[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-2502":3,"related-2502":85},{"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":63,"slug":66,"view_count":21,"doi":67,"paper":68,"created_at":84},2502,"CBAM-LeafGAN: A selective attention-guided StyleGAN framework for mango leaf disease image synthesis and explainable recognition","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.compag.2026.112409","The development of robust deep learning systems for plant disease diagnosis is constrained by limited expert-annotated datasets, class imbalance, and insufficient visual diversity, particularly in real-world field mango leaf disease classification. To address these challenges, this paper proposes CBAM-LeafGAN, a Selective Residual-Gated CBAM-enhanced StyleGAN3-t framework for disease-specific image synthesis and augmentation. Lightweight Residual-Gated Convolutional Block Attention Modules are selectively integrated into the intermediate synthesis layers to enhance disease-relevant feature generation while preserving alias-free synthesis. Evaluated on the MangoLeafDS2025 dataset, CBAM-LeafGAN was compared with StyleGAN3-t, StyleGAN3-r, Vanilla Latent Diffusion Model, DiffusionPix2Pix, Variational Autoencoder, and LeafGAN. It achieved the lowest Fréchet Inception Distance (FID) of 12.44 among the evaluated models. Ablation, t-SNE, nearest-neighbour diversity, Turing tests, and statistical analyses further validated the realism and diversity of the images. Synthetic augmentation improved disease classification across 23 deep learning architectures and mitigated class imbalance. Evaluation of the proposed approach on the comparable dataset MangoLeafBD further demonstrated improved performance over state-of-the-art methods. Grad-CAM visualisation and quantitative explainability analysis confirmed the consistent localisation of disease-relevant regions, supporting the reliability and interpretability of the proposed framework for mango leaf disease diagnosis under data-constrained agricultural conditions and indicating its potential for practical field deployment.","为植物病害诊断构建稳健的深度学习系统受到专家标注数据集有限、类别不平衡以及视觉多样性不足的制约，尤其是在真实田间芒果叶片病害分类场景中。为应对这些挑战，本文提出CBAM-LeafGAN，一种用于病害特异性图像合成与增强的选择性残差门控CBAM增强StyleGAN3-t框架。轻量级残差门控卷积块注意力模块（Residual-Gated Convolutional Block Attention Modules）被选择性地集成到中间合成层中，以增强病害相关特征的生成，同时保持无混叠合成。在MangoLeafDS2025数据集上进行评估时，CBAM-LeafGAN与StyleGAN3-t、StyleGAN3-r、原始潜在扩散模型（Vanilla Latent Diffusion Model）、DiffusionPix2Pix、变分自编码器（Variational Autoencoder）和LeafGAN进行了比较。在受评估模型中，它取得了最低的Fréchet Inception距离（FID），为12.44。消融实验、t-SNE、最近邻多样性、图灵测试和统计分析进一步验证了图像的真实性和多样性。合成增强在23种深度学习架构上改善了病害分类，并缓解了类别不平衡。在可比数据集MangoLeafBD上对所提方法进行评估，进一步证明其性能优于当前最先进方法。Grad-CAM可视化和定量可解释性分析证实了病害相关区域定位的一致性，支持了所提框架在数据受限农业条件下用于芒果叶片病害诊断的可靠性和可解释性，并表明其在实际田间部署中的潜力。",null,"Computers and Electronics in Agriculture","2026-09-14T00:00:00Z","论文",10,false,81,{"impact":17,"substance":18,"depth":17,"authority":19,"freshness":20,"relevant":21,"comment":22},18,22,14,9,1,"提出CBAM-LeafGAN生成式数据增强框架，在芒果叶病害识别上取得FID 12.44并提升23种模型分类性能，方法新颖、验证充分，对数据受限场景下的作物病害智能诊断有参考价值。",[24],{"name":10,"url":6},[26,27,28,29,30],"智慧农业","农业人工智能","芒果","病害识别","图像合成",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},"I2","agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai","I",[],{"status":39,"locked":14,"version":40,"method":41,"evidence":42},"classified","2026-10-22.2","rules",{"candidates":43,"agricultural_evidence":46},[44,52,58],{"code":32,"score":45,"supported":46,"title":47,"abstract":48,"tags":51},8,true,[],[49,50],"深度学习","deep learning",[27,29],{"code":53,"score":54,"supported":46,"title":55,"abstract":56,"tags":57},"B2",4,[],[28],[28],{"code":59,"score":21,"supported":14,"title":60,"abstract":61,"tags":62},"I1",[],[],[26],[64,65],"农业人工智能 图像合成 智慧农业 病害识别","农业人工智能 图像合成","农业人工智能图像合成智慧农业病害识别-2502","10.1016\u002Fj.compag.2026.112409",{"doi":67,"openalex_id":69,"authors":70,"venue":10,"cited_by_count":76,"oa_url":6,"card":77,"direction":81,"ingested_from":83},"W7212973113",[71,73],{"name":72,"orcid":9},"Pankajkumar Thakre",{"name":74,"orcid":75},"Jagdish Chakole","https:\u002F\u002Forcid.org\u002F0000-0003-0242-7297",0,{"tldr":78,"method":79,"finding":80,"direction":81,"opportunity":82},"提出CBAM-LeafGAN，用选择性注意力增强StyleGAN3合成芒果叶病图像并提升识别可解释性","在StyleGAN3-t中间层选择性集成残差门控CBAM，用MangoLeafD","FID低至12.44，合成增强提升23种分类模型性能并缓解类别不平衡，Grad-CAM定位可靠。","农业人工智能与决策模型","可探索注意力引导生成模型在更多作物病害及跨域田间场景下的泛化与轻量化部署。","openalex","2026-09-15T23:30:01.657805Z",{"total":86,"page":21,"page_size":86,"items":87},6,[88,157,222,293,353,417],{"id":89,"title":90,"url":91,"summary":92,"summary_zh":93,"content":9,"source_name":94,"source_url":91,"published_at":95,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":96,"score_detail":97,"sources":100,"tags":102,"primary_topic":105,"secondary_topics":106,"classification":107,"search_phrases":142,"slug":145,"view_count":76,"doi":146,"paper":147,"created_at":156},5311,"AI SMART AGRO SYSTEM: A WEB-BASED AI-ASSISTED DECISION SUPPORT PLATFORM FOR FARMERS","https:\u002F\u002Fdoi.org\u002F10.56726\u002Firjmets103514","Farmers often have to decide on disease control, crop selection, soil management and irrigation without timely expert advice, which leads to yield loss and wastage of water and fertiliser.This paper presents AI Smart Agro System, a web-based decision-support platform that brings five services into one dashboard: leaf disease detection from an uploaded photograph, crop recommendation, soil health analysis, irrigation advisory and weather-based farming advice.The system is built with PHP, MySQL and JavaScript, while disease detection and crop recommendation are handled by an external Python Flask AI service that is called through a secured REST API.Soil, irrigation and weather advice are generated by transparent rule-based logic using soil pH, N-P-K values, soil moisture, rainfall, temperature and humidity.Every result is stored in the user's history, and notifications and an administrator panel support monitoring.The prototype is designed to return the disease name, confidence, treatment and prevention tips for an uploaded leaf image, to recommend the three most suitable crops, and to give clear advice for soil and water management.Quantitative evaluation of the AI models is left for future work.","农民常常需要在没有及时专家建议的情况下决定病害防治、作物选择、土壤管理和灌溉，这会导致产量损失以及水和肥料的浪费。本文提出AI Smart Agro System，这是一个基于Web的决策支持平台，将五项服务整合到一个仪表板中：根据上传照片进行叶片病害检测、作物推荐、土壤健康分析、灌溉建议以及基于天气的农事建议。该系统使用PHP、MySQL和JavaScript构建，而病害检测和作物推荐由外部Python Flask AI服务处理，并通过安全的REST API调用。土壤、灌溉和天气建议由透明的基于规则的逻辑生成，使用土壤pH、N-P-K值、土壤湿度、降雨量、温度和湿度。每个结果都存储在用户历史记录中，通知和管理员面板支持监控。该原型旨在为上传的叶片图像返回病害名称、置信度、治疗和预防提示，推荐三种最合适的作物，并为土壤和水管理提供明确建议。AI模型的定量评估留待未来工作。","International Research Journal of Modernization in Engineering Technology and Science","2026-10-17T00:00:00Z",49,{"impact":45,"substance":19,"depth":98,"authority":86,"freshness":20,"relevant":21,"comment":99},12,"原型系统整合病害识别、作物推荐与灌溉建议，思路清晰但缺乏定量评估，属细分技术进展，公共影响有限。",[101],{"name":94,"url":91},[26,27,103,29,104],"决策支持系统","灌溉管理",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[],{"status":39,"locked":14,"version":40,"method":41,"evidence":108},{"candidates":109,"agricultural_evidence":46},[110,116,124,131,138],{"code":32,"score":111,"supported":46,"title":112,"abstract":114,"tags":115},11,[113],"decision support",[],[27,103,29],{"code":117,"score":118,"supported":46,"title":119,"abstract":120,"tags":123},"H3",7,[],[121,122],"灌溉","irrigation",[104],{"code":125,"score":86,"supported":46,"title":126,"abstract":127,"tags":130},"H2",[],[128,129],"土壤","soil",[],{"code":132,"score":133,"supported":46,"title":134,"abstract":135,"tags":137},"F3",3,[],[136],"肥料",[],{"code":59,"score":21,"supported":14,"title":139,"abstract":140,"tags":141},[],[],[26],[143,144],"Web 农业决策支持平台","农业人工智能 决策支持系统 智慧农业 灌溉管理","Web农业决策支持平台-5311","10.56726\u002Firjmets103514",{"doi":146,"openalex_id":148,"authors":149,"venue":94,"cited_by_count":76,"oa_url":91,"card":150,"direction":155,"ingested_from":83},"W7222785888",[],{"tldr":151,"method":152,"finding":153,"direction":81,"opportunity":154},"构建一个面向农民的网页版AI决策支持平台，集成病害识别、选种、土壤、灌溉与天气建议。","PHP\u002FMySQL\u002FJS前端，Python Flask提供病害与选种AI服务，规","原型可输出病害名称、置信度与防治建议，推荐三种适宜作物并给出水土管理建议。","AI模型尚未定量评估，可补充精度验证与田间实测，并扩展多语言、离线与本地化作物模型。","智慧农业 \u002F 农业物联网","2026-10-19T23:30:08.043132Z",{"id":158,"title":159,"url":160,"summary":161,"summary_zh":162,"content":9,"source_name":163,"source_url":160,"published_at":164,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":165,"score_detail":166,"sources":169,"tags":171,"primary_topic":174,"secondary_topics":175,"classification":176,"search_phrases":200,"slug":203,"view_count":76,"doi":204,"paper":205,"created_at":221},4994,"Deep learning-based estimation of Kent mango yield by commercial export category under real field conditions in northern Peru","https:\u002F\u002Fdoi.org\u002F10.3389\u002Ffpls.2026.1943711","Manual visual counting remains the standard procedure for estimating Kent mango ( Mangifera indica L.) yield in northern Peru, an inconsistent practice that compromises harvest planning, transport dimensioning and commercial negotiation. Existing computer vision approaches to mango detection resolve fruit-versus-background discrimination but rarely classify fruit into the commercial categories that determine its market value. This study developed and validated a deep learning method that simultaneously detects and classifies Kent mango fruit into three export categories (air freight, sea freight and reject) from RGB images acquired under real field conditions in the Motupe district, Lambayeque. A purpose-built dataset of 400 photographs was compiled from 200 trees across ten commercial farms using two opposite views per tree; 6,544 patches of 1,280 × 1,280 pixels with 20% overlap were generated, containing 66,465 manually annotated instances. Five architectures were trained under identical experimental conditions: YOLOv8m, YOLO11m, YOLO26m, RT-DETR-l and Mango-YOLO, the latter a proposed architecture built on YOLO26m through the incorporation of a CBAM attention module, a BiFPN neck and the MPDIoU regression loss. Mango-YOLO achieved the best performance across all five metrics, with a precision of 0.930, a recall of 0.940, an F1-score of 0.930, a mAP@0.5 of 0.940 and a mAP@0.5:0.95 of 0.730, at an inference time of 114.94 ms per image. Per-category analysis revealed heterogeneous behaviour, with F1-scores of 0.960, 0.845 and 0.792 for sea freight, air freight and reject respectively. The selected model was integrated into a three-tier web application that scales fruit counts from sampled trees to the whole farm and converts them into kilograms through Monte Carlo simulation with category mean weights, returning a 90% interval. Validation against harvested production recorded on the processing-plant scale across the ten farms yielded a mean absolute percentage error of 2.63% and a coefficient of determination of 0.9993, outperforming the field supervisor’s manual estimate by a factor of 3.3.","在秘鲁北部，人工目视计数仍是估算肯特芒果（Mangifera indica L.）产量的标准方法，但这种做法缺乏一致性，影响了采收计划、运输规模测算和商业谈判。现有的芒果检测计算机视觉方法解决了果实与背景的区分问题，但很少将果实归入决定其市场价值的商业类别。本研究开发并验证了一种深度学习方法，能够从兰巴耶克省莫图佩地区真实田间条件下采集的RGB图像中，同时检测肯特芒果并将其分为三个出口类别（空运、海运和拒收）。研究构建了一个专用数据集，来自十个商业果园的200棵果树，每棵树采集两个相对视角，共400张照片；生成了6,544个1,280 × 1,280像素、重叠率20%的图像块，包含66,465个人工标注实例。在相同实验条件下训练了五种架构：YOLOv8m、YOLO11m、YOLO26m、RT-DETR-l和Mango-YOLO，后者是在YOLO26m基础上引入CBAM注意力模块、BiFPN颈部和MPDIoU回归损失而提出的架构。Mango-YOLO在全部五项指标上表现最佳，精确率为0.930，召回率为0.940，F1分数为0.930，mAP@0.5为0.940，mAP@0.5:0.95为0.730，每张图像推理时间为114.94 ms。分类别分析显示出异质性表现，海运、空运和拒收类别的F1分数分别为0.960、0.845和0.792。所选模型被集成到一个三层Web应用程序中，可将采样果树的果实计数扩展至整个果园，并通过使用类别平均权重的蒙特卡洛模拟将其转换为千克，返回90%区间。在十个果园中，以加工厂地磅记录的采收产量进行验证，平均绝对百分比误差为2.63%，决定系数为0.9993，比田间主管的人工估算优3.3倍。","Frontiers in Plant Science","2026-10-13T00:00:00Z",82,{"impact":17,"substance":167,"depth":17,"authority":19,"freshness":20,"relevant":21,"comment":168},23,"深度学习实现芒果出口分级与产量估算，MAPE仅2.63%，方法新颖、数据扎实，对果园智能测产有参考价值。",[170],{"name":163,"url":160},[26,27,172,173,28],"计算机视觉","产量预测",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[],{"status":39,"locked":14,"version":40,"method":41,"evidence":177},{"candidates":178,"agricultural_evidence":46},[179,186,191,195],{"code":32,"score":180,"supported":46,"title":181,"abstract":182,"tags":185},19,[50],[49,172,50,183,184],"computer vision","YOLO",[27,172],{"code":53,"score":118,"supported":46,"title":187,"abstract":188,"tags":190},[],[189,28],"果树",[28],{"code":59,"score":21,"supported":14,"title":192,"abstract":193,"tags":194},[],[],[26],{"code":196,"score":21,"supported":14,"title":197,"abstract":198,"tags":199},"I3",[],[],[173],[201,202],"秘鲁 Kent 芒果 产量预测","Mango-YOLO 芒果 分级","秘鲁Kent芒果产量预测-4994","10.3389\u002Ffpls.2026.1943711",{"doi":204,"openalex_id":206,"authors":207,"venue":163,"cited_by_count":76,"oa_url":215,"card":216,"direction":155,"ingested_from":83},"W7221685456",[208,210,213],{"name":209,"orcid":9},"Víctor Yoel Huamán Asencio",{"name":211,"orcid":212},"Heber I. Mejía-Cabrera","https:\u002F\u002Forcid.org\u002F0000-0002-0007-0928",{"name":214,"orcid":9},"Juan Carlos Arcila Diaz","https:\u002F\u002Fwww.frontiersin.org\u002Fjournals\u002Fplant-science\u002Farticles\u002F10.3389\u002Ffpls.2026.1943711\u002Fpdf",{"tldr":217,"method":218,"finding":219,"direction":81,"opportunity":220},"开发深度学习模型在真实田间条件下检测并分类肯特芒果为三个出口等级，实现产量估计。","构建400张田间RGB图像数据集，提出Mango-YOLO（YOLO26m+CB","Mango-YOLO最优，mAP@0.5达0.940，产量估计MAPE为2.63%，比人工估计准确3","可探索多品种、多生长阶段及不同光照条件下的泛化能力，并融合时序遥感数据提升区域产量预测精度。","2026-10-15T23:30:11.559035Z",{"id":223,"title":224,"url":225,"summary":226,"summary_zh":227,"content":9,"source_name":163,"source_url":225,"published_at":228,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":15,"score_detail":229,"sources":231,"tags":233,"primary_topic":236,"secondary_topics":242,"classification":244,"search_phrases":260,"slug":263,"view_count":21,"doi":264,"paper":265,"created_at":292},4806,"A study on the identification of wheat powdery mildew using chlorophyll fluorescence imaging","https:\u002F\u002Fdoi.org\u002F10.3389\u002Ffpls.2026.1837843","Introduction Wheat powdery mildew poses a serious threat to global wheat production, and its accurate identification is key to smart agriculture. Methods This study combines chlorophyll fluorescence imaging with deep learning algorithms to investigate the temporal dynamics of key fluorescence parameters (such as maximum photochemical efficiency, QYmax) during the progression of powderymildew. A QYmax-based chlorophyll fluorescence image dataset for wheat powdery mildew was constructed. Building on this, the LCRNet model was proposed, which effectively processes fluorescence image data through a collaborative mechanism integrating the Large Selective Adaptive Fusion (LSAF) and Coordinated Attention (CA) modules. Results On an independent test set, the model achieved an accuracy of 94% and an F1 score of 94.1%, outperforming comparison models. Discussion This study proposes a method for classifying the severity of powdery mildew in wheat that reduces the computational cost of fluorescence data processing, providing important methodological support and technical insights for advancing the intelligent, precise, and green transformation of crop disease prevention and control systems.","引言 小麦白粉病对全球小麦生产构成严重威胁，其准确识别是实现智慧农业的关键。方法 本研究将叶绿素荧光成像与深度学习算法相结合，探究白粉病发展过程中关键荧光参数（如最大光化学效率QYmax）的时间动态变化，构建了基于QYmax的小麦白粉病叶绿素荧光图像数据集。在此基础上，提出了LCRNet模型，该模型通过融合大选择性自适应融合（LSAF）模块与协调注意力（CA）模块的协同机制，有效处理荧光图像数据。结果 在独立测试集上，该模型达到了94%的准确率和94.1%的F1分数，优于对比模型。讨论 本研究提出了一种小麦白粉病严重程度分类方法，降低了荧光数据处理的计算成本，为推动作物病害防控体系向智能化、精准化和绿色化转型提供了重要的方法学支撑和技术启示。","2026-10-12T00:00:00Z",{"impact":17,"substance":18,"depth":17,"authority":19,"freshness":20,"relevant":21,"comment":230},"将叶绿素荧光成像与深度学习结合用于小麦白粉病严重度识别，方法新颖、数据可靠，对智慧植保有实质参考价值。",[232],{"name":163,"url":225},[26,27,29,234,235],"叶绿素荧光","小麦白粉病",{"code":237,"slug":238,"name":239,"path":240,"parent_code":241},"A2","wheat","小麦","\u002Ftopic\u002Ffield-crops\u002Fwheat","A",[243],{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},{"status":39,"locked":14,"version":40,"method":41,"evidence":245},{"candidates":246,"agricultural_evidence":46},[247,251,255],{"code":237,"score":19,"supported":46,"title":248,"abstract":249,"tags":250},[238],[239,238],[],{"code":32,"score":45,"supported":46,"title":252,"abstract":253,"tags":254},[],[49,50],[27,29],{"code":59,"score":118,"supported":46,"title":256,"abstract":257,"tags":259},[],[26,258],"smart agriculture",[26],[261,262],"小麦白粉病 叶绿素荧光 深度学习","LCRNet 小麦病害识别","小麦白粉病叶绿素荧光深度学习-4806","10.3389\u002Ffpls.2026.1837843",{"doi":264,"openalex_id":266,"authors":267,"venue":163,"cited_by_count":76,"oa_url":286,"card":287,"direction":155,"ingested_from":83},"W7221549180",[268,270,273,276,278,280,282,284],{"name":269,"orcid":9},"Yijia Chen",{"name":271,"orcid":272},"Chengyuan Hao","https:\u002F\u002Forcid.org\u002F0000-0002-6002-3963",{"name":274,"orcid":275},"Zhiyu Li","https:\u002F\u002Forcid.org\u002F0000-0003-3771-0199",{"name":277,"orcid":9},"Xupeng An",{"name":279,"orcid":9},"Jingxiang Huang",{"name":281,"orcid":9},"Xinchang Chen",{"name":283,"orcid":9},"Shidong Wang",{"name":285,"orcid":9},"ZhiChao Chen","https:\u002F\u002Fwww.frontiersin.org\u002Fjournals\u002Fplant-science\u002Farticles\u002F10.3389\u002Ffpls.2026.1837843\u002Fpdf",{"tldr":288,"method":289,"finding":290,"direction":81,"opportunity":291},"该研究利用叶绿素荧光成像与深度学习识别小麦白粉病并分级。","构建QYmax荧光图像数据集，提出融合LSAF与CA模块的LCRNet模型。","模型在独立测试集上准确率达94%、F1为94.1%，优于对比模型。","可探索多病害、多作物荧光图像通用模型，并融合田间时序数据实现早期预警。","2026-10-13T23:30:09.060156Z",{"id":294,"title":295,"url":296,"summary":297,"summary_zh":298,"content":9,"source_name":299,"source_url":296,"published_at":228,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":300,"score_detail":301,"sources":305,"tags":307,"primary_topic":310,"secondary_topics":311,"classification":312,"search_phrases":329,"slug":332,"view_count":21,"doi":333,"paper":334,"created_at":352},4804,"Comparative Deep Learning and RISE-Based Explainable AI Framework for Orange Fruit Disease Detection","https:\u002F\u002Fdoi.org\u002F10.15832\u002Fankutbd.1850038","Orange diseases significantly threaten fruit quality and agricultural productivity, resulting in substantial economic losses. Traditional diagnostic methods are limited by their labor-intensive nature, high costs, and reliance on subjective expert evaluation, increasing the demand for automated solutions. This study systematically compared three state-of the-art deep learning architectures-EfficientNet-B0, ConvNeXt-Tiny, and Vision Transformer (ViT-B\u002F16)-on the Orange Fruit dataset, which comprises healthy, citrus canker, and melanose samples. To enhance generalization and mitigate overfitting, transfer learning with ImageNet pretraining was combined with advanced data augmentation strategies like Mixup and Cutout. Experimental results demonstrated that the ConvNeXt-Tiny architecture, supported by these techniques, achieved superior performance, reaching 98.76% Accuracy, 99% F1-Score, and 99.95% ROC-AUC. Beyond numerical metrics, model transparency was investigated using Explainable Artificial Intelligence (XAI). The study evaluated the widely used Gradient-weighted Class Activation Mapping (Grad-CAM) method alongside Randomized Input Sampling for Explanation (RISE), which, to the best of our knowledge, represents one of the first applications of RISE in plant disease image analysis. RISE provided more comprehensive visualizations of the fruit surface regions influencing decisions, thereby strengthening interpretability. These findings confirm the high potential of modern deep learning for early disease diagnosis and highlight that XAI integration significantly enhances the reliability of agricultural decision support systems. Five fold cross-validation experiments and paired t-test analysis further confirmed the robustness and statistical significance of these findings.","橙病严重威胁果实品质和农业生产力，造成重大经济损失。传统诊断方法受限于劳动强度大、成本高以及依赖专家主观评估，增加了对自动化解决方案的需求。本研究在包含健康、柑橘溃疡病和黑斑病样本的橙果数据集上，系统比较了三种先进的深度学习架构——EfficientNet-B0、ConvNeXt-Tiny和Vision Transformer（ViT-B\u002F16）。为增强泛化能力并缓解过拟合，将基于ImageNet预训练的迁移学习与Mixup和Cutout等先进数据增强策略相结合。实验结果表明，在这些技术的支持下，ConvNeXt-Tiny架构取得了最优性能，达到98.76%的准确率、99%的F1分数和99.95%的ROC-AUC。除数值指标外，本研究还利用可解释人工智能（XAI）探讨了模型透明度。研究评估了广泛使用的梯度加权类激活映射（Grad-CAM）方法以及随机输入采样解释方法（RISE），据我们所知，这是RISE在植物病害图像分析中的首批应用之一。RISE提供了更全面的果实表面区域可视化，展示了影响决策的区域，从而增强了可解释性。这些发现证实了现代深度学习在早期病害诊断中的巨大潜力，并表明XAI的整合显著提升了农业决策支持系统的可靠性。五折交叉验证实验和配对t检验分析进一步证实了这些发现的稳健性和统计显著性。","Tarım Bilimleri Dergisi",72,{"impact":98,"substance":302,"depth":17,"authority":303,"freshness":20,"relevant":21,"comment":304},20,13,"系统比较三种深度学习架构并首次将RISE用于植物病害图像分析，方法新颖、数据可靠，对智慧农业病害诊断有参考价值。",[306],{"name":299,"url":296},[26,27,308,309,29],"可解释AI","柑橘",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[],{"status":39,"locked":14,"version":40,"method":41,"evidence":313},{"candidates":314,"agricultural_evidence":46},[315,320,325],{"code":32,"score":302,"supported":46,"title":316,"abstract":317,"tags":319},[50],[49,50,318,113],"artificial intelligence",[27,308,29],{"code":53,"score":118,"supported":46,"title":321,"abstract":322,"tags":324},[],[309,323],"citrus",[309],{"code":59,"score":21,"supported":14,"title":326,"abstract":327,"tags":328},[],[],[26],[330,331],"ConvNeXt 柑橘病害检测","RISE 可解释AI 植物病害","ConvNeXt柑橘病害检测-4804","10.15832\u002Fankutbd.1850038",{"doi":333,"openalex_id":335,"authors":336,"venue":299,"cited_by_count":76,"oa_url":346,"card":347,"direction":155,"ingested_from":83},"W7221532433",[337,340,343],{"name":338,"orcid":339},"Kürşad Buğrahan Yapar","https:\u002F\u002Forcid.org\u002F0009-0006-7666-6177",{"name":341,"orcid":342},"Nurettin Doğan","https:\u002F\u002Forcid.org\u002F0000-0002-8267-8469",{"name":344,"orcid":345},"Özgür Öksüz","https:\u002F\u002Forcid.org\u002F0000-0001-5568-6116","https:\u002F\u002Fdergipark.org.tr\u002Fen\u002Fdownload\u002Farticle-file\u002F5543457",{"tldr":348,"method":349,"finding":350,"direction":81,"opportunity":351},"比较三种深度学习模型检测橙子病害，并用RISE提升可解释性。","EfficientNet-B0、ConvNeXt-Tiny、ViT-B\u002F16，迁","ConvNeXt-Tiny达98.76%准确率，RISE比Grad-CAM提供更全面可解释可视化。","RISE在植物病害分析中应用尚少，可探索更多XAI方法提升农业决策系统可信度。","2026-10-13T23:30:08.943247Z",{"id":354,"title":355,"url":356,"summary":357,"summary_zh":358,"content":9,"source_name":359,"source_url":356,"published_at":360,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":361,"score_detail":362,"sources":364,"tags":366,"primary_topic":368,"secondary_topics":369,"classification":370,"search_phrases":389,"slug":392,"view_count":21,"doi":393,"paper":394,"created_at":416},4675,"Deep learning–based computer vision for rapid and accurate quantitative field assessment of powdery mildew severity: model comparison, validation, and agreement with human raters","https:\u002F\u002Fdoi.org\u002F10.1007\u002Fs40858-026-00845-1","Abstract Quantifying foliar disease severity accurately and reproducibly remains a fundamental challenge in plant pathology, as traditional visual assessment is inherently subjective and prone to rater variability. Although computer vision has enabled automated disease detection, validated end-to-end frameworks that integrate reference generation, model training, and field-deployable percent severity estimation are lacking. Here, we developed and validated a computer vision pipeline for quantifying powdery mildew ( Oidium spp.) severity in pea ( Pisum sativum ) using smartphone-acquired images under field conditions, with tomato ( Solanum lycopersicum ) powdery mildew serving as a cross-species validation system. A semi-automated multichannel segmentation approach generated continuous percent severity estimates as reproducible training references. Among evaluated deep learning architectures, a YOLO-based model was selected for its superior balance of predictive accuracy, computational efficiency, and deployment flexibility and was integrated into PowderyComputerVision, a purpose-built web application enabling real-time inference. Under field conditions, the automated system showed substantially higher agreement with reference severity values than trained human raters (concordance correlation coefficient (CCC), 0.952 vs. 0.513). A linear mixed-effects model revealed that human evaluators produced absolute estimation errors 280.96 times greater than the automated system ( p \u003C 0.0001; 95% CI 120.34–655.91). Performance remained robust across variable lighting conditions, imaging angles, and plant structures, supporting applicability beyond controlled settings. The proposed four-stage pipeline comprising image capture, host and disease classification via YOLOv11m-cls, leaf localization via YOLOv11n-seg, and lesion quantification via multichannel segmentation offers a scalable, objective, and reproducible framework for continuous field-based disease severity estimation.","摘要 准确且可重复地量化叶部病害严重度仍是植物病理学的一项基本挑战，因为传统的视觉评估本质上具有主观性，且容易受到评估者间变异的影响。尽管计算机视觉已使病害自动检测成为可能，但仍缺乏经过验证的端到端框架，将参考标准生成、模型训练和可田间部署的百分比严重度估算整合在一起。在此，我们开发并验证了一套计算机视觉流程，用于在田间条件下利用智能手机采集的图像量化豌豆(Pisum sativum)白粉病(Oidium spp.)严重度，并以番茄(Solanum lycopersicum)白粉病作为跨物种验证系统。一种半自动多通道分割方法生成了连续的百分比严重度估算值，作为可重复的训练参考标准。在评估的深度学习架构中，基于YOLO的模型因其在预测准确性、计算效率和部署灵活性之间的优越平衡而被选中，并被整合到PowderyComputerVision这一专门构建的Web应用程序中，以实现实时推理。在田间条件下，该自动化系统与参考严重度值的一致性显著高于受过训练的人类评估者(一致性相关系数(CCC)，0.952 vs. 0.513)。线性混合效应模型显示，人类评估者产生的绝对估算误差是自动化系统的280.96倍(p \u003C 0.0001；95% CI 120.34–655.91)。其性能在多变的光照条件、成像角度和植物结构下均保持稳健，支持其在受控环境之外的适用性。所提出的四阶段流程——包括图像采集、通过YOLOv11m-cls进行寄主与病害分类、通过YOLOv11n-seg进行叶片定位，以及通过多通道分割进行病斑量化——为基于田间的连续病害严重度估算提供了一个可扩展、客观且可重复的框架。","Tropical Plant Pathology","2026-10-09T00:00:00Z",80,{"impact":17,"substance":18,"depth":17,"authority":303,"freshness":20,"relevant":21,"comment":363},"该研究提出并验证了基于深度学习的白粉病严重度田间量化流水线，与人工评估对比数据扎实，对智慧植保具有较高参考价值。",[365],{"name":359,"url":356},[26,27,172,29,367],"豌豆白粉病",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[],{"status":39,"locked":14,"version":40,"method":41,"evidence":371},{"candidates":372,"agricultural_evidence":46},[373,378,385],{"code":32,"score":374,"supported":46,"title":375,"abstract":376,"tags":377},28,[50,183],[49,172,50,183,184],[27,29,172],{"code":379,"score":86,"supported":46,"title":380,"abstract":381,"tags":384},"B1",[],[382,383],"番茄","tomato",[],{"code":59,"score":21,"supported":14,"title":386,"abstract":387,"tags":388},[],[],[26],[390,391],"豌豆白粉病 计算机视觉","YOLO 病害严重度评估","豌豆白粉病计算机视觉-4675","10.1007\u002Fs40858-026-00845-1",{"doi":393,"openalex_id":395,"authors":396,"venue":359,"cited_by_count":76,"oa_url":409,"card":410,"direction":414,"ingested_from":83},"W7221108017",[397,399,402,404,406],{"name":398,"orcid":9},"Jesus Enrique Flores-Riera",{"name":400,"orcid":401},"William Fernando Cárdenas-Urrego","https:\u002F\u002Forcid.org\u002F0009-0006-4753-719X",{"name":403,"orcid":9},"Víctor Hugo Aguilar",{"name":405,"orcid":9},"Emerson Medeiros Del Ponte",{"name":407,"orcid":408},"Joaquín Guillermo Ramírez‐Gil","https:\u002F\u002Forcid.org\u002F0000-0002-0162-3598","https:\u002F\u002Flink.springer.com\u002Fcontent\u002Fpdf\u002F10.1007\u002Fs40858-026-00845-1.pdf",{"tldr":411,"method":412,"finding":413,"direction":414,"opportunity":415},"开发并验证基于深度学习的计算机视觉流程，用手机图像快速定量评估豌豆白粉病严重度。","YOLO模型与多通道分割，手机田间图像，豌豆和番茄白粉病数据。","自动系统与参考值一致性CCC达0.952，远高于人工评估的0.513，误差显著更低。","农业遥感与作物表型","可扩展至多作物多病害的田间连续严重度估计，并探索轻量化模型与实时决策集成。","2026-10-11T23:30:21.770403Z",{"id":418,"title":419,"url":420,"summary":421,"summary_zh":422,"content":9,"source_name":423,"source_url":420,"published_at":424,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":300,"score_detail":425,"sources":429,"tags":431,"primary_topic":434,"secondary_topics":435,"classification":436,"search_phrases":457,"slug":460,"view_count":21,"doi":461,"paper":462,"created_at":491},4338,"Bridging the performance–efficiency gap: a resource-efficient hybrid DL-ML framework for jackfruit disease detection","https:\u002F\u002Fdoi.org\u002F10.3389\u002Ffrai.2026.1884567","Complex environment variability and limited computational resources in agricultural settings create challenges in automated detection in tropical regions, where jackfruit diseases pose major threats to harvest productivity and food security. Conventional machine learning models often fail to generalize across diverse image conditions. For sustainable, field-deployable diagnostic systems, it is essential to close the performance–efficiency gap, as traditional deep learning methods require large amounts of training data and high-end hardware. We propose a hybrid deep learning machine learning (DL-ML) framework that has been tailored for accurate and resource-efficient jackfruit disease identification. This framework discusses deep visual representations that are extracted using a lightweight custom Convolutional Neural Network (CNN) combined with a previously trained EfficientNet-B0, capturing both localized and nested image features. Principal Component Analysis (PCA) is utilized for feature optimization to enhance discriminative power and eliminate redundancy, while SMOTE minimizes dataset imbalance by generating synthetic minority samples. These selected features are then classified by employing a variety of machine learning algorithms, including Random Forest (RF), Support Vector Machine (SVM), XGBoost, CatBoost, and k-Nearest Neighbors (KNN). A stacked ensemble model incorporates the strengths of these classifiers, enhancing generalization and robustness under diverse lighting and background conditions. Experiments across multiple dataset configurations show a noticeable improvement in accuracy from 76.80% on raw data to 99.39% on the PCA-augmented dataset, illustrating that this hybrid ensemble consistently outperforms all individual deep and machine learning baselines. Although the approach is currently tailored to jackfruit disease datasets, its applicability to other crops or stress conditions requires greater domain adaptation studies. However, the model's high accuracy highlights its scalability and utility in low-resource agricultural environments. Additionally, it establishes a cost-effective diagnostic paradigm where efficient ML classification and deep feature extraction collaborate, laying the foundation for next-generation smart farming and precision agriculture systems.","热带地区农业环境中复杂的环境多变性和有限的计算资源给自动化检测带来了挑战，其中菠萝蜜病害对收获产量和粮食安全构成重大威胁。传统机器学习模型往往难以在不同图像条件下泛化。对于可持续、可田间部署的诊断系统而言，弥合性能与效率之间的差距至关重要，因为传统深度学习方法需要大量训练数据和高端硬件。我们提出了一种混合深度学习-机器学习（DL-ML）框架，专为准确且资源高效的菠萝蜜病害识别而设计。该框架利用轻量级自定义卷积神经网络（CNN）结合预训练的EfficientNet-B0提取深度视觉表征，同时捕获局部和嵌套图像特征。采用主成分分析（PCA）进行特征优化，以增强判别能力并消除冗余，同时利用SMOTE通过生成合成少数类样本以缓解数据集不平衡问题。随后，采用多种机器学习算法对所选特征进行分类，包括随机森林（RF）、支持向量机（SVM）、XGBoost、CatBoost和k近邻（KNN）。堆叠集成模型融合了这些分类器的优势，增强了在不同光照和背景条件下的泛化能力和鲁棒性。在多种数据集配置上的实验表明，准确率从原始数据上的76.80%显著提升至PCA增强数据集上的99.39%，说明该混合集成模型始终优于所有单一深度学习和机器学习基线。尽管该方法目前针对菠萝蜜病害数据集进行了定制，但其对其他作物或胁迫条件的适用性仍需更多领域自适应研究。然而，该模型的高准确率凸显了其在低资源农业环境中的可扩展性和实用性。此外，它建立了一种经济高效的诊断范式，使高效的机器学习分类与深度特征提取协同工作，为下一代智慧农业和精准农业系统奠定了基础。","Frontiers in Artificial Intelligence","2026-10-06T00:00:00Z",{"impact":98,"substance":426,"depth":427,"authority":303,"freshness":20,"relevant":21,"comment":428},21,17,"提出轻量CNN与EfficientNet-B0混合特征加PCA与集成分类的菠萝蜜病害识别框架，准确率提升显著，对低资源农业场景有参考价值，但属单作物方法研究，影响面有限。",[430],{"name":423,"url":420},[26,27,432,29,433],"精准农业","热带作物",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[],{"status":39,"locked":14,"version":40,"method":41,"evidence":437},{"candidates":438,"agricultural_evidence":46},[439,444,450],{"code":59,"score":111,"supported":46,"title":440,"abstract":441,"tags":443},[],[26,432,442],"precision agriculture",[26,432],{"code":32,"score":111,"supported":46,"title":445,"abstract":446,"tags":449},[],[49,447,448,50],"机器学习","machine learning",[27,29],{"code":451,"score":86,"supported":46,"title":452,"abstract":453,"tags":456},"A8",[],[454,455],"粮食安全","food security",[],[458,459],"菠萝蜜 病害检测 深度学习","EfficientNet 病害识别","菠萝蜜病害检测深度学习-4338","10.3389\u002Ffrai.2026.1884567",{"doi":461,"openalex_id":463,"authors":464,"venue":423,"cited_by_count":76,"oa_url":420,"card":486,"direction":155,"ingested_from":83},"W7220649782",[465,468,471,474,477,480,483],{"name":466,"orcid":467},"Rittik Chandra Das Turjy","https:\u002F\u002Forcid.org\u002F0009-0004-3545-8457",{"name":469,"orcid":470},"Fahmid Al Farid","https:\u002F\u002Forcid.org\u002F0000-0003-2625-2348",{"name":472,"orcid":473},"Md. Hasan Imam Bijoy","https:\u002F\u002Forcid.org\u002F0000-0002-4756-3025",{"name":475,"orcid":476},"Sarbajit Paul Bappy","https:\u002F\u002Forcid.org\u002F0009-0006-7551-0461",{"name":478,"orcid":479},"Amir Sohel","https:\u002F\u002Forcid.org\u002F0009-0006-3493-0929",{"name":481,"orcid":482},"Jia Uddin","https:\u002F\u002Forcid.org\u002F0000-0002-3403-4095",{"name":484,"orcid":485},"Hezerul Bin Abdul Karim","https:\u002F\u002Forcid.org\u002F0000-0002-7613-4596",{"tldr":487,"method":488,"finding":489,"direction":81,"opportunity":490},"提出轻量CNN与EfficientNet-B0混合特征提取、PCA降维和堆叠集成分类的菠萝蜜病害检测","轻量CNN+EfficientNet-B0提取特征，PCA优化，SMOTE平衡，","PCA增强数据集上准确率从76.80%提升至99.39%，优于所有单一深度与机器学习基线。","可研究跨作物与胁迫条件的域适应迁移，并探索边缘设备上更低成本的实时部署方案。","2026-10-07T23:30:08.546665Z"]