[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-1404":3,"related-1404":88},{"id":4,"title":5,"url":6,"summary":7,"summary_zh":8,"content":8,"source_name":9,"source_url":6,"published_at":10,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":14,"score_detail":15,"sources":21,"tags":23,"primary_topic":28,"secondary_topics":34,"classification":41,"search_phrases":65,"slug":68,"view_count":69,"doi":70,"paper":71,"created_at":87},1404,"Explainable Deep Learning-Based Potato Leaf Disease Detection and Severity Assessment for Smart Agriculture in Bangladesh","https:\u002F\u002Fdoi.org\u002F10.21203\u002Frs.3.rs-10879535\u002Fv1","Explainable Deep Learning-Based Potato Leaf Disease Detection and Severity Assessment for Smart Agriculture in Bangladesh。Research Square",null,"Research Square","2026-09-01T00:00:00Z","论文",10,false,53,{"impact":16,"substance":17,"depth":17,"authority":16,"freshness":18,"relevant":19,"comment":20},8,15,7,1,"针对孟加拉国马铃薯病害的可解释深度学习检测研究，方法新颖但地域性强，影响有限。",[22],{"name":9,"url":6},[24,25,26,27],"智慧农业","农业人工智能","马铃薯","作物病害",{"code":29,"slug":30,"name":31,"path":32,"parent_code":33},"A5","tubers-grains","薯类与杂粮","\u002Ftopic\u002Ffield-crops\u002Ftubers-grains","A",[35],{"code":36,"slug":37,"name":38,"path":39,"parent_code":40},"I2","agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai","I",{"status":42,"locked":13,"version":43,"method":44,"evidence":45},"classified","2026-10-22.2","rules",{"candidates":46,"agricultural_evidence":49},[47,54,60],{"code":29,"score":48,"supported":49,"title":50,"abstract":52,"tags":53},12,true,[51],"potato",[51],[26],{"code":55,"score":48,"supported":49,"title":56,"abstract":58,"tags":59},"I1",[57],"smart agriculture",[57],[24],{"code":36,"score":48,"supported":49,"title":61,"abstract":63,"tags":64},[62],"deep learning",[62],[25],[66,67],"农业人工智能 作物病害 智慧农业 马铃薯","农业人工智能 作物病害","农业人工智能作物病害智慧农业马铃薯-1404",4,"10.21203\u002Frs.3.rs-10879535\u002Fv1",{"doi":70,"openalex_id":72,"authors":73,"venue":9,"cited_by_count":83,"oa_url":84,"card":8,"direction":85,"ingested_from":86},"W7204884709",[74,77,80],{"name":75,"orcid":76},"Kazi Abdul Mannan","https:\u002F\u002Forcid.org\u002F0000-0002-7123-132X",{"name":78,"orcid":79},"Sinha Nafiz Hasan","https:\u002F\u002Forcid.org\u002F0009-0007-8906-4586",{"name":81,"orcid":82},"Ahmed Shafin","https:\u002F\u002Forcid.org\u002F0009-0000-4162-6913",0,"https:\u002F\u002Fwww.researchsquare.com\u002Farticle\u002Frs-10879535\u002Flatest.pdf","智慧农业 \u002F 农业物联网","openalex","2026-09-02T23:30:10.900822Z",{"total":89,"page":19,"page_size":89,"items":90},6,[91,161,266,334,394,475],{"id":92,"title":93,"url":94,"summary":95,"summary_zh":96,"content":8,"source_name":97,"source_url":94,"published_at":98,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":99,"score_detail":100,"sources":105,"tags":107,"primary_topic":110,"secondary_topics":111,"classification":113,"search_phrases":131,"slug":134,"view_count":83,"doi":135,"paper":136,"created_at":160},4818,"Hybrid Deep Feature Extraction and Ensemble Machine Learning for Potato Leaf Disease Classification in Uncontrolled Field Conditions","https:\u002F\u002Fdoi.org\u002F10.15832\u002Fankutbd.1837896","Early and accurate diagnosis of potato leaf diseases is critical for preventing yield losses under real field conditions. This study proposes a hybrid deep feature–based machine learning framework that combines transfer learning and ensemble classifiers to automatically identify seven potato leaf disease classes from images. A total of 4,260 images from the publicly available Potato Leaf Disease Dataset were organized into seven categories: Bacteria, Virus, Pest, Fungi, Nematode, Phytophthora, and Healthy. Real-time image augmentation was performed using Keras ImageDataGenerator to improve strength to field variability. Deep feature vectors were extracted from three pre-trained convolutional neural networks (ResNet50, EfficientNetB0, and MobileNetV2), each followed by a lightweight dense head, and concatenated into 768-dimensional feature representations. After standardization, dimensionality reduction was applied via principal component analysis, and class imbalance in the training set was addressed using the SMOTE-ENN hybrid resampling method. On these deep features, 32 classical machine learning and ensemble models including SVMs, tree-based methods, boosting algorithms, multilayer perceptrons, and four ensemble schemes, were trained. Model performance was assessed by three-fold cross-validation on the training set and final evaluation on the held-out validation set using accuracy, precision, recall, and macro F1-score. Ensemble models outperformed individual classifiers. The best performance was obtained with the StackingClassifier_L2 model, which combines XGBoost, LightGBM, Random Forest, SVM with RBF kernel, and Extra Trees within a stacking framework. This model achieved an accuracy of 81.34% and a macro F1-score of 81.74% on the validation set. The proposed hybrid system, which integrates multi-backbone deep feature extraction with comprehensive classical and ensemble classifiers, provides a robust and practical solution for automatic potato leaf disease classification in uncontrolled field conditions. The results suggest that stacking-based ensembles on top of deep features can capture complex visual patterns more effectively than single models and may support decision-making in precision agriculture.","在马铃薯实际田间条件下，早期准确地诊断叶片病害对于防止产量损失至关重要。本研究提出了一种基于混合深度特征的机器学习框架，该框架结合迁移学习和集成分类器，从图像中自动识别七类马铃薯叶片病害。来自公开可用的马铃薯叶片病害数据集的4,260张图像被分为七个类别：细菌（Bacteria）、病毒（Virus）、害虫（Pest）、真菌（Fungi）、线虫（Nematode）、疫霉（Phytophthora）和健康（Healthy）。使用Keras ImageDataGenerator进行实时图像增强，以提高对田间变异的鲁棒性。从三个预训练的卷积神经网络（ResNet50、EfficientNetB0和MobileNetV2）中提取深度特征向量，每个网络后接一个轻量级密集头，并将其拼接为768维特征表示。经过标准化后，通过主成分分析进行降维，并使用SMOTE-ENN混合重采样方法解决训练集中的类别不平衡问题。在这些深度特征上，训练了32种经典机器学习和集成模型，包括支持向量机、基于树的方法、提升算法、多层感知机以及四种集成方案。通过训练集上的三折交叉验证和保留验证集上的最终评估，使用准确率、精确率、召回率和宏F1分数评估模型性能。集成模型优于单个分类器。最佳性能由StackingClassifier_L2模型获得，该模型在堆叠框架内结合了XGBoost、LightGBM、随机森林、带RBF核的支持向量机和极端随机树。该模型在验证集上达到了81.34%的准确率和81.74%的宏F1分数。所提出的混合系统将多骨干深度特征提取与全面的经典分类器和集成分类器相结合，为无控制田间条件下的马铃薯叶片病害自动分类提供了一种稳健且实用的解决方案。结果表明，基于深度特征的堆叠集成能够比单一模型更有效地捕捉复杂的视觉模式，并可能支持精准农业中的决策。","Tarım Bilimleri Dergisi","2026-10-12T00:00:00Z",65,{"impact":16,"substance":101,"depth":102,"authority":103,"freshness":16,"relevant":19,"comment":104},20,16,13,"方法组合较全面、数据规模明确，但验证集准确率仅81.34%，属细分领域技术进展，可作主题聚合素材而非头条精选。",[106],{"name":97,"url":94},[24,25,26,108,109],"病虫害识别","精准农业",{"code":29,"slug":30,"name":31,"path":32,"parent_code":33},[112],{"code":36,"slug":37,"name":38,"path":39,"parent_code":40},{"status":42,"locked":13,"version":43,"method":44,"evidence":114},{"candidates":115,"agricultural_evidence":49},[116,122,126],{"code":36,"score":102,"supported":49,"title":117,"abstract":119,"tags":121},[118],"machine learning",[120,118],"机器学习",[25,108],{"code":29,"score":17,"supported":49,"title":123,"abstract":124,"tags":125},[51],[26,51],[26],{"code":55,"score":16,"supported":49,"title":127,"abstract":128,"tags":130},[],[109,129],"precision agriculture",[24,109],[132,133],"马铃薯 叶片病害 深度学习","Potato Leaf Disease Dataset 集成学习","马铃薯叶片病害深度学习-4818","10.15832\u002Fankutbd.1837896",{"doi":135,"openalex_id":137,"authors":138,"venue":97,"cited_by_count":83,"oa_url":153,"card":154,"direction":85,"ingested_from":86},"W7221546496",[139,142,144,147,150],{"name":140,"orcid":141},"Cihan Ünal","https:\u002F\u002Forcid.org\u002F0000-0002-5255-4078",{"name":143,"orcid":8},"Cengizhan Tumer",{"name":145,"orcid":146},"Elham Tahsin Yasin","https:\u002F\u002Forcid.org\u002F0000-0003-3246-6000",{"name":148,"orcid":149},"İlkay Çınar","https:\u002F\u002Forcid.org\u002F0000-0003-0611-3316",{"name":151,"orcid":152},"Murat Köklü","https:\u002F\u002Forcid.org\u002F0000-0002-2737-2360","https:\u002F\u002Fdergipark.org.tr\u002Fen\u002Fdownload\u002Farticle-file\u002F5487015",{"tldr":155,"method":156,"finding":157,"direction":158,"opportunity":159},"融合三种CNN深度特征与集成学习，实现田间马铃薯叶片七类病害分类。","ResNet50\u002FEfficientNetB0\u002FMobileNetV2特征拼接+","StackingClassifier_L2最优，验证集准确率81.34%、宏F1为81.74%。","农业人工智能与决策模型","可探索轻量化模型与多模态数据融合，提升非受控田间小样本病害识别鲁棒性。","2026-10-13T23:30:16.021066Z",{"id":162,"title":163,"url":164,"summary":165,"summary_zh":166,"content":8,"source_name":167,"source_url":164,"published_at":168,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":169,"score_detail":170,"sources":172,"tags":174,"primary_topic":177,"secondary_topics":178,"classification":186,"search_phrases":237,"slug":240,"view_count":69,"doi":241,"paper":242,"created_at":265},3918,"UAV-based RGB imagery and supervised–unsupervised machine learning for estimating sorghum biomass and grain yield under water and nitrogen stress","https:\u002F\u002Fdoi.org\u002F10.3389\u002Ffenvs.2026.1928610","Precision agriculture emerged as a key strategy for rational management of water and nutrients amid growing food demand driven by global population expansion. Therefore, timely monitoring of field crops is crucial in maintaining adequate nutrient and water levels in the crop fields. Vegetation indices are increasingly being studied for their use in plant nutrient management and in guiding fertilizer recommendations. However, these indices were not adequately evaluated for nutrient management of crops grown with different water and nitrogen application levels. This study investigated the potential application of RGB imagery from low-cost Unmanned Aerial Vehicles (UAVs), combined with machine learning techniques, for quantifying water and nutrient stress in sorghum [ Sorghum bicolor (L.) Moench] crop performance. The field experimental setup at Prairie View A&M University, located in Southeast Texas, applied four irrigation levels (0, 75, 100, and 125% of irrigation; the 100% irrigation level is a fully irrigated treatment with no water stress) and four nitrogen application rates (0, 90, 180, and 360 kg N ha −1 ). The RGB images were captured before harvesting the grain sorghum. Image segmentation for filtering sorghum crops from soil and weeds was performed using K-means clustering and a Convolutional Neural Network (CNN). The vegetation indices, including the Green-Red Vegetation Index (GRVI), Green Leaf Index (GLI), and Excess Green Index (ExG), derived from the segmented images were used to predict sorghum grain and biomass yields using multiple linear regression, Partial least squares regression, Ridge regression, and XGBoost. The results indicated that no single model consistently outperformed the others across both segmentation approaches and response variables, suggesting that the predictive information in the RGB indices is meaningful with selected ML approaches. For example, biomass prediction using ridge regression yielded a Pearson correlation coefficient (r) of 0.65 and an RMSE of 0.72 for K-means clustering-derived indices. The models exhibited moderate, stage-specific predictive performance, with no single regression method consistently outperforming the others across both segmentation approaches and response variables. Because imagery was acquired only once during the late reproductive period, the RGB indices primarily reflected canopy condition, maturity, and senescence at that time. The findings demonstrate the feasibility of a low-cost, single-date UAV RGB workflow for late-season sorghum assessment, while emphasizing the need for multi-temporal imagery and independent validation before application to season-long stress diagnosis or operational yield forecasting.","精准农业已成为在水资源与养分合理管理中的关键策略，以应对全球人口增长所驱动的日益增长的粮食需求。因此，及时监测田间作物对于维持作物田块中适宜的养分和水分水平至关重要。植被指数在植物养分管理和指导施肥推荐中的应用正日益受到研究。然而，这些指数尚未在不同水分和氮素施用水平下对作物养分管理进行充分评估。本研究探讨了来自低成本无人机（UAV）的RGB影像结合机器学习技术在量化高粱[Sorghum bicolor (L.) Moench]作物表现中水分和养分胁迫方面的潜在应用。位于德克萨斯州东南部的Prairie View A&M大学的田间试验设置了四个灌溉水平（灌溉量的0%、75%、100%和125%；100%灌溉水平为充分灌溉处理，无水分胁迫）和四个氮素施用量（0、90、180和360 kg N ha⁻¹）。RGB图像在收获籽粒高粱前采集。使用K-means聚类和卷积神经网络（CNN）进行图像分割，以将高粱作物与土壤和杂草分离。从分割图像中提取的植被指数，包括绿红植被指数（GRVI）、绿叶指数（GLI）和过量绿指数（ExG），被用于通过多元线性回归、偏最小二乘回归、岭回归和XGBoost预测高粱籽粒和生物量产量。结果表明，在两种分割方法和响应变量下，没有单一模型始终优于其他模型，这表明RGB指数中的预测信息在所选机器学习方法下具有意义。例如，使用岭回归进行生物量预测时，基于K-means聚类所得指数的Pearson相关系数（r）为0.65，RMSE为0.72。这些模型表现出中等、特定阶段的预测性能，在两种分割方法和响应变量下，没有单一回归方法始终优于其他方法。由于影像仅在生殖后期采集一次，RGB指数主要反映当时的冠层状况、成熟度和衰老程度。研究结果证明了低成本、","Frontiers in Environmental Science","2026-09-30T00:00:00Z",69,{"impact":48,"substance":101,"depth":102,"authority":103,"freshness":16,"relevant":19,"comment":171},"低成本无人机RGB影像结合机器学习估算高粱生物量与产量的方法学研究，结论审慎，对精准农业养分水分管理有参考价值，但属单点单期试验，影响范围有限。",[173],{"name":167,"url":164},[24,25,109,175,176],"遥感","高粱",{"code":29,"slug":30,"name":31,"path":32,"parent_code":33},[179,180],{"code":36,"slug":37,"name":38,"path":39,"parent_code":40},{"code":181,"slug":182,"name":183,"path":184,"parent_code":185},"F3","fertilizers","肥料与水肥一体化","\u002Ftopic\u002Finputs-equipment\u002Ffertilizers","F",{"status":42,"locked":13,"version":43,"method":44,"evidence":187},{"candidates":188,"agricultural_evidence":49},[189,194,198,206,214,218,225,232],{"code":29,"score":17,"supported":49,"title":190,"abstract":192,"tags":193},[191],"sorghum",[176,191],[176],{"code":36,"score":17,"supported":49,"title":195,"abstract":196,"tags":197},[118],[120,118],[25],{"code":181,"score":199,"supported":49,"title":200,"abstract":201,"tags":205},9,[],[202,203,204],"施肥","养分管理","fertilizer",[],{"code":207,"score":199,"supported":49,"title":208,"abstract":209,"tags":213},"H3",[],[210,211,212],"灌溉","水资源","irrigation",[],{"code":55,"score":16,"supported":49,"title":215,"abstract":216,"tags":217},[],[109,129],[24,109],{"code":219,"score":89,"supported":49,"title":220,"abstract":221,"tags":224},"H2",[],[222,223],"土壤","soil",[],{"code":226,"score":227,"supported":49,"title":228,"abstract":229,"tags":231},"A7",3,[],[230],"间作",[],{"code":233,"score":19,"supported":13,"title":234,"abstract":235,"tags":236},"I3",[],[],[175],[238,239],"Prairie View A&M University 高粱 无人机","高粱 生物量 产量 无人机 RGB","PrairieViewAMUniversity高粱无人机-3918","10.3389\u002Ffenvs.2026.1928610",{"doi":241,"openalex_id":243,"authors":244,"venue":167,"cited_by_count":83,"oa_url":164,"card":259,"direction":158,"ingested_from":86},"W7214994063",[245,247,250,252,255,257],{"name":246,"orcid":8},"Anoop Valiya Veettil",{"name":248,"orcid":249},"Ripendra Awal","https:\u002F\u002Forcid.org\u002F0000-0002-2453-2592",{"name":251,"orcid":8},"Ali Fares",{"name":253,"orcid":254},"Binita Thapa","https:\u002F\u002Forcid.org\u002F0000-0001-7093-4915",{"name":256,"orcid":8},"Almoutaz Elhassan",{"name":258,"orcid":8},"Atikur Rahman",{"tldr":260,"method":261,"finding":262,"direction":263,"opportunity":264},"用无人机RGB影像与机器学习估算水分氮胁迫下高粱生物量和产量。","低成本无人机RGB影像，K-means与CNN分割，多种回归模型预测。","单一模型未持续最优，RGB指数在特定时期具中等预测能力。","农业遥感与作物表型","可研究多时相RGB影像与深度学习融合，提升胁迫下作物产量预测精度。","2026-10-01T23:30:54.379150Z",{"id":267,"title":268,"url":269,"summary":270,"summary_zh":271,"content":8,"source_name":272,"source_url":269,"published_at":168,"category":11,"cover_url":8,"hotness":273,"is_selected":13,"score":274,"score_detail":275,"sources":278,"tags":282,"primary_topic":286,"secondary_topics":287,"classification":289,"search_phrases":318,"slug":321,"view_count":227,"doi":322,"paper":323,"created_at":333},3913,"Lightweight Edge-AI Model for Real-Time Sorghum Disease Detection on Resource-Constrained Mobile Devices","https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23054218","Lightweight Edge-AI Model for Real-Time Sorghum Disease Detection on Resource-Constrained Mobile Devices presents an efficient artificial intelligence approach for identifying sorghum diseases directly on mobile and edge-computing platforms. The study focuses on developing a lightweight machine learning model capable of performing rapid and accurate disease detection while operating under limited computational resources, memory, and power constraints. Computer vision and deep learning techniques are used to analyse sorghum leaf images and distinguish healthy plants from diseased conditions. Model optimization strategies are considered to reduce computational complexity and improve inference speed without substantially compromising detection performance. The proposed approach supports real-time, on-device disease diagnosis, reducing dependence on cloud connectivity and enabling timely intervention in agricultural environments. The work highlights the potential of Edge AI and mobile computing to make intelligent crop monitoring more accessible, practical, and scalable, particularly for farmers and agricultural users operating in resource-constrained settings.","面向资源受限移动设备的实时高粱病害检测轻量级边缘人工智能模型提出了一种高效的人工智能方法，用于直接在移动和边缘计算平台上识别高粱病害。该研究聚焦于开发一种轻量级机器学习模型，能够在有限的计算资源、内存和功耗约束下进行快速、准确的病害检测。研究采用计算机视觉和深度学习技术分析高粱叶片图像，区分健康植株与患病状态。研究考虑了模型优化策略，以降低计算复杂度并提高推理速度，同时不过度损害检测性能。所提出的方法支持实时、设备端病害诊断，减少了对云连接的依赖，并能够在农业环境中实现及时干预。该工作凸显了边缘人工智能和移动计算在使智能作物监测更加可及、实用和可扩展方面的潜力，特别是对于在资源受限环境中作业的农民和农业用户而言。","Zenodo (CERN European Organization for Nuclear Research)",25,75,{"impact":102,"substance":101,"depth":276,"authority":103,"freshness":199,"relevant":19,"comment":277},17,"面向资源受限移动端的轻量边缘AI高粱病害实时检测研究，方法实用、契合智慧农业场景，但属细分作物技术进展，影响力有限。",[279,280],{"name":272,"url":269},{"name":272,"url":281},"https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23054219",[24,25,283,284,285],"边缘计算","作物监测","高粱病害",{"code":29,"slug":30,"name":31,"path":32,"parent_code":33},[288],{"code":36,"slug":37,"name":38,"path":39,"parent_code":40},{"status":42,"locked":13,"version":43,"method":44,"evidence":290},{"candidates":291,"agricultural_evidence":49},[292,297,305,310,314],{"code":29,"score":293,"supported":49,"title":294,"abstract":295,"tags":296},14,[191],[176,191],[],{"code":36,"score":12,"supported":49,"title":298,"abstract":299,"tags":304},[],[300,120,301,118,62,302,303],"深度学习","计算机视觉","artificial intelligence","computer vision",[25],{"code":306,"score":69,"supported":49,"title":307,"abstract":308,"tags":309},"I5",[],[283],[283],{"code":55,"score":19,"supported":13,"title":311,"abstract":312,"tags":313},[],[],[24],{"code":233,"score":19,"supported":13,"title":315,"abstract":316,"tags":317},[],[],[284],[319,320],"高粱病害 边缘AI 检测","高粱叶片 病害识别 移动端","高粱病害边缘AI检测-3913","10.5281\u002Fzenodo.23054218",{"doi":322,"openalex_id":324,"authors":325,"venue":272,"cited_by_count":83,"oa_url":269,"card":328,"direction":158,"ingested_from":86},"W7214912247",[326],{"name":327,"orcid":8},"Anantjit Publication",{"tldr":329,"method":330,"finding":331,"direction":158,"opportunity":332},"开发轻量级边缘AI模型，在资源受限移动设备上实时检测高粱病害。","计算机视觉与深度学习分析高粱叶片图像，并做模型轻量化优化。","轻量模型可在移动端实时准确识别病害，减少对云连接依赖。","可探索多作物、多病害通用轻量模型及田间复杂光照下的鲁棒性。","2026-10-01T23:30:48.902282Z",{"id":335,"title":336,"url":337,"summary":338,"summary_zh":339,"content":8,"source_name":340,"source_url":337,"published_at":168,"category":11,"cover_url":8,"hotness":341,"is_selected":13,"score":342,"score_detail":343,"sources":345,"tags":353,"primary_topic":356,"secondary_topics":357,"classification":359,"search_phrases":379,"slug":382,"view_count":227,"doi":383,"paper":384,"created_at":393},3912,"An Integrated Computer Vision and Machine Learning Framework for Sorghum Disease Severity Assessment and Yield Prediction","https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23054047","An Integrated Computer Vision and Machine Learning Framework for Sorghum Disease Severity Assessment and Yield Prediction presents a technology-driven approach for automated crop health monitoring and yield estimation in sorghum cultivation. The study integrates computer vision techniques and machine learning algorithms to identify and quantify disease severity from crop images and to predict potential yield based on relevant visual and agronomic characteristics. The framework aims to reduce the limitations of conventional manual disease assessment, which can be time-consuming and subject to human variability. Image processing and feature extraction techniques are employed to capture disease-related characteristics, while machine learning models support classification, severity estimation, and yield prediction. By combining disease assessment with yield forecasting, the proposed framework provides a comprehensive approach to precision agriculture and data-driven crop management. The work demonstrates the potential of artificial intelligence and computer vision to support timely decision-making, improve crop monitoring, and contribute to sustainable and efficient sorghum production.","一个集成的计算机视觉与机器学习框架用于高粱病害严重度评估与产量预测，提出了一种技术驱动的方法，用于高粱种植中的自动化作物健康监测与产量估算。该研究整合了计算机视觉技术和机器学习算法，从作物图像中识别和量化病害严重度，并基于相关视觉和农艺特征预测潜在产量。该框架旨在减少传统人工病害评估的局限性，传统方法可能耗时且受人为变异影响。研究采用图像处理和特征提取技术捕捉病害相关特征，同时机器学习模型支持分类、严重度估计和产量预测。通过将病害评估与产量预测相结合，所提出的框架为精准农业和数据驱动的作物管理提供了一种综合方法。该工作展示了人工智能和计算机视觉在支持及时决策、改善作物监测以及促进可持续和高效高粱生产方面的潜力。","International Journal of Emerging Technologies and Innovative Research",55,56,{"impact":48,"substance":102,"depth":293,"authority":89,"freshness":16,"relevant":19,"comment":344},"论文提出计算机视觉与机器学习结合的高粱病害严重度评估与产量预测框架，方法有一定新意但期刊权威性一般，属细分领域技术进展。",[346,347,349,351],{"name":340,"url":337},{"name":340,"url":348},"https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23054048",{"name":340,"url":350},"https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23053752",{"name":340,"url":352},"https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23053753",[24,25,301,354,176,355],"产量预测","病害识别",{"code":29,"slug":30,"name":31,"path":32,"parent_code":33},[358],{"code":36,"slug":37,"name":38,"path":39,"parent_code":40},{"status":42,"locked":13,"version":43,"method":44,"evidence":360},{"candidates":361,"agricultural_evidence":49},[362,367,371,375],{"code":36,"score":363,"supported":49,"title":364,"abstract":365,"tags":366},28,[118,303],[120,301,118,302,303],[25,355,301],{"code":29,"score":17,"supported":49,"title":368,"abstract":369,"tags":370},[191],[176,191],[176],{"code":55,"score":18,"supported":49,"title":372,"abstract":373,"tags":374},[],[109,129],[24],{"code":233,"score":19,"supported":13,"title":376,"abstract":377,"tags":378},[],[],[354],[380,381],"高粱 病害识别 产量预测","计算机视觉 作物病害评估","高粱病害识别产量预测-3912","10.5281\u002Fzenodo.23054047",{"doi":383,"openalex_id":385,"authors":386,"venue":340,"cited_by_count":83,"oa_url":337,"card":388,"direction":158,"ingested_from":86},"W7214900332",[387],{"name":327,"orcid":8},{"tldr":389,"method":390,"finding":391,"direction":158,"opportunity":392},"融合计算机视觉与机器学习，实现高粱病害严重度评估与产量预测。","图像处理与特征提取结合机器学习分类回归模型。","该框架可自动量化病害严重度并预测产量，减少人工评估偏差。","可探索多作物迁移学习与田间实时轻量化部署，提升泛化性。","2026-10-01T23:30:48.742698Z",{"id":395,"title":396,"url":397,"summary":398,"summary_zh":399,"content":8,"source_name":400,"source_url":397,"published_at":401,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":274,"score_detail":402,"sources":405,"tags":407,"primary_topic":410,"secondary_topics":411,"classification":413,"search_phrases":443,"slug":446,"view_count":19,"doi":447,"paper":448,"created_at":474},3153,"Explainable optimized deep learning and generative AI based framework for finger millet disease detection in smart agriculture","https:\u002F\u002Fdoi.org\u002F10.1007\u002Fs44163-026-02222-y","Abstract Eleusine coracana, locally known as finger millet (ragi), is a wholesome, climate-resilient crop. It is a staple food in semi-dry and dry areas of Asia and Africa. The productivity of this crop can be severely reduced by diseases such as downy, mottle, seedling, smut, and wilt. Conventional diagnosis methods are often time-consuming and unsuitable for large-scale farming. Therefore,Deep Learning (DL) offers an alternative for smart commercial farming that can streamline disease screening and improve productivity. This work offers an AI-integrated framework for early disease intervention combining DL, Explainable AI (XAI), and an image-based decision-support interface. The finger millet (ragi) dataset from Kaggle was used to train and evaluate custom CNN, VGG16, and ResNet50 models. To reduce feature redundancy and improve generalization, Grey Wolf Optimizer (GWO) based feature selection was applied to the penultimate-layer features of each model. Custom CNN, VGG16, and ResNet50 achieved accuracies of 89.40%, 83.11%, and 91.83%, respectively. Further, the use of GWO improved the accuracies to 97.71%, 90.84%, and 98.36%, respectively. ResNet50 achieved the highest accuracy 98.36% and was selected as the final backbone. The final ResNet50 model is integrated with the Gemini API to provide disease-specific recommendations based on the predicted class and user queries. To increase prediction transparency, Grad-CAM was used to highlight image regions that influenced the model output. The integration of DL, XAI, and generative AI provides a scalable approach for detecting and managing finger millet diseases. The proposed work boosts precision agriculture by providing AI-based and XAI-supported decision-making, while fostering sustainable agricultural practices.","摘要 穇子（Eleusine coracana），当地称为指黍（ragi），是一种有益健康且气候适应性强的作物。它是亚洲和非洲半干旱及干旱地区的主食。霜霉病、斑驳病、苗枯病、黑穗病和枯萎病等病害可严重降低该作物的产量。传统诊断方法往往耗时且不适合大规模种植。因此，深度学习（DL）为智能商业农业提供了一种替代方案，可简化病害筛查并提高生产力。本研究提出了一种人工智能集成框架，用于早期病害干预，结合了深度学习、可解释人工智能（XAI）和基于图像的决策支持界面。使用来自Kaggle的指黍（ragi）数据集训练和评估了自定义CNN、VGG16和ResNet50模型。为减少特征冗余并提高泛化能力，将灰狼优化器（GWO）基于特征选择应用于每个模型的倒数第二层特征。自定义CNN、VGG16和ResNet50分别达到了89.40%、83.11%和91.83%的准确率。此外，使用GWO将准确率分别提高到97.71%、90.84%和98.36%。ResNet50达到了最高准确率98.36%，并被选为最终骨干网络。最终的ResNet50模型与Gemini API集成，根据预测类别和用户查询提供针对特定病害的建议。为提高预测透明度，使用Grad-CAM突出显示影响模型输出的图像区域。深度学习、可解释人工智能和生成式人工智能的集成提供了一种可扩展的方法，用于检测和管理指黍病害。所提出的工作通过提供基于人工智能和可解释人工智能支持的决策，促进了精准农业，同时推动了可持续农业实践。","Discover Artificial Intelligence","2026-09-22T00:00:00Z",{"impact":17,"substance":101,"depth":403,"authority":48,"freshness":12,"relevant":19,"comment":404},18,"将深度学习、可解释AI与生成式AI结合用于指状粟病害识别，方法新颖、数据充分，对智慧农业有参考价值。",[406],{"name":400,"url":397},[24,25,408,355,409],"可解释AI","小米作物",{"code":29,"slug":30,"name":31,"path":32,"parent_code":33},[412],{"code":36,"slug":37,"name":38,"path":39,"parent_code":40},{"status":42,"locked":13,"version":43,"method":44,"evidence":414},{"candidates":415,"agricultural_evidence":49},[416,420,424,430,437],{"code":36,"score":276,"supported":49,"title":417,"abstract":418,"tags":419},[62],[300,62],[25,408,355],{"code":55,"score":17,"supported":49,"title":421,"abstract":422,"tags":423},[57],[109,129],[24],{"code":29,"score":425,"supported":49,"title":426,"abstract":428,"tags":429},11,[427],"millet",[427],[],{"code":431,"score":89,"supported":49,"title":432,"abstract":433,"tags":436},"H7",[],[434,435],"气候适应","干旱",[],{"code":438,"score":227,"supported":49,"title":439,"abstract":440,"tags":442},"H8",[],[441],"可持续农业",[],[444,445],"ResNet50 GWO 病害识别","农业人工智能 小米作物 智慧农业 病害识别","ResNet50GWO病害识别-3153","10.1007\u002Fs44163-026-02222-y",{"doi":447,"openalex_id":449,"authors":450,"venue":400,"cited_by_count":83,"oa_url":397,"card":469,"direction":85,"ingested_from":86},"W7213978567",[451,454,456,458,460,463,466],{"name":452,"orcid":453},"Sunil Kumar Mohapatra","https:\u002F\u002Forcid.org\u002F0000-0002-5865-095X",{"name":455,"orcid":8},"A. Sanjib Kumar Patro",{"name":457,"orcid":8},"Lulen Kumar Sahu",{"name":459,"orcid":8},"Chinmaye Dora",{"name":461,"orcid":462},"Sujata Chakravarty","https:\u002F\u002Forcid.org\u002F0000-0002-1293-5378",{"name":464,"orcid":465},"Kshira Sagar Sahoo","https:\u002F\u002Forcid.org\u002F0000-0002-6435-5738",{"name":467,"orcid":468},"Byomakesh Mahapatra","https:\u002F\u002Forcid.org\u002F0000-0002-9126-1729",{"tldr":470,"method":471,"finding":472,"direction":158,"opportunity":473},"提出融合深度学习、可解释AI与生成式AI的指状粟病害检测框架，实现高精度识别与决策支持。","用Kaggle指状粟图像训练CNN、VGG16、ResNet50，并用灰狼优化做","GWO特征选择显著提升精度，ResNet50达98.36%，结合Grad-CAM与Gemini AP","可探索轻量化模型与多作物泛化，并将XAI与生成式建议在田间移动端实时验证。","2026-09-22T23:30:10.255790Z",{"id":476,"title":477,"url":478,"summary":479,"summary_zh":480,"content":8,"source_name":272,"source_url":478,"published_at":481,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":274,"score_detail":482,"sources":485,"tags":487,"primary_topic":489,"secondary_topics":490,"classification":491,"search_phrases":507,"slug":510,"view_count":83,"doi":511,"paper":512,"created_at":526},5584,"Distributionally-Robust Evaluation of Computer Vision Agriculture Disease Classifiers Under Climate-Driven Pathogen Range Shifts: An Artificial Intelligence Worst-Case Generalization Bound","https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23544150","Computer vision classifiers for crop disease are evaluated almost exclusively on held-out samples drawn from the same distribution as their training data, yet climate-driven pathogen range shifts mean that deployment environments are systematically out of distribution. The framework treats the deployment distribution as an adversarial perturbation of the training distribution within a Wasserstein ball whose radius is calibrated to observed climate displacement. Using NASA POWER monthly reanalysis for Harare, Zimbabwe (2000-2023, 312 monthly observations), the analysis estimates a mean air temperature of 18.914 C, a standard deviation of 3.107 C, and a warming trend of 0.2001 C per decade. These estimates anchor the radius of the ambiguity set and the resulting bound. The bound decomposes worst-case risk into an empirical risk term, a Wasserstein radius term scaled by the Lipschitz constant of the classifier, and a concentration term. Under plausible Lipschitz constants for convolutional backbones, the radius term dominates the empirical risk term once the climate displacement exceeds a threshold that the Harare record already approaches. The contribution is a formal evaluation criterion that replaces the i. i. d. assumption with a climate-calibrated ambiguity set, together with the analytical machinery needed to compute it.","作物病害的计算机视觉分类器几乎完全在与训练数据同分布的留出样本上进行评估，然而气候驱动的病原体分布范围变化意味着部署环境系统性地处于分布之外。该框架将部署分布视为训练分布在Wasserstein球内的对抗性扰动，球的半径根据观测到的气候位移进行校准。利用NASA POWER对津巴布韦哈拉雷的月度再分析数据（2000—2023年，312个月度观测值），分析估计平均气温为18.914°C，标准差为3.107°C，每十年升温趋势为0.2001°C。这些估计值锚定了模糊集的半径及由此得到的界。该界将最坏情况风险分解为经验风险项、由分类器Lipschitz常数缩放后的Wasserstein半径项以及集中项。在卷积骨干网络合理的Lipschitz常数下，一旦气候位移超过哈拉雷记录已接近的阈值，半径项便主导经验风险项。本文的贡献在于提出了一种形式化评估准则，以气候校准的模糊集替代独立同分布（i.i.d.）假设，并提供了计算该准则所需的分析工具。","2026-10-21T00:00:00Z",{"impact":48,"substance":483,"depth":403,"authority":103,"freshness":12,"relevant":19,"comment":484},22,"方法新颖、数据具体，但属方法论预印论文，产业影响有限，适合作为专业精选而非头条。",[486],{"name":272,"url":478},[24,25,301,27,488],"气候变化",{"code":36,"slug":37,"name":38,"path":39,"parent_code":40},[],{"status":42,"locked":13,"version":43,"method":44,"evidence":492},{"candidates":493,"agricultural_evidence":49},[494,499,503],{"code":36,"score":495,"supported":49,"title":496,"abstract":497,"tags":498},24,[302,303],[301,303],[25,301],{"code":431,"score":19,"supported":13,"title":500,"abstract":501,"tags":502},[],[],[488],{"code":55,"score":19,"supported":13,"title":504,"abstract":505,"tags":506},[],[],[24],[508,509],"NASA POWER 哈拉雷 气温","Wasserstein 分布鲁棒 病害分类","NASAPOWER哈拉雷气温-5584","10.5281\u002Fzenodo.23544150",{"doi":511,"openalex_id":513,"authors":514,"venue":272,"cited_by_count":83,"oa_url":478,"card":521,"direction":158,"ingested_from":86},"W7223087892",[515,517,519],{"name":516,"orcid":8},"Kossi Afeli",{"name":518,"orcid":8},"Akouvi Amegashie",{"name":520,"orcid":8},"Komlan Sélom Dossou",{"tldr":522,"method":523,"finding":524,"direction":158,"opportunity":525},"提出气候校准的分布鲁棒评估准则，替代i.i.d.假设评价作物病害视觉分类器。","用Wasserstein球建模部署分布偏移，以NASA POWER哈拉雷气温数据","气候位移超阈值后，半径项主导经验风险，现有评估会高估模型泛化能力。","可将气候校准的鲁棒评估扩展到多作物多区域，并研究适应分布偏移的鲁棒训练方法。","2026-10-22T11:21:00.036937Z"]