[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-3199":3,"related-3199":135},{"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":25,"primary_topic":31,"secondary_topics":36,"classification":37,"search_phrases":88,"slug":91,"view_count":92,"doi":93,"paper":94,"created_at":134},3199,"Recommendation of Suitable Planting Locations for Hybrid Maize Varieties Based on Graph Collaborative Filtering","https:\u002F\u002Fdoi.org\u002F10.1093\u002Finsilicoplants\u002Fdiag026","Abstract Enhancing the precision of crop variety adaptability evaluation is critical for maximizing yields and ensuring food security under climate change. Traditional methods relying on expert knowledge or process-based models face scalability challenges and difficulties in capturing complex variety–environment interactions. This study proposes a scalable graph-based collaborative filtering framework for assessing suitable planting locations for hybrid maize varieties, reformulating adaptability evaluation as a ranking task. The framework integrates two key innovations: a multi-layer perceptron (MLP) encoding module that leverages 15 variety trait features and 24 environmental variables to address the cold-start problem, enabling recommendations for previously unseen varieties; and a BPR+ loss function that distinguishes unsuitable environments from merely unplanted ones to accommodate the unique sparsity structure of agricultural trial data. Evaluated against six state-of-the-art graph-based recommendation models on a dataset of 189 locations and 1,102 hybrid maize varieties across China's major agro-ecological zones, our method consistently outperforms existing approaches, achieving approximately 5% improvement over the strongest baseline. The MLP module attains cold-start NDCG above 0.8, and BPR+ yields significant gains in Recall@1–4 and NDCG@1–5. A case study of the widely cultivated variety ZD958, validated against both national promotion data (2019–2024) and multi-year field-trial yields from 40 locations (2017–2024), confirms that the model's rank-ordered recommendations align closely with independent yield evidence: top-ranked locations achieve both the highest productivity and the greatest yield stability. This framework provides an AI-based decision support tool for variety placement optimization and climate-resilient agriculture.","摘要 提高作物品种适应性评价的精度，对于在气候变化背景下最大化产量和保障粮食安全至关重要。依赖专家知识或过程模型的传统方法面临可扩展性挑战，且难以捕捉复杂的品种–环境互作。本研究提出了一种可扩展的基于图协同过滤框架，用于评估杂交玉米品种的适宜种植地点，将适应性评价重新表述为排序任务。该框架融合了两项关键创新：一是多层感知机（MLP）编码模块，利用15个品种性状特征和24个环境变量来缓解冷启动问题，从而能够对未见过的品种进行推荐；二是BPR+损失函数，将不适宜环境与仅未种植环境区分开来，以适应农业试验数据特有的稀疏结构。在中国主要农业生态区189个地点和1，102个杂交玉米品种的数据集上，与六种最先进的基于图的推荐模型进行对比评估，我们的方法始终优于现有方法，相较最强基线提升约5%。MLP模块的冷启动NDCG达到0.8以上，BPR+在Recall@1–4和NDCG@1–5上均带来显著提升。对广泛种植品种ZD958的案例研究，结合国家推广数据（2019–2024年）和40个地点多年田间试验产量数据（2017–2024年）进行验证，证实模型的排序推荐与独立产量证据高度一致：排名靠前的地点既具有最高生产力，也具有最大产量稳定性。该框架为品种布局优化和气候韧性农业提供了基于人工智能的决策支持工具。",null,"in silico Plants","2026-09-20T00:00:00Z","论文",10,false,79,{"impact":17,"substance":18,"depth":17,"authority":19,"freshness":20,"relevant":21,"comment":22},18,22,13,8,1,"提出基于图协同过滤的杂交玉米品种适宜种植区推荐框架，覆盖全国189个地点、1102个品种，冷启动与排序指标均有提升，对品种布局优化有实用价值。",[24],{"name":10,"url":6},[26,27,28,29,30],"智慧农业","农业人工智能","种业振兴","玉米","品种布局",{"code":32,"slug":33,"name":29,"path":34,"parent_code":35},"A3","maize","\u002Ftopic\u002Ffield-crops\u002Fmaize","A",[],{"status":38,"locked":14,"version":39,"method":40,"evidence":41},"classified","2026-10-22.2","rules",{"candidates":42,"agricultural_evidence":45},[43,49,57,64,71,78,83],{"code":32,"score":44,"supported":45,"title":46,"abstract":47,"tags":48},15,true,[33],[29,33],[29],{"code":50,"score":51,"supported":45,"title":52,"abstract":53,"tags":56},"A8",6,[],[54,55],"粮食安全","food security",[],{"code":58,"score":51,"supported":45,"title":59,"abstract":60,"tags":63},"H7",[],[61,62],"气候变化","climate change",[],{"code":65,"score":66,"supported":45,"title":67,"abstract":68,"tags":70},"I2",4,[],[69],"decision support",[27],{"code":72,"score":73,"supported":45,"title":74,"abstract":75,"tags":77},"H6",3,[],[76],"农业生态",[],{"code":79,"score":21,"supported":14,"title":80,"abstract":81,"tags":82},"F1",[],[],[28],{"code":84,"score":21,"supported":14,"title":85,"abstract":86,"tags":87},"I1",[],[],[26],[89,90],"杂交玉米 品种 适宜种植区","图协同过滤 玉米 推荐","杂交玉米品种适宜种植区-3199",5,"10.1093\u002Finsilicoplants\u002Fdiag026",{"doi":93,"openalex_id":95,"authors":96,"venue":10,"cited_by_count":126,"oa_url":6,"card":127,"direction":131,"ingested_from":133},"W7213886755",[97,100,103,106,108,111,114,117,120,123],{"name":98,"orcid":99},"Yanyun Han","https:\u002F\u002Forcid.org\u002F0000-0003-4333-8565",{"name":101,"orcid":102},"Zhongqiang Liu","https:\u002F\u002Forcid.org\u002F0009-0008-4290-8718",{"name":104,"orcid":105},"Aiwen Wang","https:\u002F\u002Forcid.org\u002F0000-0003-0031-5857",{"name":107,"orcid":9},"Yong Zhang",{"name":109,"orcid":110},"Shouhui Pan","https:\u002F\u002Forcid.org\u002F0000-0003-4917-9921",{"name":112,"orcid":113},"Xiangyu Zhao","https:\u002F\u002Forcid.org\u002F0000-0001-7668-1509",{"name":115,"orcid":116},"Qiusi Zhang","https:\u002F\u002Forcid.org\u002F0000-0001-5469-1802",{"name":118,"orcid":119},"Qi Zhang","https:\u002F\u002Forcid.org\u002F0009-0004-5061-5422",{"name":121,"orcid":122},"Xinglin Piao","https:\u002F\u002Forcid.org\u002F0000-0003-3774-5789",{"name":124,"orcid":125},"Kaiyi Wang","https:\u002F\u002Forcid.org\u002F0009-0000-5365-7178",0,{"tldr":128,"method":129,"finding":130,"direction":131,"opportunity":132},"提出图协同过滤框架，将玉米品种适种区评估转化为排序任务，推荐适宜种植地点。","图协同过滤+MLP编码15个品种性状与24个环境变量，BPR+损失，189地点1","方法优于六个基线约5%，冷启动NDCG超0.8，推荐排名与独立产量证据高度一致。","农业人工智能与决策模型","可拓展至多作物多品种跨区域迁移推荐，并融合气候情景预测未来适种区变化。","openalex","2026-09-22T23:30:49.262877Z",{"total":51,"page":21,"page_size":51,"items":136},[137,205,265,342,416,458],{"id":138,"title":139,"url":140,"summary":141,"summary_zh":9,"content":9,"source_name":142,"source_url":9,"published_at":143,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":144,"score_detail":145,"sources":151,"tags":153,"primary_topic":155,"secondary_topics":156,"classification":162,"search_phrases":190,"slug":193,"view_count":194,"doi":195,"paper":196,"created_at":204},2492,"BiFormer-Enhanced YOLOv11n for Accurate Maize Ear Detection in Seed Production Fields(玉米穗识别 BiF-YOLO)","https:\u002F\u002Fdoi.org\u002F10.3390\u002Fagronomy16181780","青岛农业大学机电工程学院 2026 年 9 月 10 日在《Agronomy》(Section Precision and Digital Agriculture)发表。提出 BiF-YOLO 模型基于 YOLOv11n 和 BiFormer 架构,三项目标性增强策略:①在特征提取网络嵌入 C3k2_AdditiveBlock 模块以增强田间被遮挡和小型玉米穗的识别能力;②构建 C2BRA 双层路由注意力机制(Cross-stage partial 2 Bottleneck with Residual Attention)以抑制背景干扰;③采用 ShapeIoU 损失函数。从胶州和张掖种子生产田采集玉米穗图像。BiF-YOLO 模型有效适应复杂田间作业场景,为制种玉米收获机的实时识别与产量估计提供技术支撑。","MDPI Agronomy 2026-09-10","2026-09-09T16:00:00Z",74,{"impact":146,"substance":147,"depth":148,"authority":19,"freshness":149,"relevant":21,"comment":150},16,21,17,7,"面向制种玉米收获的轻量化检测模型改进，方法有针对性增强且数据来自真实制种田，对智能农机与产量估计有实用价值，但属细分技术进展，影响面有限。",[152],{"name":142,"url":140},[26,27,28,29,154],"目标检测",{"code":32,"slug":33,"name":29,"path":34,"parent_code":35},[157],{"code":65,"slug":158,"name":159,"path":160,"parent_code":161},"agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai","I",{"status":38,"locked":14,"version":39,"method":40,"evidence":163},{"candidates":164,"agricultural_evidence":45},[165,170,175,180,186],{"code":32,"score":166,"supported":45,"title":167,"abstract":168,"tags":169},20,[29,33],[29],[29],{"code":65,"score":19,"supported":45,"title":171,"abstract":173,"tags":174},[172],"YOLO",[172],[27,154],{"code":84,"score":66,"supported":45,"title":176,"abstract":177,"tags":179},[],[178],"digital agriculture",[26],{"code":181,"score":73,"supported":45,"title":182,"abstract":183,"tags":185},"A7",[],[184],"间作",[],{"code":79,"score":21,"supported":14,"title":187,"abstract":188,"tags":189},[],[],[28],[191,192],"农业人工智能 智慧农业 目标检测 种业振兴","农业人工智能 智慧农业","农业人工智能智慧农业目标检测种业振兴-2492",2,"10.3390\u002Fagronomy16181780",{"doi":195,"openalex_id":9,"authors":197,"venue":9,"cited_by_count":126,"oa_url":9,"card":198,"direction":131,"ingested_from":203},[],{"tldr":199,"method":200,"finding":201,"direction":131,"opportunity":202},"提出BiF-YOLO模型，在制种田实现玉米穗精准检测。","基于YOLOv11n，融合BiFormer、C3k2_AdditiveBlock","BiF-YOLO有效适应复杂田间场景，提升遮挡和小目标玉米穗检测精度。","可探索轻量化模型在收获机边缘设备上的实时部署与产量估计集成。","agent","2026-09-15T00:04:27.207493Z",{"id":206,"title":207,"url":208,"summary":209,"summary_zh":210,"content":9,"source_name":211,"source_url":208,"published_at":212,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":213,"score_detail":214,"sources":218,"tags":220,"primary_topic":223,"secondary_topics":224,"classification":226,"search_phrases":245,"slug":248,"view_count":126,"doi":249,"paper":250,"created_at":264},4759,"Machine Learning-based Prediction of SVI Parameter in Maize: A Comparative Study and Feature Importance Analysis","https:\u002F\u002Fdoi.org\u002F10.24925\u002Fturjaf.v14i10.2870-2879.9043","Background: Accurate estimation of seedling vigor indices (SVI) is critical for monitoring plant health and predicting yields in precision agriculture. This study comparatively evaluates the performance of different machine learning algorithms in estimating the SVI parameter. Methods: Germination data were collected from 39 maize genotypes under three different salinity conditions (0 mM, 100 mM, and 200 mM NaCl), including the parameters Gmax, MGT, GI, SL, RL, SdL, SMR, RMR, and SVI. The dataset was preprocessed through normalization to the [0–1] range. The SVI parameter was estimated using four different machine learning methods: Random Forest, CatBoost, LightGBM, and Support Vector Regression (SVR). To evaluate the performance of the models, both 10-fold cross-validation and 80%–20% hold-out validation techniques were employed. The model performance was examined using RMSE, MSE, MAE, and R² metrics. Results: The findings show that all models have a high explanatory power in the training dataset. Specifically, the CatBoost model showed the highest success in the training phase (R²=0.9962), whereas the SVR model achieved the highest generalization performance on the test dataset (R²=0.9712, RMSE=0.0582). Feature importance analysis results revealed that the 'App' parameter had the highest impact on the target variable in most models, but different structural interactions were prominent in the LightGBM model. The results demonstrate that machine learning techniques provide high accuracy in SVI estimation and that the tendency towards overfitting is limited. Conclusions: In conclusion, this study confirms that machine learning techniques are powerful and reliable tools for estimating spectral indices in precision agriculture applications. The integration of SVR and ensemble-based models into agricultural decision support systems is expected to significantly contribute to optimizing resource use and improving yield predictions.","背景：准确估计幼苗活力指数（SVI）对于精准农业中监测植物健康和预测产量至关重要。本研究比较评估了不同机器学习算法在估计SVI参数方面的性能。方法：从39个玉米基因型在三种不同盐度条件（0 mM、100 mM和200 mM NaCl）下收集发芽数据，包括Gmax、MGT、GI、SL、RL、SdL、SMR、RMR和SVI等参数。数据集通过归一化至[0–1]范围进行预处理。采用四种不同的机器学习方法估计SVI参数：随机森林、CatBoost、LightGBM和支持向量回归（SVR）。为评估模型性能，采用了10折交叉验证和80%–20%留出验证技术。使用RMSE、MSE、MAE和R²指标检验模型性能。结果：研究结果表明，所有模型在训练数据集中均具有较高的解释力。具体而言，CatBoost模型在训练阶段表现出最高的成功率（R²=0.9962），而SVR模型在测试数据集上取得了最高的泛化性能（R²=0.9712，RMSE=0.0582）。特征重要性分析结果显示，在大多数模型中，“App”参数对目标变量的影响最大，但在LightGBM模型中不同的结构交互作用更为突出。结果表明，机器学习技术在SVI估计中提供了高精度，且过拟合倾向有限。结论：总之，本研究证实机器学习技术是精准农业应用中估计光谱指数的强大而可靠的工具。将SVR和基于集成的模型整合到农业决策支持系统中，预计将显著有助于优化资源利用和改善产量预测。","Turkish Journal of Agriculture - Food Science and Technology","2026-10-11T00:00:00Z",63,{"impact":20,"substance":17,"depth":146,"authority":215,"freshness":216,"relevant":21,"comment":217},12,9,"方法对比扎实、结论可靠，但属细分领域论文，产业影响有限，适合主题聚合而非头条精选。",[219],{"name":211,"url":208},[26,27,221,29,222],"机器学习","耐盐育种",{"code":32,"slug":33,"name":29,"path":34,"parent_code":35},[225],{"code":65,"slug":158,"name":159,"path":160,"parent_code":161},{"status":38,"locked":14,"version":39,"method":40,"evidence":227},{"candidates":228,"agricultural_evidence":45},[229,235,239],{"code":65,"score":230,"supported":45,"title":231,"abstract":233,"tags":234},19,[232],"machine learning",[221,232,69],[27,221],{"code":32,"score":44,"supported":45,"title":236,"abstract":237,"tags":238},[33],[29,33],[29],{"code":84,"score":149,"supported":45,"title":240,"abstract":241,"tags":244},[],[242,243],"精准农业","precision agriculture",[26],[246,247],"玉米 SVI 机器学习","玉米 耐盐 幼苗活力指数","玉米SVI机器学习-4759","10.24925\u002Fturjaf.v14i10.2870-2879.9043",{"doi":249,"openalex_id":251,"authors":252,"venue":211,"cited_by_count":126,"oa_url":208,"card":259,"direction":131,"ingested_from":133},"W7221462188",[253,256],{"name":254,"orcid":255},"Mustafa Turan Arslan","https:\u002F\u002Forcid.org\u002F0000-0001-5498-6571",{"name":257,"orcid":258},"Gülay Zülkadir","https:\u002F\u002Forcid.org\u002F0000-0003-3488-4011",{"tldr":260,"method":261,"finding":262,"direction":131,"opportunity":263},"用四种机器学习模型基于39个玉米基因型盐胁迫发芽数据预测幼苗活力指数SVI。","39个玉米基因型在三种盐浓度下的发芽数据，归一化后用RF、CatBoost、Li","SVR泛化最优（R²=0.9712），CatBoost训练拟合最高；App参数对SVI影响最大。","可扩展到多作物多胁迫环境，结合光谱或图像表型数据提升SVI预测的田间适用性。","2026-10-12T23:30:44.762497Z",{"id":266,"title":267,"url":268,"summary":269,"summary_zh":270,"content":9,"source_name":271,"source_url":268,"published_at":272,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":273,"score_detail":274,"sources":277,"tags":279,"primary_topic":282,"secondary_topics":283,"classification":284,"search_phrases":306,"slug":309,"view_count":51,"doi":310,"paper":311,"created_at":341},4018,"UGV based multi-RGBD SLAM framework for high-throughput in-situ 3D phenotyping of field maize","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.aiia.2026.08.019","Accurate 3D reconstruction of maize populations under real-world field conditions is a prerequisite for individual-level trait in-situ extraction. Conventional visual mapping techniques frequently encounter challenges in complex field environments, which are characterized by weak textures and significant occlusions. To address these limitations, we propose a novel field maize in-situ phenotyping system utilizing synchronized multi-view RGBD cameras mounted on an Unmanned Ground Vehicle (UGV) platform. Specifically, a fixed multi-camera system is first established to introduce cross-view geometric constraints. Then, a semantic-guided feature distribution regulation strategy is proposed to balance feature sampling between crop foreground regions and background areas to alleviate localization degradation caused by spatially imbalanced feature distributions. Then, multi-view observations are subsequently fused to generate a field-scale dense point cloud with high geometric consistency. Finally, an automated processing pipeline is developed by integrating point cloud density projection and three-dimensional topological analysis. This pipeline enables accurate instance segmentation of individual plants within densely planted maize populations and supports robust in-situ quantification of key morphological traits, including plant height and stem diameter. Field experiments conducted under diverse environmental conditions demonstrated a 94.55% accuracy in plant counting. For morphological traits, the pipeline estimated plant height with an R 2 of 0.85 and an RMSE of 41.29 mm. Stem diameter estimation along two orthogonal axes yields R 2 values of 0.79 and 0.76, with corresponding RMSEs of 1.21 mm and 1.03 mm, respectively. The results validate the feasibility and robustness of the proposed framework for three-dimensional reconstruction in complex agricultural environments, and demonstrate its potential as a reliable solution for high-throughput in-situ crop phenotyping.","在真实田间条件下对玉米群体进行精确三维重建，是实现个体尺度性状原位提取的前提。传统视觉建图技术在弱纹理和严重遮挡为特征的复杂田间环境中经常面临挑战。为解决这些局限，我们提出了一种新型田间玉米原位表型分析系统，利用搭载于无人地面车辆（UGV）平台上的同步多视角RGBD相机。具体而言，首先建立固定多相机系统以引入跨视角几何约束。然后，提出一种语义引导的特征分布调控策略，以平衡作物前景区域与背景区域之间的特征采样，从而缓解由空间不平衡特征分布引起的定位退化。随后，融合多视角观测以生成具有高几何一致性的田块尺度稠密点云。最后，通过集成点云密度投影和三维拓扑分析，开发了一套自动化处理流程。该流程能够对密集种植的玉米群体中的单株植物进行准确实例分割，并支持对株高和茎粗等关键形态性状的稳健原位量化。在多种环境条件下进行的田间试验表明，植株计数准确率达到94.55%。对于形态性状，该流程估计株高的R²为0.85，RMSE为41.29 mm。沿两个正交轴估计茎粗的R²分别为0.79和0.76，相应的RMSE分别为1.21 mm和1.03 mm。结果验证了所提出框架在复杂农业环境中进行三维重建的可行性和鲁棒性，并展示了其作为高通量原位作物表型分析可靠解决方案的潜力。","Artificial Intelligence in Agriculture","2026-10-01T00:00:00Z",81,{"impact":17,"substance":18,"depth":17,"authority":275,"freshness":216,"relevant":21,"comment":276},14,"提出UGV多RGBD SLAM框架实现田间玉米高通量原位三维表型，方法新颖、数据扎实，对智慧农业育种与表型研究有参考价值。",[278],{"name":271,"url":268},[26,27,29,280,281],"三维表型","无人地面车",{"code":32,"slug":33,"name":29,"path":34,"parent_code":35},[],{"status":38,"locked":14,"version":39,"method":40,"evidence":285},{"candidates":286,"agricultural_evidence":45},[287,291,298,302],{"code":32,"score":44,"supported":45,"title":288,"abstract":289,"tags":290},[33],[29,33],[29],{"code":292,"score":51,"supported":45,"title":293,"abstract":294,"tags":297},"I3",[],[295,296],"作物表型","crop phenotyping",[],{"code":84,"score":21,"supported":14,"title":299,"abstract":300,"tags":301},[],[],[26],{"code":65,"score":21,"supported":14,"title":303,"abstract":304,"tags":305},[],[],[27],[307,308],"UGV 多相机 SLAM 玉米 三维表型","田间玉米 原位表型 点云","UGV多相机SLAM玉米三维表型-4018","10.1016\u002Fj.aiia.2026.08.019",{"doi":310,"openalex_id":312,"authors":313,"venue":271,"cited_by_count":126,"oa_url":268,"card":335,"direction":339,"ingested_from":133},"W7215041090",[314,316,319,321,324,327,329,331,333],{"name":315,"orcid":9},"Si Yang",{"name":317,"orcid":318},"Yuqiang Liang","https:\u002F\u002Forcid.org\u002F0009-0005-9307-6091",{"name":320,"orcid":9},"Guanming Huang",{"name":322,"orcid":323},"Guangjie Qiu","https:\u002F\u002Forcid.org\u002F0000-0001-9477-9593",{"name":325,"orcid":326},"Weiliang Wen","https:\u002F\u002Forcid.org\u002F0000-0001-7184-5121",{"name":328,"orcid":9},"Wang Chuanyu",{"name":330,"orcid":9},"Wenbo Gou",{"name":332,"orcid":9},"Xinyu Guo",{"name":334,"orcid":9},"Chunjiang Zhao",{"tldr":336,"method":337,"finding":338,"direction":339,"opportunity":340},"提出UGV搭载多RGBD相机的田间玉米原位3D表型系统，实现高通量三维重建与性状提取。","UGV多RGBD同步采集，语义引导特征分布调节与多视角融合，密度投影和拓扑分析分","植株计数准确率94.55%，株高R²=0.85、RMSE=41.29mm，茎径R²约0.76-0.7","农业遥感与作物表型","可探索多模态传感器融合与轻量化SLAM，提升密植冠层遮挡下的实时单株表型精度。","2026-10-02T23:30:19.673397Z",{"id":343,"title":344,"url":345,"summary":346,"summary_zh":347,"content":9,"source_name":348,"source_url":345,"published_at":272,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":273,"score_detail":349,"sources":351,"tags":353,"primary_topic":356,"secondary_topics":357,"classification":362,"search_phrases":382,"slug":385,"view_count":194,"doi":386,"paper":387,"created_at":415},3973,"A physical-context-conditioned multimodal fusion network for estimating maize canopy nitrogen content using multi-source UAV remote sensing data","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.compag.2026.112465","A physical-context-conditioned multimodal fusion network for estimating maize canopy nitrogen content using multi-source UAV remote sensing data。Computers and Electronics in Agriculture","一种基于物理情境条件调节的多模态融合网络，用于利用多源无人机遥感数据估算玉米冠层氮含量。《计算机与电子农业》","Computers and Electronics in Agriculture",{"impact":17,"substance":18,"depth":17,"authority":275,"freshness":216,"relevant":21,"comment":350},"多源无人机遥感与物理上下文多模态融合估算玉米冠层氮素，方法新颖、数据可靠，对精准施肥有参考价值。",[352],{"name":348,"url":345},[26,27,29,354,355],"遥感","氮素营养",{"code":32,"slug":33,"name":29,"path":34,"parent_code":35},[358],{"code":292,"slug":359,"name":360,"path":361,"parent_code":161},"remote-sensing","遥感与农情监测","\u002Ftopic\u002Fdigital-agriculture\u002Fremote-sensing",{"status":38,"locked":14,"version":39,"method":40,"evidence":363},{"candidates":364,"agricultural_evidence":45},[365,369,374,378],{"code":32,"score":44,"supported":45,"title":366,"abstract":367,"tags":368},[33],[29,33],[29],{"code":292,"score":44,"supported":45,"title":370,"abstract":372,"tags":373},[371],"remote sensing",[354,371],[354],{"code":84,"score":21,"supported":14,"title":375,"abstract":376,"tags":377},[],[],[26],{"code":65,"score":21,"supported":14,"title":379,"abstract":380,"tags":381},[],[],[27],[383,384],"无人机 遥感 玉米 氮含量","多模态融合 冠层氮素","无人机遥感玉米氮含量-3973","10.1016\u002Fj.compag.2026.112465",{"doi":386,"openalex_id":388,"authors":389,"venue":348,"cited_by_count":126,"oa_url":9,"card":410,"direction":339,"ingested_from":133},"W7215084428",[390,392,394,397,400,402,404,407],{"name":391,"orcid":9},"Wang Baoju",{"name":393,"orcid":9},"Jing Zhao",{"name":395,"orcid":396},"Yu Yan","https:\u002F\u002Forcid.org\u002F0009-0004-4202-6935",{"name":398,"orcid":399},"Junke Zhu","https:\u002F\u002Forcid.org\u002F0009-0005-7778-9614",{"name":401,"orcid":9},"Xinlong Li",{"name":403,"orcid":9},"Huizheng Wang",{"name":405,"orcid":406},"Weiguang Yang","https:\u002F\u002Forcid.org\u002F0000-0002-9862-5839",{"name":408,"orcid":409},"Yubin Lan","https:\u002F\u002Forcid.org\u002F0000-0001-6664-5571",{"tldr":411,"method":412,"finding":413,"direction":339,"opportunity":414},"提出物理上下文条件化多模态融合网络，用多源无人机遥感估算玉米冠层氮含量。","多源无人机遥感数据与物理上下文条件化多模态融合网络。","该网络能提升玉米冠层氮含量估算精度，融合多源数据优于单一模态。","可探索物理先验与多模态融合在不同作物、生育期及跨区域迁移中的泛化能力。","2026-10-02T23:30:02.413294Z",{"id":417,"title":418,"url":419,"summary":420,"summary_zh":9,"content":421,"source_name":422,"source_url":9,"published_at":423,"category":424,"cover_url":9,"hotness":13,"is_selected":14,"score":15,"score_detail":425,"sources":427,"tags":429,"primary_topic":431,"secondary_topics":432,"classification":433,"search_phrases":453,"slug":456,"view_count":66,"doi":9,"paper":9,"created_at":457},3933,"大数据智能决策技术助力黑龙江玉米生产数字化转型","https:\u002F\u002Fnews.sciencenet.cn\u002Fhtmlnews\u002F2026\u002F9\u002F572335.shtm","9月19日，国家重点研发计划项目\"粮食生产大数据平台研发与应用\"玉米智慧生产现场观摩会在北大荒集团红星农场举办。项目组首次将深度学习应用于规模化农场种植布局优化，研发基于PPO强化学习的农场尺度智能种植规划算法；研发自适应玉米生育进程动态监测预报算法，分辨率10m~1km，可逐日动态监测，关键生育期可提前20天精准预报；构建生产、生态维度综合效益评估算法。","[![Image 1: 科学网新闻频道](https:\u002F\u002Fnews.sciencenet.cn\u002Fimages\u002Fnews.jpg)](https:\u002F\u002Fnews.sciencenet.cn\u002F)\n\n[生命科学](https:\u002F\u002Fwww.sciencenet.cn\u002Flife\u002F) | [医学科学](https:\u002F\u002Fwww.sciencenet.cn\u002Fmedicine\u002F) | [化学科学](https:\u002F\u002Fwww.sciencenet.cn\u002Fchemistry\u002F) | [工程材料](https:\u002F\u002Fwww.sciencenet.cn\u002Fmaterial\u002F) | [信息科学](https:\u002F\u002Fwww.sciencenet.cn\u002Finformation\u002F) | [地球科学](https:\u002F\u002Fwww.sciencenet.cn\u002Fearth\u002F) | [数理科学](https:\u002F\u002Fwww.sciencenet.cn\u002Fmathematics\u002F) | [管理综合](https:\u002F\u002Fwww.sciencenet.cn\u002Fpolicy\u002F)[站内规定](https:\u002F\u002Fblog.sciencenet.cn\u002Fblog-45-1064777.html) | [手机版](https:\u002F\u002Fwap.sciencenet.cn\u002F)\n\n[首页](https:\u002F\u002Fwww.sciencenet.cn\u002F) | [新闻](https:\u002F\u002Fnews.sciencenet.cn\u002F) | [博客](https:\u002F\u002Fblog.sciencenet.cn\u002Fblog.php) | [院士](https:\u002F\u002Fnews.sciencenet.cn\u002Fys\u002F) | [人才](https:\u002F\u002Ftalent.sciencenet.cn\u002F) | [会议](https:\u002F\u002Fmeeting.sciencenet.cn\u002F) | [基金·项目](https:\u002F\u002Ffund.sciencenet.cn\u002F) | [论文](https:\u002F\u002Fpaper.sciencenet.cn\u002F) | [绘图](https:\u002F\u002Fvisual-web.sciencenet.cn\u002F) | [视频·直播](https:\u002F\u002Fblog.sciencenet.cn\u002Fblog.php?mod=video) | [小柯机器人](https:\u002F\u002Fpaper.sciencenet.cn\u002FAInews) | [医学科普](https:\u002F\u002Fyxkp.sciencenet.cn\u002F)\n\n作者：李晨 来源：中国科学报 发布时间：2026\u002F9\u002F29 20:15:16 选择字号：小 中 大 大数据智能决策技术助力黑龙江玉米生产数字化转型 9月19日，国家重点研发计划项目“粮食生产大数据平台研发与应用”玉米智慧生产现场观摩会在北大荒集团红星农场举办。会议由中国农业科学院农业资源与农业区划研究所、北京市农林科学院信息技术研究中心主办，北大荒集团北安分公司、红星农场协办，共70余人参会。 ![Image 2](https:\u002F\u002Frmtzx.sciencenet.cn\u002Fkxwsprint\u002F6abb7e11e4b078fce44b1363.jpg)现场展示的不同决策方案的应用效果。中国农科院供图 会议聚焦玉米生产全程大数据智能决策技术示范应用，围绕玉米产前、产中、产后各关键环节，现场展示了智能分析算法在农机作业调度、无人机出苗率调查、变量精准施肥、苗情长势、病虫害监测和预警及估产、种植规划和适宜收获期等场景的应用。项目负责人、北京市农林科学院信息技术研究中心杨贵军研究员系统介绍了粮食生产大数据平台的研发与应用进展。 课题负责人、中国农业科学院农业资源与农业区划研究所研究员孙亮重点介绍了玉米智能决策技术领域取得的突破性成果。 在产前种植规划方面，项目组首次将深度学习应用于规模化农场种植布局优化，研发了基于PPO强化学习的农场尺度智能种植规划算法，在保障国家粮食安全的前提下，统筹经济效益与农业绿色发展，实现多目标协同。 ![Image 3](https:\u002F\u002Frmtzx.sciencenet.cn\u002Fkxwsprint\u002F6abb7e55e4b078fce44b1365.png)观摩会现场。中国农科院供图 在产中动态监测方面，项目组研发适配粮食主产区、农场多尺度的自适应玉米生育进程动态监测预报算法，分辨率10m~1km，可实现逐日动态监测，关键生育期可提前20天完成精准预报。该算法为玉米关键生育期追肥处方的动态实时更新、产量与适宜收获期预测提供了坚实技术基础。 在产后效益评价方面，项目组构建生产、生态维度综合效益评估算法，围绕北大荒玉米生产全环节社会化服务模式，精准量化其在提升生产效益、改善农田生态等方面核心优势，为优化农场要素配置、推动农业绿色发展提供理论与实践支撑。 据介绍，通过大数据平台与田间生产管理深度融合，玉米生产全过程实现可视化、精准化、智能化管控，智能决策技术在玉米生产管理中的应用水平进一步提升。相关成果将为黑龙江玉米生产数字化转型提供有力技术支撑，助力国家粮食安全保障能力持续提升。 版权声明：凡本网注明“来源：中国科学报、科学网、科学新闻杂志”的所有作品，网站转载，请在正文上方注明来源和作者，且不得对内容作实质性改动；微信公众号、头条号等新媒体平台，转载请联系授权。邮箱：shouquan@stimes.cn。\n\n打印 发E-mail给：\n\n以下评论只代表网友个人观点，不代表科学网观点。\n\n[![Image 4](https:\u002F\u002Fnews.sciencenet.cn\u002Fimages\u002Fnewcomm.gif)](https:\u002F\u002Fnews.sciencenet.cn\u002Fhtml\u002Fcomment.aspx?id=572335)\n\n相关新闻 相关论文\n*   1\n*   [北京理工大学：解除与教师裴某聘用关系](https:\u002F\u002Fnews.sciencenet.cn\u002Fhtmlnews\u002F2026\u002F9\u002F572327.shtm)\n\n*   2\n*   [中国科学院启动实施科学前沿极限突破计划](https:\u002F\u002Fnews.sciencenet.cn\u002Fhtmlnews\u002F2026\u002F9\u002F572326.shtm)\n\n*   3\n*   [中国科技期刊卓越行动计划高起点新刊项目启动申报](https:\u002F\u002Fnews.sciencenet.cn\u002Fhtmlnews\u002F2026\u002F9\u002F572325.shtm)\n\n*   4\n*   [2026年诺奖即将揭晓，权威专家为您全面解读！](https:\u002F\u002Fnews.sciencenet.cn\u002Fhtmlnews\u002F2026\u002F9\u002F572324.shtm)\n\n*   5\n*   [科技部：持续打击买卖论文等违规行为](https:\u002F\u002Fnews.sciencenet.cn\u002Fhtmlnews\u002F2026\u002F9\u002F572320.shtm)\n\n*   6\n*   [不再“一本书教十年” 他们让本科生站上科学前沿](https:\u002F\u002Fnews.sciencenet.cn\u002Fhtmlnews\u002F2026\u002F9\u002F572319.shtm)\n\n*   7\n*   [聚焦自身免疫病与胰腺癌，瑞金医院发表两项新成果](https:\u002F\u002Fnews.sciencenet.cn\u002Fhtmlnews\u002F2026\u002F9\u002F572312.shtm)\n\n*   8\n*   [科学家首次观测到共振荧光中的多光子纠缠](https:\u002F\u002Fnews.sciencenet.cn\u002Fhtmlnews\u002F2026\u002F9\u002F572310.shtm)\n\n*   1\n*   [E&E | 西南林业大学李兆元课题组解释雉科鸟类多样性的空间维持机制](https:\u002F\u002Fnews.sciencenet.cn\u002Fhtmlpaper\u002F2026\u002F9\u002F2026929103559990156110.shtm)\n\n*   2\n*   [NPH | 马泽清团队提出外生菌根形态与觅食空间理论框架](https:\u002F\u002Fnews.sciencenet.cn\u002Fhtmlpaper\u002F2026\u002F9\u002F2026929102613990156108.shtm)\n\n*   3\n*   [GCB Communications | 南极地衣会成为气候变化的“赢家”吗？](https:\u002F\u002Fnews.sciencenet.cn\u002Fhtmlpaper\u002F2026\u002F9\u002F202692910224380156107.shtm)\n\n*   4\n*   [江苏科技大学施敏杰、杨骏，中国科学院合肥物质院周宏建Angew：基于有机分子电极亲核调控策略的高效低碳电化学脱盐技术](https:\u002F\u002Fnews.sciencenet.cn\u002Fhtmlpaper\u002F2026\u002F9\u002F2026929101854492156106.shtm)\n\n*   5\n*   [南京大学黄宏文、胡征Angew：氢键驱动界面调控构筑高性能低铂质子交换膜燃料电池](https:\u002F\u002Fnews.sciencenet.cn\u002Fhtmlpaper\u002F2026\u002F9\u002F202692910155352156105.shtm)\n\n*   6\n*   [Angew. Chem.: 一维多酸-铜链协同“静\u002F动”双调控！实现CO2→CH4高效转化](https:\u002F\u002Fnews.sciencenet.cn\u002Fhtmlpaper\u002F2026\u002F9\u002F2026929101218833156104.shtm)\n\n*   7\n*   [清华大学张如范团队 Nat. Commun.：多功能辐射制冷气凝胶纤维织物](https:\u002F\u002Fnews.sciencenet.cn\u002Fhtmlpaper\u002F2026\u002F9\u002F2026928181459630156091.shtm)\n\n*   8\n*   [2024-2025高引文章|MDPI编辑荐读](https:\u002F\u002Fnews.sciencenet.cn\u002Fhtmlpaper\u002F2026\u002F9\u002F2026928143220693156090.shtm)\n\n[图片新闻](https:\u002F\u002Fnews.sciencenet.cn\u002Fimgnews.aspx)\n\n[![Image 5: “鹏城首发星”成功入轨](https:\u002F\u002Fnews.sciencenet.cn\u002F\u002Fupload\u002Fnews\u002Fimages\u002F2026\u002F9\u002F20269231515219900.png)](https:\u002F\u002Fnews.sciencenet.cn\u002Fhttps:\u002F\u002Fnews.sciencenet.cn\u002Fhtmlnews\u002F2026\u002F9\u002F571953.shtm)[![Image 6: 研究揭示甲烷部分氧化制合成气驱动因素](https:\u002F\u002Fnews.sciencenet.cn\u002F\u002Fupload\u002Fnews\u002Fimages\u002F2026\u002F9\u002F2026917125225680.jpg)](https:\u002F\u002Fnews.sciencenet.cn\u002F\u002Fhtmlnews\u002F2026\u002F9\u002F571610.shtm)\n[![Image 7: 科学家揭示火星大气物质运输规律](https:\u002F\u002Fnews.sciencenet.cn\u002F\u002Fupload\u002Fnews\u002Fimages\u002F2026\u002F9\u002F2026915174236470.jpg)](https:\u002F\u002Fnews.sciencenet.cn\u002F\u002Fhtmlnews\u002F2026\u002F9\u002F571503.shtm)[![Image 8: 超导双层镍氧化物薄膜的三维电子结构](https:\u002F\u002Fnews.sciencenet.cn\u002F\u002Fupload\u002Fnews\u002Fimages\u002F2026\u002F9\u002F20269151415154110.png)](https:\u002F\u002Fnews.sciencenet.cn\u002Fhttps:\u002F\u002Fnews.sciencenet.cn\u002Fhtmlnews\u002F2026\u002F9\u002F571484.shtm)\n[>>更多](https:\u002F\u002Fnews.sciencenet.cn\u002Fimgnews.aspx)\n\n[一周新闻排行](https:\u002F\u002Fnews.sciencenet.cn\u002Fzphnews.aspx)\n*   1\n*   [西部冰冻圈加速“退化”：冰川消融敲响高山安全警钟](https:\u002F\u002Fnews.sciencenet.cn\u002F\u002Fhtmlnews\u002F2026\u002F9\u002F572233.shtm)\n\n*   2\n*   [第六届全国农业科技发展战略智库联盟大会在兰州召开](https:\u002F\u002Fnews.sciencenet.cn\u002F\u002Fhtmlnews\u002F2026\u002F9\u002F572311.shtm)\n\n*   3\n*   [西安交通大学成立一中心、两学院](https:\u002F\u002Fnews.sciencenet.cn\u002F\u002Fhtmlnews\u002F2026\u002F9\u002F572170.shtm)\n\n*   4\n*   [《中国科学院“十五五”发展规划》发布](https:\u002F\u002Fnews.sciencenet.cn\u002F\u002Fhtmlnews\u002F2026\u002F9\u002F572194.shtm)\n\n*   5\n*   [北京理工大学：解除与教师裴某聘用关系](https:\u002F\u002Fnews.sciencenet.cn\u002F\u002Fhtmlnews\u002F2026\u002F9\u002F572327.shtm)\n\n*   6\n*   [导师要求写周报，算额外负担吗？](https:\u002F\u002Fnews.sciencenet.cn\u002F\u002Fhtmlnews\u002F2026\u002F9\u002F572133.shtm)\n\n*   7\n*   [2026年诺奖即将揭晓，权威专家为您全面解读！](https:\u002F\u002Fnews.sciencenet.cn\u002F\u002Fhtmlnews\u002F2026\u002F9\u002F572324.shtm)\n\n*   8\n*   [“分子刹车”拆解玉米“争光长高”指令](https:\u002F\u002Fnews.sciencenet.cn\u002F\u002Fhtmlnews\u002F2026\u002F9\u002F572358.shtm)\n\n*   9\n*   [拖延≠懒惰，真相也许是这样](https:\u002F\u002Fnews.sciencenet.cn\u002F\u002Fhtmlnews\u002F2026\u002F9\u002F572150.shtm)\n\n*   10\n*   [西安交通大学成立两大学院](https:\u002F\u002Fnews.sciencenet.cn\u002F\u002Fhtmlnews\u002F2026\u002F9\u002F572144.shtm)\n\n[编辑部推荐博文](http:\u002F\u002Fblog.sciencenet.cn\u002Fblog.php?mod=recommend)\n*   ![Image 9](https:\u002F\u002Fnews.sciencenet.cn\u002Fimages\u002Ft11.gif)\n*   [封面文章|《水生生物学报》：微藻蛋白的研究进展](https:\u002F\u002Fblog.sciencenet.cn\u002Fblog-45-1554679.html)\n\n*   ![Image 10](https:\u002F\u002Fnews.sciencenet.cn\u002Fimages\u002Ft11.gif)\n*   [科学网2026年8月十佳博文榜单公布！](https:\u002F\u002Fblog.sciencenet.cn\u002Fblog-45-1551789.html)\n\n*   ![Image 11](https:\u002F\u002Fnews.sciencenet.cn\u002Fimages\u002Ft11.gif)\n*   [天上最亮的星，你还看得见吗？](https:\u002F\u002Fblog.sciencenet.cn\u002Fblog-1152260-1554948.html)\n\n*   ![Image 12](https:\u002F\u002Fnews.sciencenet.cn\u002Fimages\u002Ft11.gif)\n*   [当西方城市意象理论走进东方古城后](https:\u002F\u002Fblog.sciencenet.cn\u002Fblog-38998-1554940.html)\n\n*   ![Image 13](https:\u002F\u002Fnews.sciencenet.cn\u002Fimages\u002Ft11.gif)\n*   [国自然“免疫力数字解码”重大项目](https:\u002F\u002Fblog.sciencenet.cn\u002Fblog-290052-1554895.html)\n\n*   ![Image 14](https:\u002F\u002Fnews.sciencenet.cn\u002Fimages\u002Ft11.gif)\n*   [有效交流的前提](https:\u002F\u002Fblog.sciencenet.cn\u002Fblog-729911-1554792.html)\n\n*   [更多>>](https:\u002F\u002Fblog.sciencenet.cn\u002Fblog.php?mod=recommend)","中国科学报·2026年09月29日","2026-09-29T12:15:16Z","报道",{"impact":18,"substance":166,"depth":146,"authority":275,"freshness":149,"relevant":21,"comment":426},"国家重点研发计划玉米智慧生产现场观摩，含强化学习种植规划、生育期动态监测等实质技术进展，对农业数字化转型有参考价值。",[428],{"name":422,"url":419},[26,27,54,430,29],"黑龙江",{"code":32,"slug":33,"name":29,"path":34,"parent_code":35},[],{"status":38,"locked":14,"version":39,"method":40,"evidence":434},{"candidates":435,"agricultural_evidence":45},[436,440,445,449],{"code":32,"score":215,"supported":45,"title":437,"abstract":438,"tags":439},[29],[29],[29],{"code":65,"score":66,"supported":45,"title":441,"abstract":442,"tags":444},[],[443],"深度学习",[27],{"code":50,"score":21,"supported":14,"title":446,"abstract":447,"tags":448},[],[],[54],{"code":84,"score":21,"supported":14,"title":450,"abstract":451,"tags":452},[],[],[26],[454,455],"黑龙江 玉米 智慧生产","粮食生产大数据平台 玉米","黑龙江玉米智慧生产-3933","2026-10-02T11:53:11.085935Z",{"id":459,"title":460,"url":461,"summary":462,"summary_zh":463,"content":9,"source_name":464,"source_url":461,"published_at":465,"category":12,"cover_url":9,"hotness":13,"is_selected":45,"score":273,"score_detail":466,"sources":468,"tags":470,"primary_topic":472,"secondary_topics":473,"classification":475,"search_phrases":501,"slug":504,"view_count":92,"doi":505,"paper":506,"created_at":538},3842,"Raspberry Pi Lightweight Deep Learning Models Deployed for Corn Growth Stage Classification Using PhenoCam Images","https:\u002F\u002Fdoi.org\u002F10.3390\u002Frs18193337","Identifying crop growth stages is essential for comprehensive assessment that facilitates timely and precise agronomic management. However, conventional field-based manual scouting is labor-intensive, time-consuming, and lacks scalability. Digital field images with advanced models deployed on portable edge-computing devices offer an alternative solution for real-time crop monitoring and decision support. Therefore, in this study, the standard 8 corn growth stages were defined as 8 classes based on distinct visual canopy appearance patterns observed in PhenoCam (near-surface [proximal] remote sensing network camera) imagery collected from 10 PhenoCam sites. These classes do not correspond directly to standard agronomic vegetation and reproductive growth stages. For modeling, four lightweight deep learning (DL) models, ELiteCrop0, ELiteCrop1, ELiteCrop4, and MobNetCropV2, were developed and evaluated across five image vertical clipping levels (0–40%) using a supercomputer (NDSU CCAST). The optimized model was subsequently deployed on a single-board Raspberry Pi 5 computer for edge inference. Model training accounted for the majority of the total CPU time, exceeding 97%, while the testing times ranged from 0.01–0.12 min, enabling real-time applications. Among the models, ELiteCrop0 achieved the most balanced performance with a confusion-matrix diagonal ratio (CMDR) of 0.93, followed by ELiteCrop1 (CMDR = 0.92). Overall, model performance decreased with increasing vertical clipping; therefore, a moderate image clipping (0–10%) was recommended for improved computational efficiency. Analysis with a supercomputer produced an intrasite (same training sites) accuracy of 0.90–0.93 (Raspberry Pi: 0.78–0.81) and an intersite (unseen test sites) accuracy of 0.48–0.50 (Raspberry Pi: 0.41–0.43), indicating challenges with model generalization. Raspberry Pi processed ≈ 1000 images\u002Fmin at 68 °C, with a PhenoCam:Raspberry Pi processing ratio of 1:2500 at 50% capacity over an 8 h day. Overall, this study presents a scalable and cost-effective solution for real-time corn growth stage monitoring in precision agriculture, which could be extended to other crops.","识别作物生长阶段对于全面评估、促进及时精准的农艺管理至关重要。然而，传统的田间人工巡查劳动强度大、耗时且缺乏可扩展性。部署在便携式边缘计算设备上的数字田间图像与先进模型为实时作物监测和决策支持提供了替代方案。因此，本研究基于从10个PhenoCam（近地面[近端]遥感网络相机）站点采集的影像中观察到的不同冠层视觉外观模式，将标准的8个玉米生长阶段定义为8个类别。这些类别并不直接对应于标准的农艺营养生长和生殖生长阶段。在建模方面，开发了四种轻量级深度学习（DL）模型——ELiteCrop0、ELiteCrop1、ELiteCrop4和MobNetCropV2，并利用超级计算机（NDSU CCAST）在五种图像垂直裁剪水平（0–40%）下进行了评估。随后，优化后的模型被部署到单板Raspberry Pi 5计算机上进行边缘推理。模型训练占总CPU时间的大部分，超过97%，而测试时间为0.01–0.12分钟，支持实时应用。在各模型中，ELiteCrop0实现了最均衡的性能，混淆矩阵对角线比率（CMDR）为0.93，其次是ELiteCrop1（CMDR = 0.92）。总体而言，模型性能随垂直裁剪比例的增加而下降；因此，建议采用适度的图像裁剪（0–10%）以提高计算效率。使用超级计算机的分析得出站内（相同训练站点）准确率为0.90–0.93（Raspberry Pi：0.78–0.81），站间（未见测试站点）准确率为0.48–0.50（Raspberry Pi：0.41–0.43），表明模型泛化面临挑战。Raspberry Pi在68 °C下处理约1000张图像\u002F分钟，在8小时工作日50%容量下，PhenoCam与Raspberry Pi的处理比为1:2500。总体而言，本研究为精准农业中的实时玉米生长阶段监测提供了一种可扩展且经济高效的解决方案，并可扩展至其他作物。","Remote Sensing","2026-09-29T00:00:00Z",{"impact":17,"substance":18,"depth":17,"authority":275,"freshness":216,"relevant":21,"comment":467},"该研究将轻量深度学习模型部署于树莓派实现玉米生育期实时识别，方法新颖、数据扎实，对精准农业边缘监测有参考价值。",[469],{"name":464,"url":461},[26,27,471,29,354],"边缘计算",{"code":32,"slug":33,"name":29,"path":34,"parent_code":35},[474],{"code":65,"slug":158,"name":159,"path":160,"parent_code":161},{"status":38,"locked":14,"version":39,"method":40,"evidence":476},{"candidates":477,"agricultural_evidence":45},[478,483,488,492,496],{"code":65,"score":17,"supported":45,"title":479,"abstract":481,"tags":482},[480],"deep learning",[443,480,69],[27],{"code":32,"score":44,"supported":45,"title":484,"abstract":486,"tags":487},[485],"corn",[29,485],[29],{"code":84,"score":149,"supported":45,"title":489,"abstract":490,"tags":491},[],[242,243],[26],{"code":292,"score":149,"supported":45,"title":493,"abstract":494,"tags":495},[],[354,371],[354],{"code":497,"score":66,"supported":45,"title":498,"abstract":499,"tags":500},"I5",[],[471],[471],[502,503],"Raspberry Pi 玉米 生长阶段","PhenoCam 玉米 深度学习","RaspberryPi玉米生长阶段-3842","10.3390\u002Frs18193337",{"doi":505,"openalex_id":507,"authors":508,"venue":464,"cited_by_count":126,"oa_url":461,"card":533,"direction":339,"ingested_from":133},"W7214787420",[509,512,515,518,521,524,527,530],{"name":510,"orcid":511},"Anne Joice","https:\u002F\u002Forcid.org\u002F0000-0001-7499-9222",{"name":513,"orcid":514},"Humeera Tazeen","https:\u002F\u002Forcid.org\u002F0000-0003-2411-0725",{"name":516,"orcid":517},"Talha Tufaique","https:\u002F\u002Forcid.org\u002F0000-0002-4490-5341",{"name":519,"orcid":520},"Mathala Juliet Gupta","https:\u002F\u002Forcid.org\u002F0000-0002-3886-2429",{"name":522,"orcid":523},"Cannayen Igathinathane","https:\u002F\u002Forcid.org\u002F0000-0001-8884-7959",{"name":525,"orcid":526},"Nitin Rai","https:\u002F\u002Forcid.org\u002F0000-0002-9350-9601",{"name":528,"orcid":529},"Craig W. Whippo","https:\u002F\u002Forcid.org\u002F0000-0002-7626-6919",{"name":531,"orcid":532},"David Walter Archer","https:\u002F\u002Forcid.org\u002F0000-0002-4816-7040",{"tldr":534,"method":535,"finding":536,"direction":339,"opportunity":537},"用PhenoCam图像训练轻量深度学习模型，在树莓派上实时分类玉米8个生长阶段。","基于10个PhenoCam站点图像，开发4种轻量DL模型，在树莓派5上边缘推理。","ELiteCrop0表现最佳（CMDR=0.93），站点内精度0.90-0.93，但站点间泛化差（0","站点间泛化差是主要瓶颈，可研究跨站点迁移学习或域适应以提升模型鲁棒性。","2026-09-30T23:30:29.168194Z"]