":456},"Sunchung Park","https:\u002F\u002Forcid.org\u002F0000-0002-7398-9476",{"name":458,"orcid":459},"Seunghyun Lim","https:\u002F\u002Forcid.org\u002F0000-0003-3023-4863",{"name":461,"orcid":462},"Lyndel W. Meinhardt","https:\u002F\u002Forcid.org\u002F0000-0001-8299-2629",{"name":464,"orcid":9},"Byoung-Kwan Cho",{"name":466,"orcid":467},"Insuck Baek","https:\u002F\u002Forcid.org\u002F0000-0003-1044-349X",{"tldr":469,"method":470,"finding":471,"direction":100,"opportunity":472},"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":475,"title":476,"url":477,"summary":478,"summary_zh":479,"content":9,"source_name":480,"source_url":477,"published_at":481,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":396,"score_detail":482,"sources":484,"tags":486,"primary_topic":489,"secondary_topics":490,"classification":491,"search_phrases":510,"slug":513,"view_count":68,"doi":514,"paper":515,"created_at":537},2955,"Precision agriculture applied to coffee cultivation in agroforestry systems","https:\u002F\u002Fdoi.org\u002F10.1007\u002Fs10457-026-01651-z","Abstract The search for more sustainable agricultural production systems is a goal of both agroforestry systems (AFS) and precision agriculture (PA). In this context, integrating PA technologies into coffee cultivation under AFS presents the potential to increase productive efficiency and sustainability. Thus, the objective of this study was to evaluate whether the variability of environmental conditions in coffee agroforestry systems can be estimated using technologies associated with PA. The studies were conducted on three coffee farms in Coimbra and Araponga, Minas Gerais, Brazil. For management zone delineation, combinations of apparent soil electrical conductivity (ECa), Normalized Difference Vegetation Index (NDVI), and Digital Terrain Model (DTM) were used. Microclimatic variability was evaluated using weather stations, considering different agroforestry management practices and altitude conditions in mountain coffee cultivation. Canopy cover was estimated by digital image processing techniques applied to aerial images obtained by an Unmanned Aerial Vehicle (UAV). The combination ECa + NDVI was efficient in delineating zones with distinct physical and chemical attributes. Agroforestry management practices and altitude influenced temperature and relative humidity. Tree canopy estimation proved viable to support management. It is concluded that PA is strategic for coffee cultivation in AFS, favoring site-specific and more sustainable management.","摘要 寻求更可持续的农业生产系统是农林业系统（AFS）和精准农业（PA）的共同目标。在此背景下，将精准农业技术整合到农林业系统下的咖啡种植中，具有提高生产效率和可持续性的潜力。因此，本研究旨在评估是否可以利用与精准农业相关的技术来估算咖啡农林业系统中环境条件的变异性。研究在巴西米纳斯吉拉斯州科英布拉和阿拉蓬加的三个咖啡农场进行。为划定管理区，采用了表观土壤电导率（ECa）、归一化差异植被指数（NDVI）和数字地形模型（DTM）的组合。利用气象站评估微气候变异性，考虑了山地咖啡种植中不同的农林业管理措施和海拔条件。通过将数字图像处理技术应用于无人机（UAV）获取的航空图像来估算冠层覆盖度。ECa + NDVI的组合能够有效划定具有不同物理和化学属性的区域。农林业管理措施和海拔影响了温度和相对湿度。树木冠层估算被证明可用于支持管理。结论是，精准农业对农林业系统下的咖啡种植具有战略意义，有利于因地制宜且更可持续的管理。","Agroforestry Systems","2026-09-18T00:00:00Z",{"impact":227,"substance":18,"depth":19,"authority":119,"freshness":213,"relevant":22,"comment":483},"巴西咖啡农林复合系统中应用精准农业技术的研究，方法新颖、结论可靠，对智慧农业与遥感应用有参考价值，但属区域性研究，影响范围有限。",[485],{"name":480,"url":477},[27,150,487,488,30],"遥感监测","农林复合系统",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[],{"status":40,"locked":14,"version":41,"method":42,"evidence":492},{"candidates":493,"agricultural_evidence":47},[494,498,502,506],{"code":60,"score":17,"supported":47,"title":495,"abstract":496,"tags":497},[151],[150,151],[27,150],{"code":33,"score":46,"supported":47,"title":499,"abstract":500,"tags":501},[49],[51,49],[30],{"code":288,"score":105,"supported":47,"title":503,"abstract":504,"tags":505},[],[291,292],[],{"code":161,"score":22,"supported":14,"title":507,"abstract":508,"tags":509},[],[],[487],[511,512],"巴西 咖啡 农林复合系统 精准农业","无人机 冠层覆盖 NDVI 咖啡","巴西咖啡农林复合系统精准农业-2955","10.1007\u002Fs10457-026-01651-z",{"doi":514,"openalex_id":516,"authors":517,"venue":480,"cited_by_count":95,"oa_url":477,"card":530,"direction":536,"ingested_from":102},"W7213541962",[518,520,523,526,527],{"name":519,"orcid":9},"Wagner Silva dos Santos",{"name":521,"orcid":522},"Francisco de Assis de Carvalho Pinto","https:\u002F\u002Forcid.org\u002F0000-0002-8279-9535",{"name":524,"orcid":525},"André Luiz de Freitas Coelho","https:\u002F\u002Forcid.org\u002F0000-0002-7595-9713",{"name":379,"orcid":380},{"name":528,"orcid":529},"Bruno Nery Fernandes Vasconcelos","https:\u002F\u002Forcid.org\u002F0000-0001-6298-9748",{"tldr":531,"method":532,"finding":533,"direction":534,"opportunity":535},"评估精准农业技术能否估算农林复合咖啡系统的环境变异，以支持可持续管理。","用ECa、NDVI、DTM划分管理区，气象站测微气候，无人机图像估树冠覆盖。","ECa+NDVI可有效划分土壤属性差异区，农林管理和海拔影响温湿度，树冠估算可行。","智慧农业 \u002F 农业物联网","可探索多源遥感与物联网融合的实时管理区动态划分，并验证其对咖啡产量与碳汇的长期效应。","数字乡村与农业信息化","2026-09-19T23:30:40.775897Z"]