[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-2180":3,"related-2180":131},{"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":37,"classification":38,"search_phrases":107,"slug":110,"view_count":21,"doi":111,"paper":112,"created_at":130},2180,"Crop Yield Estimation with MODIS Derived Normalized Difference Vegetation Index and Comparative Study on Crop Yield Prediction Among Linear Regression, Random Forest and Gradient Boosting as Well as CatBoost","https:\u002F\u002Fdoi.org\u002F10.3390\u002Frs18183107","This paper presents the design, development, and evaluation of a machine-learning system built to forecast agricultural crop yields across Indian states between 2000 and 2026, together with a complementary, national-scale verification of predicted crop yield using a MODIS-derived NDVI time series (MOD13A3.061 Vegetation Indices Monthly L3 Global 1 km SIN Grid). Although many prior studies address crop-yield prediction with linear regression, random forest, gradient boosting, and related methods, a complementary, aggregate-level verification method for predicted crop yield has rarely been proposed. This article contributes such a method, together with a complementary NDVI-based estimation approach for total foodgrain output. Crop yield and MODIS-derived NDVI are strongly correlated (r = 0.84 for annual maximum NDVI; r = 0.78 for annual mean NDVI), and a simple regression of total foodgrains on annual maximum NDVI alone reaches R2 = 0.70. Four modeling approaches—linear regression, random forest, gradient boosting, and CatBoost—were built and compared using a chronology-preserving, expanding-window walk-forward validation procedure with a final, untouched 2024–2026 holdout, rather than a random split; a companion leakage check confirmed that reported production is almost algebraically identical to reported yield and therefore had to be excluded from the feature set. Random forest produced the most reliable and consistent forecasts, reaching a mean absolute percentage error (MAPE) of 11.4% and R2 = 0.982 on the final holdout, ahead of CatBoost (MAPE = 11.6%, R2 = 0.969) and gradient boosting (MAPE = 13.0%, R2 = 0.908), and substantially ahead of linear regression, which failed to generalize to the holdout period (R2 = −10.67); across the walk-forward folds preceding this holdout, however, the three tree ensembles were statistically indistinguishable. A four-configuration ablation study confirms that most of this performance gain is attributable to the inclusion of MODIS-derived NDVI rather than to model choice alone. Prediction error varies considerably by crop, from under 10% MAPE for major staples (rice, wheat, maize, sugarcane, moong) to well over 80% MAPE for several lower-volume crops (soyabean, garlic, Sunn hemp, tobacco, potato). The paper also documents two consequential data-quality findings—a near-perfect algebraic relationship between production and yield, and a structural administrative reporting gap in 2020—and closes with directions for future work, including higher-resolution satellite inputs, temporal deep-learning architectures, additional environmental covariates, and explainable-AI analysis of feature contributions.","本文介绍了一个机器学习系统的设计、开发与评估，该系统用于预测2000年至2026年间印度各邦的农作物产量，并辅以一项基于MODIS衍生的NDVI时间序列（MOD13A3.061植被指数月度L3全球1 km SIN网格）在全国尺度上对预测作物产量的验证。尽管此前已有许多研究采用线性回归、随机森林、梯度提升及相关方法进行作物产量预测，但针对预测作物产量的补充性、聚合层面的验证方法却鲜有提出。本文提出了这样一种方法，并辅以一种基于NDVI的粮食总产量估算方法。作物产量与MODIS衍生的NDVI高度相关（年最大NDVI的r = 0.84；年均NDVI的r = 0.78），仅以年最大NDVI对粮食总产量进行简单回归即可达到R² = 0.70。本文构建了四种建模方法——线性回归、随机森林、梯度提升和CatBoost——并采用保持时间顺序的扩展窗口前向验证程序进行比较，最终以2024—2026年作为未触碰的留出集，而非随机划分；一项配套的泄漏检查证实，报告产量与报告单产在代数上几乎完全相同，因此必须将其从特征集中排除。随机森林产生了最可靠且一致的预测，在最终留出集上达到平均绝对百分比误差（MAPE）为11.4%、R² = 0.982，优于CatBoost（MAPE = 11.6%，R² = 0.969）和梯度提升（MAPE = 13.0%，R² = 0.908），并大幅优于线性回归，后者未能泛化至留出期（R² = −10.67）；然而在此留出集之前的各前向验证折中，三种树集成方法在统计上无法区分。一项四配置消融研究证实，大部分性能提升归因于纳入MODIS衍生的NDVI，而非仅归因于模型选择。预测误差因作物而异，主要粮食作物（水稻、小麦、玉米、甘蔗、绿豆）的MAPE低于10%，而若干低产量作物（大豆、大蒜、菽麻、烟草、马铃薯）的MAPE则远超80%。本文还记录了两项重要的数据质量发现——产量与单产之间近乎完美的代数关系，以及2020年结构性的行政报告缺口——并在结尾提出了未来研究方向。",null,"Remote Sensing","2026-09-10T00:00:00Z","论文",10,false,81,{"impact":17,"substance":18,"depth":17,"authority":19,"freshness":20,"relevant":21,"comment":22},18,23,14,8,1,"基于MODIS NDVI与多种机器学习模型的作物产量预测研究，方法严谨、结论可靠，对农业遥感估产有实质参考价值。",[24],{"name":10,"url":6},[26,27,28,29,30],"农业人工智能","粮食安全","机器学习","遥感","作物估产",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},"A6","cash-crops","棉麻糖与特色经作","\u002Ftopic\u002Ffield-crops\u002Fcash-crops","A",[],{"status":39,"locked":14,"version":40,"method":41,"evidence":42},"classified","2026-10-22.2","rules",{"candidates":43,"agricultural_evidence":46},[44,54,62,69,76,83,89,96,102],{"code":32,"score":45,"supported":46,"title":47,"abstract":48,"tags":53},9,true,[],[49,50,51,52],"甘蔗","烟草","sugarcane","tobacco",[],{"code":55,"score":56,"supported":46,"title":57,"abstract":58,"tags":61},"A1",6,[],[59,60],"水稻","rice",[],{"code":63,"score":56,"supported":46,"title":64,"abstract":65,"tags":68},"A2",[],[66,67],"小麦","wheat",[],{"code":70,"score":56,"supported":46,"title":71,"abstract":72,"tags":75},"A3",[],[73,74],"玉米","maize",[],{"code":77,"score":56,"supported":46,"title":78,"abstract":79,"tags":82},"A5",[],[80,81],"马铃薯","potato",[],{"code":84,"score":85,"supported":46,"title":86,"abstract":87,"tags":88},"I2",5,[],[28],[26,28],{"code":90,"score":91,"supported":46,"title":92,"abstract":93,"tags":95},"A4",3,[],[94],"大豆",[],{"code":97,"score":98,"supported":14,"title":99,"abstract":100,"tags":101},"I3",2,[],[],[30,29],{"code":103,"score":21,"supported":14,"title":104,"abstract":105,"tags":106},"A8",[],[],[27],[108,109],"农业人工智能 作物估产 机器学习 粮食安全","农业人工智能 作物估产","农业人工智能作物估产机器学习粮食安全-2180","10.3390\u002Frs18183107",{"doi":111,"openalex_id":113,"authors":114,"venue":10,"cited_by_count":121,"oa_url":6,"card":122,"direction":128,"ingested_from":129},"W7212191753",[115,118],{"name":116,"orcid":117},"Kohei Arai","https:\u002F\u002Forcid.org\u002F0009-0001-6433-1592",{"name":119,"orcid":120},"Sara Sanwal","https:\u002F\u002Forcid.org\u002F0009-0003-7944-0948",0,{"tldr":123,"method":124,"finding":125,"direction":126,"opportunity":127},"用MODIS NDVI与四种回归模型预测印度各邦作物产量，并做全国尺度验证。","MODIS NDVI时序、线性回归、随机森林、梯度提升、CatBoost及前向验","随机森林最优（MAPE 11.4%），NDVI贡献主要增益，主粮误差低而小作物误差高。","农业遥感与作物表型","可探索多源遥感与深度模型融合，并针对小作物和行政数据缺口改进预测。","农业人工智能与决策模型","openalex","2026-09-11T23:30:52.943993Z",{"total":56,"page":21,"page_size":56,"items":132},[133,198,269,328,382,438],{"id":134,"title":135,"url":136,"summary":137,"summary_zh":138,"content":9,"source_name":139,"source_url":136,"published_at":140,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":141,"score_detail":142,"sources":146,"tags":148,"primary_topic":151,"secondary_topics":152,"classification":158,"search_phrases":176,"slug":179,"view_count":85,"doi":180,"paper":181,"created_at":197},3179,"Multi-features and Fuzzy with Residual Lenet for Cotton Classification Using Remote Sensing Image","https:\u002F\u002Fdoi.org\u002F10.1007\u002Fs12524-026-02541-8","Multi-features and Fuzzy with Residual Lenet for Cotton Classification Using Remote Sensing Image。Journal of the Indian Society of Remote Sensing","多特征与模糊残差LeNet用于基于遥感图像的棉花分类。《印度遥感学会杂志》","Journal of the Indian Society of Remote Sensing","2026-09-21T00:00:00Z",58,{"impact":20,"substance":19,"depth":143,"authority":144,"freshness":45,"relevant":21,"comment":145},15,12,"方法类论文，提出多特征与模糊结合残差LeNet用于遥感棉花分类，专业价值尚可但影响面窄，适合作为技术参考而非每日精选头条。",[147],{"name":139,"url":136},[26,149,29,150],"棉花","作物分类",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[153],{"code":97,"slug":154,"name":155,"path":156,"parent_code":157},"remote-sensing","遥感与农情监测","\u002Ftopic\u002Fdigital-agriculture\u002Fremote-sensing","I",{"status":39,"locked":14,"version":40,"method":41,"evidence":159},{"candidates":160,"agricultural_evidence":46},[161,167,172],{"code":97,"score":162,"supported":46,"title":163,"abstract":165,"tags":166},16,[164],"remote sensing",[29,164],[150,29],{"code":32,"score":143,"supported":46,"title":168,"abstract":170,"tags":171},[169],"cotton",[149,169],[149],{"code":84,"score":21,"supported":14,"title":173,"abstract":174,"tags":175},[],[],[26],[177,178],"Residual LeNet 棉花 遥感分类","农业人工智能 作物分类 棉花 遥感","ResidualLeNet棉花遥感分类-3179","10.1007\u002Fs12524-026-02541-8",{"doi":180,"openalex_id":182,"authors":183,"venue":139,"cited_by_count":121,"oa_url":9,"card":192,"direction":126,"ingested_from":129},"W7213944040",[184,186,188,190],{"name":185,"orcid":9},"G. Suresh",{"name":187,"orcid":9},"G. Bhuvaneswari",{"name":189,"orcid":9},"G. Manikandan",{"name":191,"orcid":9},"M. Robinson Joel",{"tldr":193,"method":194,"finding":195,"direction":126,"opportunity":196},"提出多特征与模糊结合残差LeNet的遥感影像棉花分类方法。","多特征提取、模糊逻辑与残差LeNet网络，用于遥感图像分类。","该方法能有效提升棉花分类精度，验证了深度残差网络结合模糊特征的可行性。","可探索多特征与模糊深度学习在其他作物遥感分类中的泛化能力及轻量化部署。","2026-09-22T23:30:23.811658Z",{"id":199,"title":200,"url":201,"summary":202,"summary_zh":9,"content":9,"source_name":203,"source_url":201,"published_at":204,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":205,"score_detail":206,"sources":208,"tags":210,"primary_topic":213,"secondary_topics":214,"classification":215,"search_phrases":235,"slug":238,"view_count":91,"doi":239,"paper":240,"created_at":268},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":144,"substance":17,"depth":162,"authority":19,"freshness":98,"relevant":21,"comment":207},"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。",[209],{"name":203,"url":201},[211,26,149,29,212],"智慧农业","残膜污染",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[],{"status":39,"locked":14,"version":40,"method":41,"evidence":216},{"candidates":217,"agricultural_evidence":46},[218,222,227,231],{"code":32,"score":144,"supported":46,"title":219,"abstract":220,"tags":221},[169],[169],[149],{"code":223,"score":21,"supported":14,"title":224,"abstract":225,"tags":226},"I1",[],[],[211],{"code":84,"score":21,"supported":14,"title":228,"abstract":229,"tags":230},[],[],[26],{"code":97,"score":21,"supported":14,"title":232,"abstract":233,"tags":234},[],[],[29],[236,237],"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":239,"openalex_id":241,"authors":242,"venue":203,"cited_by_count":121,"oa_url":9,"card":263,"direction":126,"ingested_from":129},"W7204210197",[243,246,248,251,254,257,260],{"name":244,"orcid":245},"Zhiqiang Zhai","https:\u002F\u002Forcid.org\u002F0000-0002-5242-9779",{"name":247,"orcid":9},"Lupeng Miao",{"name":249,"orcid":250},"Huting Wang","https:\u002F\u002Forcid.org\u002F0009-0008-3605-2685",{"name":252,"orcid":253},"Jiangbo Li","https:\u002F\u002Forcid.org\u002F0000-0001-7856-9131",{"name":255,"orcid":256},"Yue Chen","https:\u002F\u002Forcid.org\u002F0009-0004-5110-6400",{"name":258,"orcid":259},"Songxin Ye","https:\u002F\u002Forcid.org\u002F0000-0002-8895-8525",{"name":261,"orcid":262},"Ruoyu Zhang","https:\u002F\u002Forcid.org\u002F0000-0003-0370-3249",{"tldr":264,"method":265,"finding":266,"direction":126,"opportunity":267},"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":270,"title":271,"url":272,"summary":273,"summary_zh":9,"content":9,"source_name":274,"source_url":9,"published_at":275,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":276,"score_detail":277,"sources":281,"tags":283,"primary_topic":287,"secondary_topics":288,"classification":293,"search_phrases":315,"slug":318,"view_count":121,"doi":9,"paper":319,"created_at":327},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":162,"substance":278,"depth":279,"authority":45,"freshness":56,"relevant":21,"comment":280},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。",[282],{"name":274,"url":272},[211,26,284,285,286],"目标检测","无人机遥感","棉花病害",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[289],{"code":84,"slug":290,"name":291,"path":292,"parent_code":157},"agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai",{"status":39,"locked":14,"version":40,"method":41,"evidence":294},{"candidates":295,"agricultural_evidence":46},[296,302,307,311],{"code":84,"score":17,"supported":46,"title":297,"abstract":300,"tags":301},[298,299],"深度学习","YOLO",[],[26,284],{"code":32,"score":303,"supported":46,"title":304,"abstract":305,"tags":306},11,[149],[169],[],{"code":223,"score":21,"supported":14,"title":308,"abstract":309,"tags":310},[],[],[211],{"code":97,"score":21,"supported":14,"title":312,"abstract":313,"tags":314},[],[],[285],[316,317],"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":320,"venue":9,"cited_by_count":121,"oa_url":9,"card":321,"direction":126,"ingested_from":326},[],{"tldr":322,"method":323,"finding":324,"direction":126,"opportunity":325},"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":329,"title":330,"url":331,"summary":332,"summary_zh":333,"content":9,"source_name":334,"source_url":331,"published_at":335,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":276,"score_detail":336,"sources":339,"tags":341,"primary_topic":343,"secondary_topics":344,"classification":345,"search_phrases":361,"slug":364,"view_count":21,"doi":365,"paper":366,"created_at":381},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":144,"substance":337,"depth":162,"authority":144,"freshness":45,"relevant":21,"comment":338},20,"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。",[340],{"name":334,"url":331},[211,26,49,298,342],"病害诊断",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[],{"status":39,"locked":14,"version":40,"method":41,"evidence":346},{"candidates":347,"agricultural_evidence":46},[348,352,357],{"code":32,"score":143,"supported":46,"title":349,"abstract":350,"tags":351},[51],[49,51],[49],{"code":223,"score":143,"supported":46,"title":353,"abstract":355,"tags":356},[354],"smart agriculture",[211,354],[211],{"code":84,"score":98,"supported":14,"title":358,"abstract":359,"tags":360},[],[],[26,298],[362,363],"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":365,"openalex_id":367,"authors":368,"venue":334,"cited_by_count":121,"oa_url":9,"card":375,"direction":380,"ingested_from":129},"W7222832245",[369,371,373],{"name":370,"orcid":9},"Bala Subramanian Chokkalingam",{"name":372,"orcid":9},"Uvarani Meenrajan",{"name":374,"orcid":9},"Shanmugapriya Sankaran",{"tldr":376,"method":377,"finding":378,"direction":128,"opportunity":379},"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。","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":383,"title":384,"url":385,"summary":386,"summary_zh":387,"content":9,"source_name":388,"source_url":385,"published_at":335,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":389,"score_detail":390,"sources":392,"tags":394,"primary_topic":397,"secondary_topics":398,"classification":399,"search_phrases":424,"slug":427,"view_count":21,"doi":428,"paper":429,"created_at":437},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":20,"substance":17,"depth":19,"authority":56,"freshness":45,"relevant":21,"comment":391},"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。",[393],{"name":388,"url":385},[211,28,149,395,396],"农业气象","病虫害预警",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[],{"status":39,"locked":14,"version":40,"method":41,"evidence":400},{"candidates":401,"agricultural_evidence":46},[402,406,413,418],{"code":32,"score":143,"supported":46,"title":403,"abstract":404,"tags":405},[169],[149,169],[149],{"code":223,"score":407,"supported":46,"title":408,"abstract":409,"tags":412},7,[],[410,411],"精准农业","precision agriculture",[211],{"code":84,"score":407,"supported":46,"title":414,"abstract":415,"tags":417},[],[28,416],"machine learning",[28],{"code":419,"score":420,"supported":46,"title":421,"abstract":422,"tags":423},"H7",4,[],[395],[395],[425,426],"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":428,"openalex_id":430,"authors":431,"venue":388,"cited_by_count":121,"oa_url":385,"card":432,"direction":380,"ingested_from":129},"W7222826491",[],{"tldr":433,"method":434,"finding":435,"direction":128,"opportunity":436},"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":439,"title":440,"url":441,"summary":442,"summary_zh":443,"content":9,"source_name":444,"source_url":441,"published_at":275,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":445,"score_detail":446,"sources":449,"tags":451,"primary_topic":453,"secondary_topics":454,"classification":456,"search_phrases":472,"slug":475,"view_count":21,"doi":476,"paper":477,"created_at":511},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":162,"substance":278,"depth":279,"authority":447,"freshness":45,"relevant":21,"comment":448},13,"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。",[450],{"name":444,"url":441},[211,26,298,452,50],"病虫害识别",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[455],{"code":84,"slug":290,"name":291,"path":292,"parent_code":157},{"status":39,"locked":14,"version":40,"method":41,"evidence":457},{"candidates":458,"agricultural_evidence":46},[459,463,468],{"code":32,"score":143,"supported":46,"title":460,"abstract":461,"tags":462},[52],[50,52],[50],{"code":84,"score":45,"supported":46,"title":464,"abstract":465,"tags":467},[],[298,466],"deep learning",[26,298,452],{"code":223,"score":407,"supported":46,"title":469,"abstract":470,"tags":471},[],[211,354],[211],[473,474],"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":476,"openalex_id":478,"authors":479,"venue":444,"cited_by_count":121,"oa_url":505,"card":506,"direction":380,"ingested_from":129},"W7222603139",[480,482,484,486,488,490,492,494,497,500,503],{"name":481,"orcid":9},"Shun He",{"name":483,"orcid":9},"Sifan Li",{"name":485,"orcid":9},"Xinjian Li",{"name":487,"orcid":9},"Qianqian Luo",{"name":489,"orcid":9},"Dong Sufen",{"name":491,"orcid":9},"Lu Lei",{"name":493,"orcid":9},"Xingjiang Chen",{"name":495,"orcid":496},"Liuti Cai","https:\u002F\u002Forcid.org\u002F0000-0002-9151-8002",{"name":498,"orcid":499},"Jian‐Yu Meng","https:\u002F\u002Forcid.org\u002F0000-0002-5796-1086",{"name":501,"orcid":502},"Ning Lu","https:\u002F\u002Forcid.org\u002F0000-0003-0214-5038",{"name":504,"orcid":9},"Hancheng Wang","https:\u002F\u002Fwww.frontiersin.org\u002Fjournals\u002Fplant-science\u002Farticles\u002F10.3389\u002Ffpls.2026.1871587\u002Fpdf",{"tldr":507,"method":508,"finding":509,"direction":128,"opportunity":510},"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"]