[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-3850":3,"related-3850":87},{"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":86},3850,"Изследване на приложимостта на вегетационни индекси за диагностика на азотния статус при тютюн","https:\u002F\u002Fdoi.org\u002F10.61308\u002Fpxjs7160","This study evaluates the applicability of seven vegetation indices (CI Green, CI Red, CI Red Edge, LCI, DCNI, NDCI, and CNDVI) for diagnosing nitrogen status in oriental and Virginia tobacco. Leaf level spectral measurements obtained with a CI-710s SpectraVue spectrometer were compared with laboratory determined total nitrogen content.In the oriental tobacco cultivar Krumovgrad 58, all indices exhibited strong positive correlations with leaf nitrogen concentration (r = 0.772-0.965), with LCI, CI Green, and CI Red Edge showing the highest sensitivity. Similar results were observed in the Virginia 842 cultivar (r = 0.797-0.916), where four indices (CI Green, CI Red Edge, CNDVI, and NDCI) demonstrated statistically significant relationships.Linear regression models developed for both cultivars showed high predictive performance (R2 = 0.596-0.931 for oriental tobacco and R2 = 0.635-0.839 for Virginia tobacco), confirming the potential of vegetation indices for indirect nitrogen estimation. The differences between the cultivars of the two tobacco types can be explained by specific features of leaf morphology and mesophyll structure, which influence the spectral signal.The results indicate that spectral vegetation indices are a reliable tool for monitoring nitrogen nutrition in tobacco. Further calibration and expanded studies across cultivars, growth stages, and environmental conditions are required to enhance model robustness and practical applicability.","本研究评估了七种植被指数（CI Green、CI Red、CI Red Edge、LCI、DCNI、NDCI和CNDVI）在诊断东方型烟草和弗吉尼亚型烟草氮素状况方面的适用性。利用CI-710s SpectraVue光谱仪获取的叶片尺度光谱测量数据与实验室测定的全氮含量进行了比较。在东方型烟草品种Krumovgrad 58中，所有植被指数均与叶片氮浓度呈强正相关（r = 0.772–0.965），其中LCI、CI Green和CI Red Edge表现出最高的敏感性。在弗吉尼亚型品种Virginia 842中观察到了类似的结果（r = 0.797–0.916），其中四个植被指数（CI Green、CI Red Edge、CNDVI和NDCI）表现出统计学显著的关系。为两个品种建立的线性回归模型显示出较高的预测性能（东方型烟草R² = 0.596–0.931，弗吉尼亚型烟草R² = 0.635–0.839），证实了植被指数在间接估算氮素方面的潜力。两种烟草类型品种间的差异可归因于叶片形态和叶肉结构的特异性，这些特性影响了光谱信号。结果表明，光谱植被指数是监测烟草氮素营养的可靠工具。仍需进一步校准并在品种、生育期和环境条件方面开展扩展研究，以增强模型的稳健性和实际适用性。",null,"Bulgarian Journal of Soil Science Agrochemistry and Ecology","2026-09-28T00:00:00Z","论文",10,false,66,{"impact":17,"substance":18,"depth":19,"authority":20,"freshness":21,"relevant":22,"comment":23},8,20,16,13,9,1,"该研究验证了七种植被指数在烟草氮素诊断中的适用性，方法扎实、数据可靠，对智慧农业遥感监测有参考价值，但属细分作物基础研究，影响范围有限。",[25],{"name":10,"url":6},[27,28,29,30,31],"智慧农业","遥感","植被指数","烟草种植","氮素诊断",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},"A6","cash-crops","棉麻糖与特色经作","\u002Ftopic\u002Ffield-crops\u002Fcash-crops","A",[],{"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":49,"tags":52},6,true,[],[50,51],"烟草","tobacco",[],{"code":54,"score":55,"supported":14,"title":56,"abstract":57,"tags":58},"I3",2,[],[],[29,28],{"code":60,"score":22,"supported":14,"title":61,"abstract":62,"tags":63},"I1",[],[],[27],[65,66],"烟草 氮素 植被指数","Krumovgrad 58 烟草","烟草氮素植被指数-3850",3,"10.61308\u002Fpxjs7160",{"doi":69,"openalex_id":71,"authors":72,"venue":10,"cited_by_count":77,"oa_url":78,"card":79,"direction":83,"ingested_from":85},"W7214629574",[73,75],{"name":74,"orcid":9},"Радка Божинова",{"name":76,"orcid":9},"Yovko Dyulgerski",0,"https:\u002F\u002Fagriacad.eu\u002Fojs\u002Findex.php\u002Fbjssae\u002Farticle\u002Fdownload\u002F6863\u002F5937",{"tldr":80,"method":81,"finding":82,"direction":83,"opportunity":84},"评估七种植被指数诊断两种烟草氮素状况的适用性并建立回归模型。","CI-710s光谱仪叶片光谱与实验室全氮含量对比，线性回归建模。","多指数与叶片氮浓度强相关，LCI、CI Green、CI Red Edge最敏感，模型R²达0.93","农业遥感与作物表型","可扩展至冠层尺度、多品种多生育期及环境条件，提升模型鲁棒性与田间实用性。","openalex","2026-09-30T23:30:31.090531Z",{"total":46,"page":22,"page_size":46,"items":88},[89,148,222,280,336,393],{"id":90,"title":91,"url":92,"summary":93,"summary_zh":94,"content":9,"source_name":95,"source_url":92,"published_at":96,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":97,"score_detail":98,"sources":100,"tags":102,"primary_topic":106,"secondary_topics":107,"classification":113,"search_phrases":132,"slug":135,"view_count":68,"doi":136,"paper":137,"created_at":147},3838,"Forecasting Cotton Productivity in the Mil-Mugan Economic Region using Climate Variables and Remote Sensing Data","https:\u002F\u002Fdoi.org\u002F10.52171\u002Fherald.342","Satellite imagery is considered as one of the most valuable and functional data sources for evaluating and forecasting production in agricultural systems. This study aims to forecast productivity in cotton fields within the Mil-Mugan economic region by incorporating vegetation indices and climatic variables. NDVI indicators were derived from Sentinel-2 data, temperature (T) was obtained from ERA5-Land data, and precipitation (R) was collected from the CHIRPS database. The data were analyzed using Google Earth Engine, ArcGIS Pro, and Python environments, consolidating them into an integrated temporal-spatial dimension. The correlation and regression analysis assessed the link among NDVI, temperature, precipitation, and productivity, leading to the development of a multivariate model. The findings derived from the model indicate that NDVI is the primary factor influencing productivity. The model developed for one region was implemented in the other three locations, and the correlation between the actual and predicted indicators was assessed using R² and RMSE metrics. The results indicate that the model's forecast accuracy fluctuates within the 90% range, considered appropriate for predicting regional cotton yield. The study's findings are significant for enhancing operational monitoring and planning of agrotechnical decisions with satellite data.","卫星影像被认为是评估和预测农业系统产量最有价值且最实用的数据来源之一。本研究旨在通过整合植被指数与气候变量，预测米尔-穆甘经济区棉田的生产力。NDVI指标源自Sentinel-2数据，温度（T）取自ERA5-Land数据，降水（R）来自CHIRPS数据库。数据在Google Earth Engine、ArcGIS Pro和Python环境中进行分析，并将其整合为统一的时空维度。通过相关性与回归分析评估了NDVI、温度、降水与生产力之间的关系，进而构建了多变量模型。模型结果表明，NDVI是影响生产力的主要因素。将为某一区域开发的模型应用于其他三个地点，并使用R²和RMSE指标评估实际值与预测值之间的相关性。结果表明，模型的预测精度在90%范围内波动，被认为适用于预测区域棉花产量。本研究结果对于利用卫星数据加强农艺决策的业务化监测与规划具有重要意义。","Herald of the Azerbaijan Engineering Academy","2026-09-29T00:00:00Z",63,{"impact":17,"substance":18,"depth":19,"authority":13,"freshness":21,"relevant":22,"comment":99},"基于Sentinel-2与气候变量的棉花产量预测模型，方法扎实、精度达90%，但属区域案例研究，公共影响有限。",[101],{"name":95,"url":92},[27,103,104,28,105],"产量预测","棉花","NDVI",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[108],{"code":54,"slug":109,"name":110,"path":111,"parent_code":112},"remote-sensing","遥感与农情监测","\u002Ftopic\u002Fdigital-agriculture\u002Fremote-sensing","I",{"status":40,"locked":14,"version":41,"method":42,"evidence":114},{"candidates":115,"agricultural_evidence":47},[116,122,128],{"code":33,"score":117,"supported":47,"title":118,"abstract":120,"tags":121},15,[119],"cotton",[104,119],[104],{"code":54,"score":123,"supported":47,"title":124,"abstract":126,"tags":127},11,[125],"remote sensing",[],[105,103,28],{"code":60,"score":22,"supported":14,"title":129,"abstract":130,"tags":131},[],[],[27],[133,134],"Mil-Mugan 棉花 遥感","Sentinel-2 NDVI 棉花产量","Mil-Mugan棉花遥感-3838","10.52171\u002Fherald.342",{"doi":136,"openalex_id":138,"authors":139,"venue":95,"cited_by_count":77,"oa_url":92,"card":142,"direction":83,"ingested_from":85},"W7214748696",[140],{"name":141,"orcid":9},"B.H. Babashli",{"tldr":143,"method":144,"finding":145,"direction":83,"opportunity":146},"结合Sentinel-2 NDVI与气候变量，构建多元模型预测米尔-穆甘地区棉花产量。","Sentinel-2 NDVI、ERA5-Land温度、CHIRPS降水，GEE","NDVI是产量主控因子，模型跨区预测精度约90%，可用于区域估产。","可引入多时相NDVI与机器学习，提升跨区域迁移性和极端气候下的估产鲁棒性。","2026-09-30T23:30:27.786044Z",{"id":149,"title":150,"url":151,"summary":152,"summary_zh":9,"content":9,"source_name":153,"source_url":151,"published_at":154,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":155,"score_detail":156,"sources":161,"tags":163,"primary_topic":166,"secondary_topics":167,"classification":168,"search_phrases":188,"slug":191,"view_count":68,"doi":192,"paper":193,"created_at":221},1208,"Restoration of occlusion in residual film images and curling degree prediction based on UAV in pre-sowing cotton fields","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.compag.2026.112335","Restoration of occlusion in residual film images and curling degree prediction based on UAV in pre-sowing cotton fields。Computers and Electronics in Agriculture","Computers and Electronics in Agriculture","2026-08-25T00:00:00Z",62,{"impact":157,"substance":158,"depth":19,"authority":159,"freshness":55,"relevant":22,"comment":160},12,18,14,"基于无人机影像的残膜污染检测与卷曲度预测研究，方法新颖，对农业环保有实际价值，但时效性较低。",[162],{"name":153,"url":151},[27,164,104,28,165],"农业人工智能","残膜污染",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[],{"status":40,"locked":14,"version":41,"method":42,"evidence":169},{"candidates":170,"agricultural_evidence":47},[171,175,179,184],{"code":33,"score":157,"supported":47,"title":172,"abstract":173,"tags":174},[119],[119],[104],{"code":60,"score":22,"supported":14,"title":176,"abstract":177,"tags":178},[],[],[27],{"code":180,"score":22,"supported":14,"title":181,"abstract":182,"tags":183},"I2",[],[],[164],{"code":54,"score":22,"supported":14,"title":185,"abstract":186,"tags":187},[],[],[28],[189,190],"农业人工智能 智慧农业 残膜污染 棉花","农业人工智能 智慧农业","农业人工智能智慧农业残膜污染棉花-1208","10.1016\u002Fj.compag.2026.112335",{"doi":192,"openalex_id":194,"authors":195,"venue":153,"cited_by_count":77,"oa_url":9,"card":216,"direction":83,"ingested_from":85},"W7204210197",[196,199,201,204,207,210,213],{"name":197,"orcid":198},"Zhiqiang Zhai","https:\u002F\u002Forcid.org\u002F0000-0002-5242-9779",{"name":200,"orcid":9},"Lupeng Miao",{"name":202,"orcid":203},"Huting Wang","https:\u002F\u002Forcid.org\u002F0009-0008-3605-2685",{"name":205,"orcid":206},"Jiangbo Li","https:\u002F\u002Forcid.org\u002F0000-0001-7856-9131",{"name":208,"orcid":209},"Yue Chen","https:\u002F\u002Forcid.org\u002F0009-0004-5110-6400",{"name":211,"orcid":212},"Songxin Ye","https:\u002F\u002Forcid.org\u002F0000-0002-8895-8525",{"name":214,"orcid":215},"Ruoyu Zhang","https:\u002F\u002Forcid.org\u002F0000-0003-0370-3249",{"tldr":217,"method":218,"finding":219,"direction":83,"opportunity":220},"用无人机图像修复残膜遮挡并预测卷曲程度，评估棉田残膜污染。","无人机图像、遮挡修复、卷曲度预测模型。","成功修复遮挡并预测卷曲度，支持残膜污染评估。","可扩展至不同作物、不同残膜类型，结合深度学习提升修复精度和预测鲁棒性。","2026-09-01T04:03:02.336224Z",{"id":223,"title":224,"url":225,"summary":226,"summary_zh":9,"content":9,"source_name":227,"source_url":9,"published_at":228,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":229,"score_detail":230,"sources":234,"tags":236,"primary_topic":240,"secondary_topics":241,"classification":246,"search_phrases":267,"slug":270,"view_count":77,"doi":9,"paper":271,"created_at":279},5532,"精准棉花病害管理：融合无人机影像与定制YOLO深度学习框架","https:\u002F\u002Farxiv.org\u002Fabs\u002F2610.24182","研究提出单阶段全卷积检测器Cotton-Vision，融合幽灵瓶颈通道注意力与跨阶段部分融合三项轻量化设计，参数减少20%。在巴基斯坦旁遮普省7个棉田采集的2700余幅高分辨率无人机影像上测试，精确率97.7%、召回率77.7%、mAP@50为85.6%，较主流基线提升10至20个百分点。预印本，仅在巴基斯坦单一区域验证。","arXiv 预印本 2026-10-16","2026-10-16T00:00:00Z",69,{"impact":19,"substance":231,"depth":232,"authority":21,"freshness":46,"relevant":22,"comment":233},21,17,"方法轻量化与实测数据有增量，但仅单区域预印本验证，属细分技术进展，可入精选备选。",[235],{"name":227,"url":225},[27,164,237,238,239],"目标检测","无人机遥感","棉花病害",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[242],{"code":180,"slug":243,"name":244,"path":245,"parent_code":112},"agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai",{"status":40,"locked":14,"version":41,"method":42,"evidence":247},{"candidates":248,"agricultural_evidence":47},[249,255,259,263],{"code":180,"score":158,"supported":47,"title":250,"abstract":253,"tags":254},[251,252],"深度学习","YOLO",[],[164,237],{"code":33,"score":123,"supported":47,"title":256,"abstract":257,"tags":258},[104],[119],[],{"code":60,"score":22,"supported":14,"title":260,"abstract":261,"tags":262},[],[],[27],{"code":54,"score":22,"supported":14,"title":264,"abstract":265,"tags":266},[],[],[238],[268,269],"Cotton-Vision 棉花病害","无人机影像 棉花 检测","Cotton-Vision棉花病害-5532",{"doi":9,"openalex_id":9,"authors":272,"venue":9,"cited_by_count":77,"oa_url":9,"card":273,"direction":83,"ingested_from":278},[],{"tldr":274,"method":275,"finding":276,"direction":83,"opportunity":277},"提出轻量化检测器Cotton-Vision，用无人机影像精准识别棉花病害。","基于YOLO改进，融合幽灵瓶颈通道注意力与跨阶段部分融合，用2700余幅无人机影","精确率97.7%、mAP@50达85.6%，较主流基线提升10至20个百分点。","仅在巴基斯坦单一区域验证，可扩展至多区域多品种，并探索边缘设备实时部署。","agent","2026-10-22T00:07:09.816790Z",{"id":281,"title":282,"url":283,"summary":284,"summary_zh":285,"content":9,"source_name":286,"source_url":283,"published_at":287,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":229,"score_detail":288,"sources":290,"tags":292,"primary_topic":295,"secondary_topics":296,"classification":297,"search_phrases":314,"slug":317,"view_count":22,"doi":318,"paper":319,"created_at":335},5234,"A smart agriculture-oriented cross-attention multiscale CNN for sugarcane disease diagnosis","https:\u002F\u002Fdoi.org\u002F10.1080\u002F23080477.2026.2749739","Sugarcane leaf diseases can adversely affect crop health and yield, making timely disease identification important for smart agriculture. Manual inspection of leaf diseases is time-consuming and requires expert knowledge, particularly when large cultivation areas need to be monitored. To tackle this problem, this study proposes a Cross-Attention Multiscale Parallel Convolutional Neural Network for automatic classification of sugarcane leaf diseases. This framework comprises a Parallel Feature Extraction Stem, cross-attention mechanism, Dense Connection Blocks, and Inception Blocks to learn disease features at various spatial scales. The PFES extracts fine and broad disease patterns using parallel convolutional branches, while the cross-attention mechanism refines the extracted features by emphasizing disease-related information. The DCB supports feature propagation and reuse, and the IB further captures multi-scale feature representations. The proposed model was tested on two benchmark datasets of leaf diseases for sugarcane plants. The proposed framework obtained an accuracy (98.79%), precision (98.31%), recall (97.81%), specificity (98.60%), and F1-Score (98.05%). For assessing the classification accuracy and consistency of the framework, the comparative study, ablation studies, and five-fold cross-validation are carried out. The results show that CA-MPCNN provides accurate sugarcane leaf disease classification and can support automated disease diagnosis in smart agriculture applications.","甘蔗叶片病害会对作物健康与产量造成不利影响，因此及时识别病害对智慧农业具有重要意义。人工检查叶片病害耗时且需要专业知识，尤其是在需要监测大面积种植区域时。为解决该问题，本研究提出一种用于甘蔗叶片病害自动分类的交叉注意力多尺度并行卷积神经网络（Cross-Attention Multiscale Parallel Convolutional Neural Network）。该框架由并行特征提取主干（Parallel Feature Extraction Stem）、交叉注意力机制、密集连接块（Dense Connection Blocks）和Inception块组成，用于学习不同空间尺度上的病害特征。PFES通过并行卷积分支提取细粒度与广域病害模式，而交叉注意力机制通过强调病害相关信息对提取的特征进行细化。DCB支持特征传播与复用，IB进一步捕获多尺度特征表示。所提出的模型在两个甘蔗植株叶片病害基准数据集上进行了测试。该框架取得了准确率（98.79%）、精确率（98.31%）、召回率（97.81%）、特异性（98.60%）和F1分数（98.05%）。为评估该框架的分类准确性和一致性，开展了对比研究、消融研究和五折交叉验证。结果表明，CA-MPCNN能够实现准确的甘蔗叶片病害分类，并可支持智慧农业应用中的自动化病害诊断。","Smart Science","2026-10-17T00:00:00Z",{"impact":157,"substance":18,"depth":19,"authority":157,"freshness":21,"relevant":22,"comment":289},"该研究提出一种面向甘蔗叶部病害自动分类的交叉注意力多尺度CNN，在两个基准数据集上准确率达98.79%，方法新颖、实验充分，但属于细分作物应用研究，产业影响有限，时效性强，值得进入每日精选。",[291],{"name":286,"url":283},[27,164,293,251,294],"甘蔗","病害诊断",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[],{"status":40,"locked":14,"version":41,"method":42,"evidence":298},{"candidates":299,"agricultural_evidence":47},[300,305,310],{"code":33,"score":117,"supported":47,"title":301,"abstract":303,"tags":304},[302],"sugarcane",[293,302],[293],{"code":60,"score":117,"supported":47,"title":306,"abstract":308,"tags":309},[307],"smart agriculture",[27,307],[27],{"code":180,"score":55,"supported":14,"title":311,"abstract":312,"tags":313},[],[],[164,251],[315,316],"甘蔗 叶部病害 智能诊断","CA-MPCNN 甘蔗 病害","甘蔗叶部病害智能诊断-5234","10.1080\u002F23080477.2026.2749739",{"doi":318,"openalex_id":320,"authors":321,"venue":286,"cited_by_count":77,"oa_url":9,"card":328,"direction":334,"ingested_from":85},"W7222832245",[322,324,326],{"name":323,"orcid":9},"Bala Subramanian Chokkalingam",{"name":325,"orcid":9},"Uvarani Meenrajan",{"name":327,"orcid":9},"Shanmugapriya Sankaran",{"tldr":329,"method":330,"finding":331,"direction":332,"opportunity":333},"提出交叉注意力多尺度并行CNN，实现甘蔗叶病自动分类。","并行卷积分支、交叉注意力、密集连接与Inception块，两个基准数据集验证。","模型准确率达98.79%，消融与五折交叉验证证明其有效稳定。","农业人工智能与决策模型","可探索田间复杂背景与轻量化部署，并扩展到多作物病害实时诊断。","智慧农业 \u002F 农业物联网","2026-10-18T23:30:07.777036Z",{"id":337,"title":338,"url":339,"summary":340,"summary_zh":341,"content":9,"source_name":342,"source_url":339,"published_at":287,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":343,"score_detail":344,"sources":346,"tags":348,"primary_topic":352,"secondary_topics":353,"classification":354,"search_phrases":379,"slug":382,"view_count":22,"doi":383,"paper":384,"created_at":392},5233,"Cotton pest and disease multi-class classification with weather and temporal predictors features across Indian agroclimatic zones","https:\u002F\u002Fdoi.org\u002F10.25258\u002Fijddt.16.73s.45","The Pink Bollworm (Pectinophora gossypiella) remains one of the most destructive pests affecting cotton productivity, with infestation levels strongly driven by climatic fluctuations and crop phenology.This study presents a machine learning-based framework for early multi-classification and prediction of Pink Bollworm incidence using agro-meteorological variables.The dataset includes maximum and minimum air temperature, relative humidity, rainfall, sunshine hours, evaporation, and wind speed, all of which influence Pink Bollworm development, migration, and reproductive cycles.After preprocessing, applying cyclic encoding to temporal variables, and generating lagged climatic features, multiple machine learning algorithms-including SVM, Neural Networks, KNN, Regression models, Ensemble methods, and Naïve Bayes-were trained and systematically compared.SVM model demonstrated the best test accuracy, effectively capturing dynamic climate-pest interactions.The resulting MATLAB-based classification tool enables rapid assessment and supports climate-informed pest surveillance.A stratified 80:20 train-test split was employed, with 5-fold cross-validation on the training set for hyperparameter tuning, ensuring no data leakage between partitions.The validation and training accuracies of these models are 89.4% and 96 % respectively.Overall, the model offers a practical and interpretable solution for early detection of Pink Bollworm risk, thereby contributing to informed decision-making in integrated pest management (IPM) and precision agriculture.","棉红铃虫（Pectinophora gossypiella）仍是影响棉花产量最具破坏性的害虫之一，其发生程度受气候波动和作物物候的强烈驱动。本研究提出了一种基于机器学习的框架，利用农业气象变量对棉红铃虫的发生进行早期多分类与预测。数据集包括最高和最低气温、相对湿度、降雨量、日照时数、蒸发量和风速，这些因素均影响棉红铃虫的发育、迁移和繁殖周期。经过预处理、对时间变量应用循环编码并生成滞后气候特征后，训练并系统比较了多种机器学习算法，包括支持向量机（SVM）、神经网络、K近邻（KNN）、回归模型、集成方法和朴素贝叶斯。SVM模型表现出最佳的测试准确率，有效捕捉了气候与害虫之间的动态相互作用。所构建的基于MATLAB的分类工具可实现快速评估，并支持气候信息驱动的害虫监测。采用分层80：20的训练-测试划分，并在训练集上进行5折交叉验证以调优超参数，确保各分区之间无数据泄漏。这些模型的验证准确率和训练准确率分别为89.4%和96%。总体而言，该模型为棉红铃虫风险的早期检测提供了一种实用且可解释的解决方案，从而有助于综合害虫管理（IPM）和精准农业中的科学决策。","International Journal of Drug Delivery Technology",55,{"impact":17,"substance":158,"depth":159,"authority":46,"freshness":21,"relevant":22,"comment":345},"研究用多类机器学习模型结合气象与时序特征预测棉铃虫风险，方法扎实但属细分领域应用研究，信源期刊权威性一般，公共影响有限。",[347],{"name":342,"url":339},[27,349,104,350,351],"机器学习","农业气象","病虫害预警",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[],{"status":40,"locked":14,"version":41,"method":42,"evidence":355},{"candidates":356,"agricultural_evidence":47},[357,361,368,373],{"code":33,"score":117,"supported":47,"title":358,"abstract":359,"tags":360},[119],[104,119],[104],{"code":60,"score":362,"supported":47,"title":363,"abstract":364,"tags":367},7,[],[365,366],"精准农业","precision agriculture",[27],{"code":180,"score":362,"supported":47,"title":369,"abstract":370,"tags":372},[],[349,371],"machine learning",[349],{"code":374,"score":375,"supported":47,"title":376,"abstract":377,"tags":378},"H7",4,[],[350],[350],[380,381],"棉铃虫 机器学习 预测","棉花 病虫害 气象因子","棉铃虫机器学习预测-5233","10.25258\u002Fijddt.16.73s.45",{"doi":383,"openalex_id":385,"authors":386,"venue":342,"cited_by_count":77,"oa_url":339,"card":387,"direction":334,"ingested_from":85},"W7222826491",[],{"tldr":388,"method":389,"finding":390,"direction":332,"opportunity":391},"用农业气象变量与时间特征构建机器学习模型，实现印度棉区棉铃虫多级发生预测。","气象数据加周期编码与滞后特征，比较SVM等模型，5折交叉验证。","SVM测试精度最高达89.4%，能有效刻画气候与虫害动态关系。","可引入遥感植被指数与实时气象物联网数据，构建跨区域可迁移的病虫害预警系统。","2026-10-18T23:30:07.725427Z",{"id":394,"title":395,"url":396,"summary":397,"summary_zh":398,"content":9,"source_name":399,"source_url":396,"published_at":228,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":400,"score_detail":401,"sources":403,"tags":405,"primary_topic":407,"secondary_topics":408,"classification":410,"search_phrases":426,"slug":429,"view_count":22,"doi":430,"paper":431,"created_at":465},5150,"MSFANet: a multi-scale feature aggregation network for tobacco pest recognition","https:\u002F\u002Fdoi.org\u002F10.3389\u002Ffpls.2026.1871587","Introduction Tobacco is a globally significant economic crop, and pest infestation represents one of the primary factors contributing to declines in both yield and quality. However, tobacco pests encompass a wide variety of species that differ substantially in morphology and scale. Field-captured images are frequently subject to complex background interference and class imbalance, presenting considerable challenges for automated recognition. Existing deep learning approaches remain limited in their capacity for multi-scale feature extraction and discriminative feature representation under such conditions. To address these challenges, this study proposes a Multi-Scale Feature Aggregation Network (MSFANet) for tobacco pest recognition, formulated here as a single-label image classification task. Methods To address scale variation and complex field backgrounds, two task-oriented modules were developed. The Multi-Scale Pyramid Pooling (MSPP) module aggregates contextual information across multiple receptive fields through parallel feature extraction and adaptive cross-branch fusion. The Hierarchical Feature Fusion Block (HFFB) subsequently refines the feature representations through spatial gating and partial convolution. An auxiliary classification head and Focal Loss were incorporated during training to provide intermediate supervision and address class imbalance, respectively. Results The effectiveness of MSFANet was evaluated on a dataset comprising 13 tobacco-associated pest and natural-enemy categories. The model achieved accuracy, precision, recall, and F1-score of 97.63%, 97.89%, 97.59%, and 97.71%, respectively, with a parameter count of 1.21 M. Among the models trained from random initialization, MSFANet shared the highest accuracy while using the fewest parameters. Discussion By combining multi-scale contextual aggregation with hierarchical feature refinement, MSFANet achieved accurate classification across the 13 tobacco-associated categories under complex field conditions. The proposed method provides effective technical support for automated and precision-oriented pest monitoring in tobacco fields, offering practical value for advancing precise pest management and the broader development of smart agriculture.","引言 烟草是全球重要的经济作物，虫害侵染是导致其产量和品质下降的主要因素之一。然而，烟草害虫种类繁多，形态和尺度差异显著。田间采集的图像常受到复杂背景干扰和类别不平衡的影响，给自动化识别带来了相当大的挑战。现有深度学习方法在此类条件下进行多尺度特征提取和判别性特征表示的能力仍然有限。为应对这些挑战，本研究提出了一种用于烟草害虫识别的多尺度特征聚合网络（Multi-Scale Feature Aggregation Network，MSFANet），并将其形式化为单标签图像分类任务。方法 针对尺度变化和复杂田间背景，开发了两个面向任务的模块。多尺度金字塔池化（Multi-Scale Pyramid Pooling，MSPP）模块通过并行特征提取和自适应跨分支融合，聚合多个感受野的上下文信息。分层特征融合模块（Hierarchical Feature Fusion Block，HFFB）随后通过空间门控和部分卷积对特征表示进行细化。训练过程中引入辅助分类头和Focal Loss，分别提供中间监督和处理类别不平衡。结果 在包含13个烟草相关害虫及天敌类别的数据集上评估了MSFANet的有效性。模型的准确率、精确率、召回率和F1分数分别为97.63%、97.89%、97.59%和97.71%，参数量为1.21 M。在随机初始化训练的模型中，MSFANet以最少的参数量取得了最高准确率。讨论 通过结合多尺度上下文聚合与分层特征细化，MSFANet在复杂田间条件下实现了对13个烟草相关类别的准确分类。所提方法为烟草田间害虫的自动化和精准监测提供了有效的技术支持，对推进精准害虫管理和智慧农业的更广泛发展具有实用价值。","Frontiers in Plant Science",76,{"impact":19,"substance":231,"depth":232,"authority":20,"freshness":21,"relevant":22,"comment":402},"该论文提出轻量化多尺度特征聚合网络，在烟草害虫识别上取得97%以上精度且参数量仅1.21M，方法新颖、数据扎实，对智慧农业虫情监测有实用价值，值得进入每日精选。",[404],{"name":399,"url":396},[27,164,251,406,50],"病虫害识别",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[409],{"code":180,"slug":243,"name":244,"path":245,"parent_code":112},{"status":40,"locked":14,"version":41,"method":42,"evidence":411},{"candidates":412,"agricultural_evidence":47},[413,417,422],{"code":33,"score":117,"supported":47,"title":414,"abstract":415,"tags":416},[51],[50,51],[50],{"code":180,"score":21,"supported":47,"title":418,"abstract":419,"tags":421},[],[251,420],"deep learning",[164,251,406],{"code":60,"score":362,"supported":47,"title":423,"abstract":424,"tags":425},[],[27,307],[27],[427,428],"MSFANet 烟草害虫识别","多尺度特征聚合 害虫识别","MSFANet烟草害虫识别-5150","10.3389\u002Ffpls.2026.1871587",{"doi":430,"openalex_id":432,"authors":433,"venue":399,"cited_by_count":77,"oa_url":459,"card":460,"direction":334,"ingested_from":85},"W7222603139",[434,436,438,440,442,444,446,448,451,454,457],{"name":435,"orcid":9},"Shun He",{"name":437,"orcid":9},"Sifan Li",{"name":439,"orcid":9},"Xinjian Li",{"name":441,"orcid":9},"Qianqian Luo",{"name":443,"orcid":9},"Dong Sufen",{"name":445,"orcid":9},"Lu Lei",{"name":447,"orcid":9},"Xingjiang Chen",{"name":449,"orcid":450},"Liuti Cai","https:\u002F\u002Forcid.org\u002F0000-0002-9151-8002",{"name":452,"orcid":453},"Jian‐Yu Meng","https:\u002F\u002Forcid.org\u002F0000-0002-5796-1086",{"name":455,"orcid":456},"Ning Lu","https:\u002F\u002Forcid.org\u002F0000-0003-0214-5038",{"name":458,"orcid":9},"Hancheng Wang","https:\u002F\u002Fwww.frontiersin.org\u002Fjournals\u002Fplant-science\u002Farticles\u002F10.3389\u002Ffpls.2026.1871587\u002Fpdf",{"tldr":461,"method":462,"finding":463,"direction":332,"opportunity":464},"提出MSFANet多尺度特征聚合网络，实现复杂田间背景下烟草害虫的自动识别。","多尺度金字塔池化MSPP、分层特征融合HFFB、辅助分类头与Focal Loss","模型准确率97.63%、参数量仅1.21M，在随机初始化模型中精度最高且参数最少。","可探索轻量多尺度网络在边缘设备部署及跨作物害虫迁移识别，结合类不平衡与复杂背景的鲁棒性研究。","2026-10-17T23:30:08.695455Z"]