[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-2904":3,"related-2904":83},{"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":23,"tags":25,"primary_topic":31,"secondary_topics":37,"classification":38,"search_phrases":68,"slug":71,"view_count":59,"doi":8,"paper":72,"created_at":82},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%。",null,"MDPI Agriculture","2026-09-17T00:00:00Z","论文",10,false,78,{"impact":16,"substance":17,"depth":18,"authority":19,"freshness":20,"relevant":21,"comment":22},16,22,18,13,9,1,"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。",[24],{"name":9,"url":6},[26,27,28,29,30],"智慧农业","农业人工智能","设施农业","番茄","病害识别",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},"B1","vegetables","蔬菜","\u002Ftopic\u002Fhorticulture\u002Fvegetables","B",[],{"status":39,"locked":13,"version":40,"method":41,"evidence":42},"classified","2026-10-22.2","rules",{"candidates":43,"agricultural_evidence":46},[44,50,57,63],{"code":32,"score":45,"supported":46,"title":47,"abstract":48,"tags":49},12,true,[29],[29],[29],{"code":51,"score":52,"supported":46,"title":53,"abstract":54,"tags":56},"B7",3,[],[55],"温室",[],{"code":58,"score":59,"supported":13,"title":60,"abstract":61,"tags":62},"I2",2,[],[],[27,30],{"code":64,"score":21,"supported":13,"title":65,"abstract":66,"tags":67},"I1",[],[],[26],[69,70],"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":73,"venue":8,"cited_by_count":74,"oa_url":8,"card":75,"direction":79,"ingested_from":81},[],0,{"tldr":76,"method":77,"finding":78,"direction":79,"opportunity":80},"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%。","农业人工智能与决策模型","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。","agent","2026-09-19T00:06:09.021594Z",{"total":84,"page":21,"page_size":84,"items":85},6,[86,143,188,262,317,362],{"id":87,"title":88,"url":89,"summary":90,"summary_zh":8,"content":8,"source_name":91,"source_url":8,"published_at":92,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":93,"score_detail":94,"sources":97,"tags":99,"primary_topic":101,"secondary_topics":102,"classification":103,"search_phrases":130,"slug":133,"view_count":59,"doi":134,"paper":135,"created_at":142},5301,"MambaSeed:基于Mamba的轻量级混合视觉网络用于番茄叶病识别(PLOS ONE)","https:\u002F\u002Fjournals.plos.org\u002Fplosone\u002Farticle?id=10.1371\u002Fjournal.pone.0359350","PLOS ONE于2026-10-15发表论文MambaSeed,由Wang X, Zhang H, Chen Y, Li Y, Qin Y, Lv Z等团队提出基于MambaVisionMixer的轻量级混合视觉网络,配套局部连续分支注入(LCBI)机制;在番茄、玉米、马铃薯数据集上识别准确率分别达98.97%、98.06%和95.87%,模型仅1.13M参数、0.240G MACs,在RTX 4090上延迟3.39ms\u002F图像,为资源受限农田环境提供紧凑、高效、鲁棒的病害识别方案。","PLOS ONE(2026-10-15)","2026-10-15T00:00:00Z",79,{"impact":18,"substance":17,"depth":18,"authority":19,"freshness":95,"relevant":21,"comment":96},8,"方法新颖、指标扎实的轻量级病害识别论文，对资源受限农田场景有实用价值，值得进入每日精选。",[98],{"name":91,"url":89},[26,27,29,30,100],"轻量级模型",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[],{"status":39,"locked":13,"version":40,"method":41,"evidence":104},{"candidates":105,"agricultural_evidence":46},[106,110,116,122,126],{"code":32,"score":45,"supported":46,"title":107,"abstract":108,"tags":109},[29],[29],[29],{"code":111,"score":52,"supported":46,"title":112,"abstract":113,"tags":115},"A3",[],[114],"玉米",[],{"code":117,"score":52,"supported":46,"title":118,"abstract":119,"tags":121},"A5",[],[120],"马铃薯",[],{"code":58,"score":59,"supported":13,"title":123,"abstract":124,"tags":125},[],[],[27,30],{"code":64,"score":21,"supported":13,"title":127,"abstract":128,"tags":129},[],[],[26],[131,132],"MambaSeed 番茄叶病","MambaVisionMixer 病害识别","MambaSeed番茄叶病-5301","10.1371\u002Fjournal.pone.0359350",{"doi":134,"openalex_id":8,"authors":136,"venue":8,"cited_by_count":74,"oa_url":8,"card":137,"direction":79,"ingested_from":81},[],{"tldr":138,"method":139,"finding":140,"direction":79,"opportunity":141},"提出轻量级混合视觉网络MambaSeed，实现番茄等作物叶病高效识别。","基于MambaVisionMixer与局部连续分支注入机制，在三种作物病害数据集","番茄、玉米、马铃薯识别准确率达98.97%、98.06%、95.87%，仅1.13M参数、3.39m","可探索该轻量模型在移动端与边缘设备的田间部署，及多作物跨域泛化与病害早期识别。","2026-10-19T00:07:48.589614Z",{"id":144,"title":145,"url":146,"summary":147,"summary_zh":8,"content":148,"source_name":149,"source_url":8,"published_at":150,"category":151,"cover_url":8,"hotness":12,"is_selected":13,"score":14,"score_detail":152,"sources":156,"tags":158,"primary_topic":161,"secondary_topics":162,"classification":163,"search_phrases":182,"slug":185,"view_count":186,"doi":8,"paper":8,"created_at":187},3751,"设施技术国产化让每公斤番茄综合生产成本降至5元","https:\u002F\u002Fnync.yn.gov.cn\u002Fhtml\u002F2026\u002Fyunnongkuanxun-new_0923\u002F1428950.html","云南首个AI+农业全链路示范基地位于昆明市宜良县匡远街道温泉社区。嘉措（云南）农业科技有限公司联合云南指月科技有限公司打造国产化AI种植模式，通过核心算法本地化适配与全链条国产化硬件深度协同，基地里每公斤番茄综合生产成本降至5元，仅为海外高端温室模式的三分之一。","位于昆明市宜良县匡远街道温泉社区的番茄示范基地即将在国庆假期进入采收期。记者近日来到该基地，探访这座云南首个“AI+农业全链路”示范基地。\n\n清晨，基地里一串串番茄在简易薄膜温室内泛着诱人的青红光泽。没有泥泞的土地，没有汗流浃背的弯腰劳作，眼前这座“番茄工厂”安静得只听见蜜蜂授粉时的嗡鸣声和水肥滴灌的微响。\n\n上午8时，有工人陆续进场，偌大的基地只需三五位工人就能打理完毕。物联网感知终端静静矗立，实时采集光照强度、空气温湿度、水肥浓度和作物长势。种苗定植、枝蔓生长、开花坐果到成熟采收，每一个环节都有“数字管家”全程值守。嘉措（云南）农业科技有限公司农场场长李良坐在棚外，掏出手机轻点屏幕，棚内温湿度曲线、土壤含水量、水肥配比参数一目了然。指尖滑动间，灌溉阀自动开启，水肥顺着滴灌带精准送达番茄根部。\n\n“过去种番茄靠经验，现在靠算法。”云南指月科技有限公司项目经理尹文鑫指着棚内的国产化智慧灌溉设备说。这套系统由嘉措（云南）农业科技有限公司联合云南指月科技有限公司打造，创新构建“全域监测+AI算法+智慧灌溉”三位一体的数字化种植模式。\n\n“基地里的全套设备都是国产的。”尹文鑫提高音量信心十足地介绍。长期以来，国内高端设施农业赛道被海外技术路线主导。参照国外植物工厂及高端玻璃温室模式，依托全套进口水肥设备与闭环控制系统虽能实现稳定高产，但每公斤番茄综合生产成本高达13元至15元，加之进口设备造价高、配件周期长、操作门槛高等痛点，技术落地难度极大。而该基地探索的国产化AI种植模式，正在彻底改写这一成本结构。\n\n国产化的意义，最终落在基地每一颗番茄的身价上。通过核心算法本地化适配与全链条国产化硬件深度协同，基地里每公斤番茄综合生产成本降至5元，仅为海外高端温室模式的三分之一。同时，国产化硬件的使用，进一步压缩了采购、部署与运维成本，凸显整套方案规模化落地的性价比。\n\n“云”上管棚，“数”里种田。从传感器、灌溉阀到核心控制系统，基地核心硬件已实现全部国产配套；从数据采集、模型训练到智能决策输出，AI技术真正扎根田间。对种植户而言，AI不再是遥远的科技概念，而是手机里能调水肥、棚里能提产量、年底能多增收的实用“新农具”。（记者：王淑娟）","云南省农业农村厅","2026-09-23T12:00:00Z","报道",{"impact":17,"substance":17,"depth":153,"authority":154,"freshness":84,"relevant":21,"comment":155},17,11,"云南首个AI+农业全链路示范基地，以国产化硬件与算法将番茄综合成本从13-15元\u002F公斤降至5元，数据具体、多方信源，具备行业示范价值。",[157],{"name":149,"url":146},[26,27,28,159,29,160],"云南","国产化替代",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[],{"status":39,"locked":13,"version":40,"method":41,"evidence":164},{"candidates":165,"agricultural_evidence":46},[166,170,174,178],{"code":32,"score":45,"supported":46,"title":167,"abstract":168,"tags":169},[29],[29],[29],{"code":51,"score":52,"supported":46,"title":171,"abstract":172,"tags":173},[],[55],[],{"code":64,"score":21,"supported":13,"title":175,"abstract":176,"tags":177},[],[],[26],{"code":58,"score":21,"supported":13,"title":179,"abstract":180,"tags":181},[],[],[27],[183,184],"云南 宜良 AI番茄 示范基地","嘉措农业 指月科技 智慧灌溉","云南宜良AI番茄示范基地-3751",4,"2026-09-29T00:08:25.735852Z",{"id":189,"title":190,"url":191,"summary":192,"summary_zh":193,"content":8,"source_name":194,"source_url":191,"published_at":195,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":196,"score_detail":197,"sources":200,"tags":202,"primary_topic":205,"secondary_topics":206,"classification":212,"search_phrases":231,"slug":234,"view_count":21,"doi":235,"paper":236,"created_at":261},4494,"Tomato ripeness classification with weight-based and texture-adaptive pooling in CNNs","https:\u002F\u002Fdoi.org\u002F10.64960\u002Feasr.2026.266223","This research introduces a weight-based mixed pooling and adaptive fuzzy pooling mechanism based on feature map texture analysis for attention mechanism Squeeze-and-Excitation Networks (SE) and Convolutional Neural Networks (CNNs) in tomato ripeness classification. Traditional CNNs and SE networks rely on deterministic pooling strategies that can be suboptimal for capturing diverse visual features such as texture and color gradients in fruit. A novel fuzzy pooling system is proposed that dynamically adjusts the fusion ratio of min and max pooling based on real-time texture analysis of feature maps via a Laplacian filter. This mechanism was integrated into Darknet53, ResNet34, and ResNet101 architectures, both directly and within the SE attention block. The method was evaluated on a five-class tomato ripeness dataset using 10-fold cross-validation against traditional and fixed-weight pooling methods. The fuzzy and mixed pooling approach demonstrated superior performance on certain models which include ESEnet: Darknet53 + SE (Fuzzy pool) achieved an average F1 score across folds of 0.8429, outperforming Darknet53 + baseline SE (Average pool) at 0.8350; Extended Net: ResNet34 + Fuzzy pool achieved an average F1 score across folds of 0.8305, outperforming baseline ResNet34 at 0.8130 and ResNet101 + Fuzzy pool achieved an average F1 score across folds of 0.8284, outperforming baseline ResNet101 at 0.8267. This improvement was realized with minimal increase in computational overhead. Specifically, measurements of model parameters (M) and FLOPs (GMACs) showed only a slight increase across all architectures after integrating SEnet into the system, and zero increase after integrating the Laplacian filter and fuzzy logic. The study concludes that logic-based adaptive pooling is a robust strategy that improves feature representation in CNNs, achieving a balance between classification accuracy and computational cost. These findings offer practical value for developing more reliable automated harvesting systems and provide a generalizable enhancement for a wide range of computer vision classification tasks.","本研究提出了一种基于特征图纹理分析的权重混合池化与自适应模糊池化机制，用于番茄成熟度分类中的注意力机制挤压激励网络（Squeeze-and-Excitation Networks, SE）和卷积神经网络（Convolutional Neural Networks, CNNs）。传统的CNN和SE网络依赖确定性池化策略，在捕捉果实纹理和颜色梯度等多样化视觉特征时可能并非最优。本文提出了一种新颖的模糊池化系统，通过拉普拉斯滤波器对特征图进行实时纹理分析，动态调整最小池化和最大池化的融合比例。该机制被集成到Darknet53、ResNet34和ResNet101架构中，既可直接集成，也可嵌入SE注意力模块内。该方法在五类番茄成熟度数据集上采用10折交叉验证，与传统池化及固定权重池化方法进行了对比评估。模糊池化和混合池化方法在部分模型上表现出优越性能，包括：ESEnet：Darknet53 + SE（模糊池化）在各折上的平均F1得分为0.8429，优于Darknet53 + 基线SE（平均池化）的0.8350；扩展网络：ResNet34 + 模糊池化在各折上的平均F1得分为0.8305，优于基线ResNet34的0.8130；ResNet101 + 模糊池化在各折上的平均F1得分为0.8284，优于基线ResNet101的0.8267。这一改进仅带来了极小的计算开销增加。具体而言，模型参数量（M）和浮点运算次数（GMACs）的测量结果表明，将SEnet集成到系统后所有架构仅有轻微增加，而集成拉普拉斯滤波器和模糊逻辑后则无任何增加。研究结论认为，基于逻辑的自适应池化是一种鲁棒策略，能够改善CNN中的特征表示，在分类精度和计算成本之间实现平衡。这些发现为开发更可靠的自动化收获系统提供了实用价值，并为广泛的计算机视觉分类任务提供了一种可推广的增强方法。","Engineering and Applied Science Research","2026-10-07T00:00:00Z",66,{"impact":95,"substance":198,"depth":153,"authority":45,"freshness":20,"relevant":21,"comment":199},20,"提出基于纹理自适应模糊池化的CNN改进方法，在番茄成熟度分类上取得小幅但稳定的F1提升且算力开销几乎不增，对自动采摘与农业视觉识别有参考价值，但属细分方法类论文，影响面有限。",[201],{"name":194,"url":191},[26,27,29,203,204],"图像识别","自动采摘",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[207],{"code":58,"slug":208,"name":209,"path":210,"parent_code":211},"agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai","I",{"status":39,"locked":13,"version":40,"method":41,"evidence":213},{"candidates":214,"agricultural_evidence":46},[215,221,227],{"code":32,"score":216,"supported":46,"title":217,"abstract":219,"tags":220},15,[218],"tomato",[29,218],[29],{"code":58,"score":95,"supported":46,"title":222,"abstract":223,"tags":226},[],[224,225],"计算机视觉","computer vision",[27,203],{"code":64,"score":21,"supported":13,"title":228,"abstract":229,"tags":230},[],[],[26],[232,233],"番茄 成熟度 分类 卷积神经网络","Darknet53 ResNet 番茄 成熟度","番茄成熟度分类卷积神经网络-4494","10.64960\u002Feasr.2026.266223",{"doi":235,"openalex_id":237,"authors":238,"venue":194,"cited_by_count":74,"oa_url":253,"card":254,"direction":259,"ingested_from":260},"W7220840158",[239,241,243,245,248,250],{"name":240,"orcid":8},"Rut Yatigul",{"name":242,"orcid":8},"Wisanu Jitviriya",{"name":244,"orcid":8},"Eiji Hayashi",{"name":246,"orcid":247},"Aran Blattler","https:\u002F\u002Forcid.org\u002F0000-0002-8684-6239",{"name":249,"orcid":8},"Watcharin Tangsuksant",{"name":251,"orcid":252},"Teppakorn Sittiwanchai","https:\u002F\u002Forcid.org\u002F0000-0001-9653-8750","https:\u002F\u002Fph01.tci-thaijo.org\u002Findex.php\u002Feasr\u002Farticle\u002Fdownload\u002F266223\u002F176569",{"tldr":255,"method":256,"finding":257,"direction":79,"opportunity":258},"提出基于纹理自适应模糊池化机制，提升CNN番茄成熟度分类性能。","用Laplacian滤波分析特征图纹理，动态调整最大最小池化融合比，集成到Dar","模糊池化在多个模型上F1提升，且计算开销几乎不增加。","可探索自适应池化在其它果蔬成熟度或田间复杂光照下的泛化能力。","智慧农业 \u002F 农业物联网","openalex","2026-10-09T23:30:09.993787Z",{"id":263,"title":264,"url":265,"summary":266,"summary_zh":267,"content":8,"source_name":268,"source_url":265,"published_at":269,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":270,"score_detail":271,"sources":273,"tags":275,"primary_topic":278,"secondary_topics":279,"classification":280,"search_phrases":298,"slug":301,"view_count":21,"doi":302,"paper":303,"created_at":316},4042,"Comparative Analysis of Decision-Level Ensemble Voting Strategies for Tomato Leaf Disease Classification Using EfficientNet-B0 and MobileNetV2","https:\u002F\u002Fdoi.org\u002F10.15294\u002Frji.v4i2.61961","Abstract. Tomato leaf diseases are a major threat to agricultural productivity, making accurate and timely disease classification essential for supporting precision agriculture. Recent advances in convolutional neural networks have significantly improved automated plant disease recognition, while ensemble learning has further enhanced classification performance by combining predictions from multiple models. Among various ensemble approaches, decision-level voting strategies, including hard voting, soft voting, and weighted soft voting, are widely adopted because of their simplicity and effectiveness. However, systematic comparisons of these voting strategies, particularly when combining lightweight CNN architectures for tomato leaf disease classification, remain limited. Therefore, this study investigates the comparative performance of different decision-level voting strategies to provide empirical evidence for selecting the most appropriate ensemble approach. Purpose: This study aims to systematically compare hard voting, soft voting, and weighted soft voting for combining EfficientNet-B0 and MobileNetV2 to identify the most effective decision-level ensemble strategy for tomato leaf disease classification. Methods\u002FStudy design\u002Fapproach: EfficientNet-B0 and MobileNetV2 were trained independently and combined using hard voting with confidence-based tie breaking, soft voting, and weighted soft voting, with voting weights optimized through grid search. Performance was evaluated using accuracy, macro-F1 score, confusion matrices, and McNemar's test. Result\u002FFindings: EfficientNet-B0 outperformed MobileNetV2 as the best single model (94.05% accuracy, 0.9266 macro-F1 vs. 90.64% accuracy, 0.8781 macro-F1). All three ensemble strategies further improved classification performance, with weighted soft voting achieving the highest performance (95.15% accuracy, 0.9417 macro-F1). McNemar's test showed that only soft voting and weighted soft voting significantly outperformed the best single model, whereas hard voting did not. No statistically significant differences were observed among the three ensemble voting strategies. Novelty: This study systematically compares hard voting, soft voting, and weighted soft voting for decision-level ensemble using EfficientNet-B0 and MobileNetV2 in tomato leaf disease classification. In addition to evaluating predictive performance, the study incorporates McNemar's statistical test to determine whether the observed performance differences among the ensemble strategies are statistically significant, providing empirical guidance for selecting appropriate voting strategies.","摘要。番茄叶片病害是农业生产力的主要威胁，因此准确、及时地进行病害分类对于支持精准农业至关重要。近年来，卷积神经网络的进展显著提升了植物病害的自动识别能力，而集成学习通过结合多个模型的预测结果进一步提高了分类性能。在各种集成方法中，决策级投票策略，包括硬投票、软投票和加权软投票，因其简单有效而被广泛采用。然而，针对这些投票策略的系统比较，尤其是在结合轻量级CNN架构进行番茄叶片病害分类时，仍然有限。因此，本研究考察了不同决策级投票策略的比较性能，为选择最合适的集成方法提供实证依据。目的：本研究旨在系统比较硬投票、软投票和加权软投票在结合EfficientNet-B0和MobileNetV2时的表现，以确定番茄叶片病害分类中最有效的决策级集成策略。方法\u002F研究设计\u002F研究路径：EfficientNet-B0和MobileNetV2分别独立训练，并通过基于置信度的平局打破硬投票、软投票和加权软投票进行结合，其中投票权重通过网格搜索进行优化。采用准确率、宏F1分数、混淆矩阵和McNemar检验评估性能。结果\u002F发现：EfficientNet-B0作为最佳单一模型优于MobileNetV2（准确率94.05%、宏F1分数0.9266，对比准确率90.64%、宏F1分数0.8781）。三种集成策略均进一步提升了分类性能，其中加权软投票表现最佳（准确率95.15%、宏F1分数0.9417）。McNemar检验表明，只有软投票和加权软投票显著优于最佳单一模型，而硬投票则没有。三种集成投票策略之间未观察到统计学显著差异。创新性：本研究系统比较了硬投票、软投票和加权软投票在番茄叶片病害分类中使用EfficientNet-B0和MobileNetV2进行决策级集成的表现。除评估预测性能外，本研究还引入McNemar统计检验，以确定所观察到的性能差异","Recursive Journal of Informatics","2026-09-30T00:00:00Z",62,{"impact":95,"substance":198,"depth":16,"authority":12,"freshness":95,"relevant":21,"comment":272},"方法对比扎实、含统计检验与量化结论，但属细分模型研究，产业影响有限，适合作为技术类精选。",[274],{"name":268,"url":265},[26,27,276,277,30],"集成学习","番茄种植",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[],{"status":39,"locked":13,"version":40,"method":41,"evidence":281},{"candidates":282,"agricultural_evidence":46},[283,287,294],{"code":32,"score":216,"supported":46,"title":284,"abstract":285,"tags":286},[218],[29,218],[277],{"code":64,"score":288,"supported":46,"title":289,"abstract":290,"tags":293},7,[],[291,292],"精准农业","precision agriculture",[26],{"code":58,"score":59,"supported":13,"title":295,"abstract":296,"tags":297},[],[],[27,30],[299,300],"EfficientNet-B0 MobileNetV2 番茄叶病","番茄叶病 集成投票 分类","EfficientNet-B0MobileNetV2番茄叶病-4042","10.15294\u002Frji.v4i2.61961",{"doi":302,"openalex_id":304,"authors":305,"venue":268,"cited_by_count":74,"oa_url":8,"card":311,"direction":79,"ingested_from":260},"W7215044289",[306,308],{"name":307,"orcid":8},"Hani Nafis Muzaki",{"name":309,"orcid":310},"Yahya Nur Ifriza","https:\u002F\u002Forcid.org\u002F0000-0003-1109-1959",{"tldr":312,"method":313,"finding":314,"direction":79,"opportunity":315},"系统比较硬投票、软投票与加权软投票三种决策级集成策略，用于番茄叶片病害分类。","EfficientNet-B0与MobileNetV2独立训练后集成，用网格搜索","加权软投票最优（准确率95.15%，宏F1 0.9417），仅软投票与加权软投票显著优于最佳单模型。","可探索更多轻量模型组合与动态权重策略，并在真实田间复杂背景下验证集成方法的鲁棒性。","2026-10-02T23:30:51.207730Z",{"id":318,"title":319,"url":320,"summary":321,"summary_zh":8,"content":8,"source_name":322,"source_url":8,"published_at":323,"category":151,"cover_url":8,"hotness":12,"is_selected":13,"score":324,"score_detail":325,"sources":327,"tags":329,"primary_topic":332,"secondary_topics":336,"classification":338,"search_phrases":357,"slug":360,"view_count":52,"doi":8,"paper":8,"created_at":361},3935,"知识与数据双驱动赋能温室番茄生长精准预测——中国农科院环发所研究成果发表","https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Fkyhd\u002F613bca00d9e84cfeb977251a606af346.htm","近日，中国农业科学院农业环境与可持续发展研究所设施植物环境工程创新团队构建了知识与数据双驱动模型，实现温室番茄生长的高精度无损预测，研究成果发表在《农业计算机与电子学（Computers and Electronics in Agriculture）》上。科研团队提出一种尺度自适应时序卷积混合模型，根据番茄不同生育期动态调整时序特征提取尺度，结合机理模型合成数据预训练与生理阈值约束，实现番茄不同器官干物质及叶面积指数的精准预测。","中国农业科学院·2026年10月","2026-10-01T00:00:00Z",82,{"impact":18,"substance":17,"depth":18,"authority":216,"freshness":20,"relevant":21,"comment":326},"国家级科研机构在核心期刊发表的知识与数据双驱动温室番茄生长预测模型，方法新颖、结论可靠，对智慧农业与设施种植有实质参考价值。",[328],{"name":322,"url":320},[26,27,28,330,331],"温室番茄","生长预测",{"code":51,"slug":333,"name":334,"path":335,"parent_code":36},"protected-horticulture","设施园艺与植物工厂","\u002Ftopic\u002Fhorticulture\u002Fprotected-horticulture",[337],{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},{"status":39,"locked":13,"version":40,"method":41,"evidence":339},{"candidates":340,"agricultural_evidence":46},[341,345,349,353],{"code":51,"score":45,"supported":46,"title":342,"abstract":343,"tags":344},[55],[55],[330],{"code":32,"score":154,"supported":46,"title":346,"abstract":347,"tags":348},[29],[29],[],{"code":64,"score":21,"supported":13,"title":350,"abstract":351,"tags":352},[],[],[26],{"code":58,"score":21,"supported":13,"title":354,"abstract":355,"tags":356},[],[],[27],[358,359],"中国农科院环发所 温室番茄 生长预测","知识与数据双驱动 番茄 模型","中国农科院环发所温室番茄生长预测-3935","2026-10-02T11:53:11.319506Z",{"id":363,"title":364,"url":365,"summary":366,"summary_zh":367,"content":8,"source_name":368,"source_url":365,"published_at":369,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":370,"score_detail":371,"sources":374,"tags":376,"primary_topic":379,"secondary_topics":380,"classification":381,"search_phrases":398,"slug":401,"view_count":84,"doi":402,"paper":403,"created_at":425},3606,"A decision-making method for light regulation of cucumber seedlings considering changes of temperature and CO2 in protected agriculture","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.biosystemseng.2026.104601","Light serves as the primary energy source for photosynthesis and significantly influences plant morphology and biomass accumulation. In protected agricultural systems, light environment parameters (intensity and spectral quality) critically determine crop productivity. A data-driven framework for light optimisation in cucumber seedlings was developed. First, an artificial neural network was established to predict net photosynthetic rate using empirical data spanning diverse environmental regimes. Then, to jointly maximise photosynthetic efficiency and minimise energy consumption, two key innovations were implemented. The U-chord method was employed to derive target light intensities, and a hybrid cubic spline-global Newton algorithm was designed for spectral quality optimisation. These subsystems were fused through support vector regression to construct adaptive decision-making models for real-time light management. The results showed that the artificial neural network model achieved exceptional Pn prediction accuracy with coefficient of determination > 0.95, root mean square error \u003C 1.6 μmol m −2 s −1 , and mean absolute error ≤ 1.2 μmol m −2 s −1 . The target light intensity points effectively delineated the light-limited phase of the Pn–PPFD response from the light-saturated phase. Final decision-making models showed high generalisability (coefficient of determination > 0.97 for both spectral and intensity decision-making models). Simulations quantified the advantages of the proposed strategy. Compared to conventional maximum Pn-oriented strategies, the proposed method reduced lighting energy consumption by 40-46%, while dynamic spectral adjustments enhanced photosynthetic efficiency by 5% relative to static controls. This study effectively improved the efficiency of light energy utilisation, and provided a theoretical method for light regulation in protected agriculture.","光作为光合作用的主要能量来源，显著影响植物形态和生物量积累。在设施农业系统中，光环境参数（光强与光谱质量）对作物生产力具有决定性作用。本研究开发了一种数据驱动的黄瓜幼苗光优化框架。首先，基于涵盖多种环境条件的经验数据，建立了人工神经网络（artificial neural network）以预测净光合速率。随后，为实现光合效率最大化与能耗最小化的协同优化，实施了两项关键创新：采用U弦法（U-chord method）确定目标光强，并设计了三次样条-全局牛顿混合算法（hybrid cubic spline-global Newton algorithm）用于光谱质量优化。通过支持向量回归（support vector regression）融合各子系统，构建了用于实时光管理的自适应决策模型。结果表明，人工神经网络模型实现了优异的净光合速率预测精度，决定系数>0.95，均方根误差\u003C1.6 μmol m⁻² s⁻¹，平均绝对误差≤1.2 μmol m⁻² s⁻¹。目标光强点有效划分了净光合速率-光合光子通量密度响应曲线中的光限制阶段与光饱和阶段。最终决策模型表现出良好的泛化能力（光谱决策模型与光强决策模型的决定系数均>0.97）。仿真量化了所提策略的优势。与传统最大净光合速率导向策略相比，该方法将补光能耗降低了40%~46%，同时动态光谱调节使光合效率较静态对照提高了5%。本研究有效提升了光能利用效率，为设施农业光调控提供了理论方法。","Biosystems Engineering","2026-09-26T00:00:00Z",81,{"impact":18,"substance":17,"depth":18,"authority":372,"freshness":20,"relevant":21,"comment":373},14,"该研究提出数据驱动的设施黄瓜育苗光调控决策方法，节能40-46%且提升光合效率，方法新颖、数据可靠，对智慧农业光环境管理有较高参考价值。",[375],{"name":368,"url":365},[26,27,28,377,378],"黄瓜育苗","光环境调控",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[],{"status":39,"locked":13,"version":40,"method":41,"evidence":382},{"candidates":383,"agricultural_evidence":46},[384,390,394],{"code":32,"score":372,"supported":46,"title":385,"abstract":387,"tags":389},[386],"cucumber",[388,386],"黄瓜",[],{"code":64,"score":21,"supported":13,"title":391,"abstract":392,"tags":393},[],[],[26],{"code":58,"score":21,"supported":13,"title":395,"abstract":396,"tags":397},[],[],[27],[399,400],"设施农业 光环境 黄瓜育苗","光合速率 光强 光谱 优化","设施农业光环境黄瓜育苗-3606","10.1016\u002Fj.biosystemseng.2026.104601",{"doi":402,"openalex_id":404,"authors":405,"venue":368,"cited_by_count":74,"oa_url":365,"card":420,"direction":79,"ingested_from":260},"W7214421951",[406,409,412,414,417],{"name":407,"orcid":408},"Pan Gao","https:\u002F\u002Forcid.org\u002F0000-0002-5184-5674",{"name":410,"orcid":411},"Huimin Li","https:\u002F\u002Forcid.org\u002F0009-0001-9124-7628",{"name":413,"orcid":8},"Jinghua Xu",{"name":415,"orcid":416},"Miao Lu","https:\u002F\u002Forcid.org\u002F0000-0001-6539-2170",{"name":418,"orcid":419},"Jin Ping Hu","https:\u002F\u002Forcid.org\u002F0000-0001-5532-6890",{"tldr":421,"method":422,"finding":423,"direction":79,"opportunity":424},"构建数据驱动光调控决策框架，优化黄瓜幼苗光强与光谱以提升光合效率并降低能耗。","人工神经网络预测净光合速率，U弦法确定光强，三次样条-全局牛顿法优化光谱，支持向","模型预测精度高，较传统策略降低能耗40-46%，动态光谱调整提升光合效率5%。","可探索多环境因子耦合的实时闭环光调控，并迁移至其他设施作物验证泛化性。","2026-09-27T23:30:09.843641Z"]