[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-4375":3,"related-4375":104},{"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":84,"slug":87,"view_count":21,"doi":88,"paper":89,"created_at":103},4375,"Exploratory Explainable Machine Learning for Crop-Yield Interpretation: A Comparative Analysis of XGBoost, LightGBM, and CatBoost","https:\u002F\u002Fdoi.org\u002F10.35414\u002Fakufemubid.1993166","Interpretable crop-yield modelling is important for agricultural planning and transparent decision support; however, the predictive value and generalizability of machine-learning explanations depend strongly on evaluation protocol and dataset quality. This study evaluates an exploratory XAI workflow combining XGBoost, LightGBM, CatBoost, SHAP, engineered features, leakage-aware feature selection, repeated cross-validation, nested cross-validation, and a cross-dataset transfer experiment. Among the three models using the common full original-plus-engineered representation, CatBoost achieved the strongest numerical held-out performance, whereas XGBoost was retained as the TreeSHAP reference model. Nested model selection chose CatBoost in all five outer folds (mean RMSE 1.0060; mean R² 0.0846). Nested SHAP selection was slightly weaker (mean RMSE 1.0199; mean R² 0.0586), supporting compact interpretation rather than a universally superior subset. The cross-dataset experiment used 7,509 mapped Maize and Rice observations and only Crop, Temperature, and Rainfall. All external R² values were negative; because the internal yield unit was undocumented and the external target was reported in hg\u002Fha, these results indicate weak, non-unit-harmonized transfer rather than definitive external validation. Overall, the workflow is appropriate for transparent exploratory interpretation, not deployment-ready yield forecasting.","可解释的作物产量建模对农业规划和透明决策支持具有重要意义；然而，机器学习解释的预测价值和泛化能力在很大程度上取决于评估方案和数据集质量。本研究评估了一种探索性XAI工作流，该工作流结合了XGBoost、LightGBM、CatBoost、SHAP、工程化特征、泄漏感知特征选择、重复交叉验证、嵌套交叉验证以及跨数据集迁移实验。在使用常见的完整原始加工程化表示的三个模型中，CatBoost取得了最强的数值留出性能，而XGBoost被保留作为TreeSHAP参考模型。嵌套模型选择在全部五个外层折中均选择了CatBoost（平均RMSE为1.0060；平均R²为0.0846）。嵌套SHAP选择略弱（平均RMSE为1.0199；平均R²为0.0586），支持紧凑解释而非普遍更优的子集。跨数据集实验使用了7，509条映射后的玉米和水稻观测数据，仅包含作物、温度和降雨量。所有外部R²值均为负；由于内部产量单位未记录，而外部目标以hg\u002Fha报告，这些结果表明迁移较弱且未进行单位协调，而非确定性的外部验证。总体而言，该工作流适用于透明的探索性解释，尚不适合直接部署用于产量预测。",null,"Afyon Kocatepe University Journal of Sciences and Engineering","2026-10-06T00:00:00Z","论文",10,false,61,{"impact":17,"substance":18,"depth":19,"authority":13,"freshness":20,"relevant":21,"comment":22},8,18,16,9,1,"方法学扎实的可解释机器学习作物产量对比研究，但外部迁移R²为负、结论偏探索性，适合作为技术参考而非重大突破。",[24],{"name":10,"url":6},[26,27,28,29,30],"智慧农业","农业人工智能","产量预测","可解释AI","玉米水稻",{"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":45},[44,52,60,67,74,79],{"code":32,"score":19,"supported":45,"title":46,"abstract":48,"tags":51},true,[47],"machine learning",[49,50],"机器学习","decision support",[27,29],{"code":53,"score":54,"supported":45,"title":55,"abstract":56,"tags":59},"A1",6,[],[57,58],"水稻","rice",[],{"code":61,"score":54,"supported":45,"title":62,"abstract":63,"tags":66},"A3",[],[64,65],"玉米","maize",[],{"code":68,"score":69,"supported":45,"title":70,"abstract":71,"tags":73},"J1",3,[],[72],"农业规划",[],{"code":75,"score":21,"supported":14,"title":76,"abstract":77,"tags":78},"I1",[],[],[26],{"code":80,"score":21,"supported":14,"title":81,"abstract":82,"tags":83},"I3",[],[],[28],[85,86],"XGBoost LightGBM CatBoost 作物产量","SHAP 作物产量 可解释机器学习","XGBoostLightGBMCatBoost作物产量-4375","10.35414\u002Fakufemubid.1993166",{"doi":88,"openalex_id":90,"authors":91,"venue":10,"cited_by_count":95,"oa_url":6,"card":96,"direction":100,"ingested_from":102},"W7220716323",[92],{"name":93,"orcid":94},"Berker Baydan","https:\u002F\u002Forcid.org\u002F0000-0003-2806-368X",0,{"tldr":97,"method":98,"finding":99,"direction":100,"opportunity":101},"比较XGBoost、LightGBM和CatBoost在作物产量可解释建模中的表现，强调评估协议与数","XGBoost、LightGBM、CatBoost结合SHAP、特征工程、嵌套交","CatBoost在嵌套选择中表现最佳，但跨数据集迁移R²为负，模型仅适合探索性解释而非部署预测。","农业人工智能与决策模型","需建立单位统一、可迁移的作物产量数据集，并开发跨区域泛化的可解释模型评估协议。","openalex","2026-10-07T23:30:47.315999Z",{"total":54,"page":21,"page_size":54,"items":105},[106,234,288,339,415,483],{"id":107,"title":108,"url":109,"summary":110,"summary_zh":111,"content":9,"source_name":112,"source_url":109,"published_at":113,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":114,"score_detail":115,"sources":120,"tags":122,"primary_topic":124,"secondary_topics":125,"classification":137,"search_phrases":217,"slug":220,"view_count":213,"doi":221,"paper":222,"created_at":233},2518,"AI-Driven Transformation in Sustainable Agriculture: A Systematic Review","https:\u002F\u002Fdoi.org\u002F10.18805\u002Fijare.a-6627","The agriculture sector is facing unprecedented pressure from a rapidly growing global population, erratic weather conditions and declining water availability. Traditional farming methods are increasingly proving insufficient in meeting the rising food demand, necessitating a shift towards technology-driven solutions. This paper provides a critical and systematic review of the application of Artificial Intelligence (AI) in agriculture, analysing its impact on productivity and sustainability, while identifying potential barriers hindering adoption. Employing a Systematic Literature Review approach, this study synthesizes recent research organized around the core domains of agricultural production - crop, soil and nutrient, water and irrigation and crop-protection (pest, disease and weed) management, alongside yield forecasting - examining how data-driven tools such as machine learning, the Internet of Things, robotics and computer vision are applied within each and prioritizing studies that reported agronomically meaningful outcomes such as yield gains and resource-use efficiency. The strongest gains identified across the review fall within three agricultural areas - crop and soil monitoring, yield forecasting and robotic field operations such as targeted detection, weeding and harvesting. The technologies enabling them, chiefly machine-learning models such as convolutional neural networks combined with field and remote sensors, have delivered high in-field precision, with reported object-detection precision of up to 95.78% in field conditions. In practical terms, this translates into more efficient use of water, fertilizer and other inputs and into improved yield and crop quality. The review concludes that prioritizing Explainable AI and interoperable farm systems is central to building farmer trust and safeguarding food security under a changing climate.","农业部门正面临着来自全球人口快速增长、天气条件不稳定以及水资源日益减少的前所未有的压力。传统耕作方法在满足不断增长的粮食需求方面日益显得力不从心，因此有必要转向技术驱动的解决方案。本文对人工智能（Artificial Intelligence, AI）在农业中的应用进行了批判性和系统性的综述，分析了其对生产力和可持续性的影响，同时识别了阻碍其采用的潜在障碍。本研究采用系统性文献综述方法，围绕农业生产的核心领域——作物、土壤与养分、水资源与灌溉、作物保护（病虫害和杂草）管理以及产量预测——对近期研究进行了综合梳理，考察了机器学习、物联网、机器人技术和计算机视觉等数据驱动工具在各领域中的应用方式，并优先关注报告了具有农学意义成果（如产量提升和资源利用效率）的研究。综述中发现的最显著收益集中在三个农业领域——作物与土壤监测、产量预测以及机器人田间作业（如靶向检测、除草和收获）。实现这些收益的技术，主要是卷积神经网络等机器学习模型与田间及遥感传感器的结合，已在田间条件下实现了高精度，所报告的目标检测精度在田间条件下高达95.78%。在实际应用中，这意味着水、肥料及其他投入品的更高效利用，以及产量和作物品质的提升。综述得出结论：优先发展可解释人工智能（Explainable AI）和可互操作的农场系统，对于在气候变化背景下建立农民信任和保障粮食安全至关重要。","Indian Journal of Agricultural Research","2026-09-14T00:00:00Z",76,{"impact":18,"substance":116,"depth":117,"authority":118,"freshness":17,"relevant":21,"comment":119},20,17,13,"系统综述梳理AI在作物土壤监测、产量预测与机器人田间作业中的应用成效与推广障碍，结论扎实，对智慧农业方向有参考价值。",[121],{"name":112,"url":109},[26,27,28,29,123],"精准农业",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[126,132],{"code":127,"slug":128,"name":129,"path":130,"parent_code":131},"H3","irrigation","水资源与灌溉","\u002Ftopic\u002Fenvironment\u002Firrigation","H",{"code":133,"slug":134,"name":135,"path":136,"parent_code":131},"H8","sustainable-agriculture","可持续与保护性农业","\u002Ftopic\u002Fenvironment\u002Fsustainable-agriculture",{"status":39,"locked":14,"version":40,"method":41,"evidence":138},{"candidates":139,"agricultural_evidence":45},[140,149,155,160,167,174,181,188,194,200,206,212],{"code":32,"score":141,"supported":45,"title":142,"abstract":143,"tags":148},11,[],[144,49,145,47,146,147],"目标检测","计算机视觉","artificial intelligence","computer vision",[27,29],{"code":127,"score":20,"supported":45,"title":150,"abstract":151,"tags":154},[],[152,153,128],"灌溉","水资源",[],{"code":133,"score":17,"supported":45,"title":156,"abstract":158,"tags":159},[157],"sustainable agriculture",[],[],{"code":161,"score":54,"supported":45,"title":162,"abstract":163,"tags":166},"A8",[],[164,165],"粮食安全","food security",[],{"code":168,"score":54,"supported":45,"title":169,"abstract":170,"tags":173},"F3",[],[171,172],"肥料","fertilizer",[],{"code":175,"score":54,"supported":45,"title":176,"abstract":177,"tags":180},"H2",[],[178,179],"土壤","soil",[],{"code":182,"score":54,"supported":45,"title":183,"abstract":184,"tags":187},"I5",[],[185,186],"物联网","internet of things",[],{"code":80,"score":189,"supported":45,"title":190,"abstract":191,"tags":193},4,[],[192],"遥感",[28],{"code":195,"score":69,"supported":45,"title":196,"abstract":197,"tags":199},"A7",[],[198],"间作",[],{"code":201,"score":69,"supported":45,"title":202,"abstract":203,"tags":205},"F4",[],[204],"除草",[],{"code":207,"score":69,"supported":45,"title":208,"abstract":209,"tags":211},"H7",[],[210],"气候变化",[],{"code":75,"score":213,"supported":14,"title":214,"abstract":215,"tags":216},2,[],[],[26,123],[218,219],"农业人工智能 产量预测 智慧农业 精准农业","农业人工智能 产量预测","农业人工智能产量预测智慧农业精准农业-2518","10.18805\u002Fijare.a-6627",{"doi":221,"openalex_id":223,"authors":224,"venue":112,"cited_by_count":95,"oa_url":109,"card":227,"direction":232,"ingested_from":102},"W7212616413",[225],{"name":226,"orcid":9},"Manish Maan",{"tldr":228,"method":229,"finding":230,"direction":100,"opportunity":231},"系统综述AI在可持续农业中的应用，分析生产力与可持续性影响及推广障碍。","系统文献综述，围绕作物、土壤、水、植保和产量预测梳理ML、IoT、机器人、计算机","作物与土壤监测、产量预测和机器人田间作业收益最大，检测精度达95.78%，需可解释AI与互操作系统。","可解释AI与互操作农场系统在农户信任和气候适应中的作用尚缺实证，是值得深入的研究空白。","智慧农业 \u002F 农业物联网","2026-09-15T23:30:13.705421Z",{"id":235,"title":236,"url":237,"summary":238,"summary_zh":239,"content":9,"source_name":240,"source_url":237,"published_at":241,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":242,"score_detail":243,"sources":246,"tags":248,"primary_topic":250,"secondary_topics":251,"classification":252,"search_phrases":268,"slug":271,"view_count":95,"doi":272,"paper":273,"created_at":287},5587,"Calibration-Shift Inference for Computer Vision Agriculture Yield Models Under Retrofit Sensor Replacement: An Artificial Intelligence Post-Deployment Recalibration Guarantee","https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23544195","Computer vision yield models deployed in agricultural fields are routinely paired with retrofit sensor packages that replace or supplement the original imaging hardware, and the resulting calibration shift can silently invalidate the model's predictive guarantees. The analysis characterises the operating characteristics of the design under a declared variance structure with between-cluster standard deviation 6.0, residual standard deviation 4.5 and intraclass correlation 0.64, using 2000 replications per design cell across cluster counts of 8, 16, 24 and 32 and effect sizes of 3, 5 and 8 percentage points. The design attains 80% power at a 5 percentage point effect with 54 clusters. The paper states a formal guarantee: under stated assumptions on the stability of the calibration map, the before-after-control-impact estimator yields a consistent and asymptotically normal estimate of the calibration shift, and the associated confidence interval covers the true shift at the nominal rate. The contribution is a design-theoretic account of how retrofit-induced calibration shift can be detected and bounded without observational data, together with the sample-size requirements that follow.","部署在农田中的计算机视觉产量模型通常与改装传感器组件配合使用，这些组件用于替换或补充原始成像硬件，由此产生的校准偏移可能悄然使模型的预测保证失效。该分析在声明的方差结构下刻画了设计的工作特性，其中组间标准差为6.0、残差标准差为4.5、组内相关系数为0.64，每个设计单元采用2000次重复，聚类数分别为8、16、24和32，效应量为3、5和8个百分点。该设计在54个聚类、5个百分点效应下达到80%的检验力。论文给出了一项形式化保证：在关于校准映射稳定性的既定假设下，前后对照影响（before-after-control-impact）估计量给出校准偏移的一致且渐近正态的估计，且相关置信区间以名义覆盖率覆盖真实偏移。其贡献在于从设计理论角度说明了如何在没有观测数据的情况下检测并界定改装引起的校准偏移，以及随之而来的样本量要求。","Zenodo (CERN European Organization for Nuclear Research)","2026-10-21T00:00:00Z",65,{"impact":17,"substance":18,"depth":117,"authority":244,"freshness":13,"relevant":21,"comment":245},12,"方法学扎实、统计设计严谨的预印本论文，但偏理论设计层面，缺乏田间实证数据，产业影响有限，适合作为技术参考而非每日精选头条。",[247],{"name":240,"url":237},[26,27,145,28,249],"传感器校准",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[],{"status":39,"locked":14,"version":40,"method":41,"evidence":253},{"candidates":254,"agricultural_evidence":45},[255,260,264],{"code":32,"score":256,"supported":45,"title":257,"abstract":258,"tags":259},24,[146,147],[145,147],[27,145],{"code":75,"score":21,"supported":14,"title":261,"abstract":262,"tags":263},[],[],[26],{"code":80,"score":21,"supported":14,"title":265,"abstract":266,"tags":267},[],[],[28],[269,270],"计算机视觉 产量预测 传感器校准","农业人工智能 传感器校准 计算机视觉 产量预测","计算机视觉产量预测传感器校准-5587","10.5281\u002Fzenodo.23544195",{"doi":272,"openalex_id":274,"authors":275,"venue":240,"cited_by_count":95,"oa_url":237,"card":282,"direction":100,"ingested_from":102},"W7223104380",[276,278,280],{"name":277,"orcid":9},"Tigist Alemu Gebremariam",{"name":279,"orcid":9},"Hawi Tesfaye Desta",{"name":281,"orcid":9},"Mulugeta Bekele Wolde",{"tldr":283,"method":284,"finding":285,"direction":100,"opportunity":286},"提出改造传感器替换后视觉产量模型校准漂移的检测与置信区间保证方法。","用前后对照影响估计量，基于方差结构做2000次重复的模拟设计分析。","5个百分点效应下54个集群可达80%检验力，估计量一致且渐近正态。","可将该设计理论扩展到多传感器异构替换与在线校准监测，填补真实田间验证空白。","2026-10-22T11:21:01.554995Z",{"id":289,"title":290,"url":291,"summary":292,"summary_zh":293,"content":9,"source_name":240,"source_url":291,"published_at":241,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":294,"score_detail":295,"sources":297,"tags":299,"primary_topic":301,"secondary_topics":302,"classification":307,"search_phrases":323,"slug":325,"view_count":95,"doi":326,"paper":327,"created_at":338},5570,"Evaluation of Counterfactual Explanation Robustness for Computer Vision Agriculture Yield-Forecast Models Under Satellite Cloud-Mask Gaps: An Interpretability Reliability Bound for Cooperative Storage Decisions","https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23545079","Counterfactual explanations are increasingly proposed as the interface through which cooperative storage managers interrogate computer-vision yield-forecast models, yet their reliability under the satellite cloud-mask gaps that routinely afflict agricultural remote sensing has not been bounded. The bound is derived from a Lipschitz continuity assumption on the forecast model and a masking-operator perturbation analysis, and it is expressed as a function of the cloud-mask gap fraction, the model's local Lipschitz constant, and the counterfactual search tolerance. A replication-based evaluation of the before-after-control-impact estimator was conducted to characterise the operating characteristics of the design under a declared variance structure (between-cluster SD = 6.0, residual SD = 4.5, intraclass correlation = 0.64), with power estimated across 2000 design cells. The reliability bound degrades non-linearly as the gap fraction rises, and that the intraclass correlation of 0.64 implies that cooperative-level aggregation of counterfactual recourse is necessary to preserve explanation stability. The contribution is a decision-relevant reliability bound that storage cooperatives can use to set a maximum tolerable cloud-mask gap before counterfactual recourse is suspended.","反事实解释日益被提出作为合作式仓储管理者质询计算机视觉产量预测模型的接口，然而在农业遥感中经常出现的卫星云掩膜缺口条件下，其可靠性尚未得到界定。该界基于预测模型上的Lipschitz连续性假设和掩膜算子扰动分析推导得出，并表示为云掩膜缺口比例、模型局部Lipschitz常数和反事实搜索容差的函数。对前后对照影响估计量进行了基于重复的评估，以表征该设计在给定方差结构（组间标准差=6.0，残差标准差=4.5，组内相关系数=0.64）下的运行特征，并在2000个设计单元中估计了检验效能。可靠性界随缺口比例上升呈非线性退化，且组内相关系数0.64意味着需要对反事实救济进行合作社层面的聚合，以保持解释的稳定性。本文的贡献在于提供了一个与决策相关的可靠性界，仓储合作社可据此设定在暂停反事实救济之前可容忍的最大云掩膜缺口。",67,{"impact":17,"substance":116,"depth":117,"authority":244,"freshness":13,"relevant":21,"comment":296},"该论文提出反事实解释在卫星云掩膜缺口下的可靠性边界，方法新颖且结论可用于合作社决策，但属细分方法学研究，产业影响有限。",[298],{"name":240,"url":291},[26,27,28,192,300],"可解释性",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[303],{"code":80,"slug":304,"name":305,"path":306,"parent_code":36},"remote-sensing","遥感与农情监测","\u002Ftopic\u002Fdigital-agriculture\u002Fremote-sensing",{"status":39,"locked":14,"version":40,"method":41,"evidence":308},{"candidates":309,"agricultural_evidence":45},[310,314,319],{"code":32,"score":244,"supported":45,"title":311,"abstract":312,"tags":313},[147],[145],[27],{"code":80,"score":17,"supported":45,"title":315,"abstract":316,"tags":318},[],[192,317],"remote sensing",[28,192],{"code":75,"score":21,"supported":14,"title":320,"abstract":321,"tags":322},[],[],[26],[324,219],"农业人工智能 产量预测 可解释性 智慧农业","农业人工智能产量预测可解释性智慧农业-5570","10.5281\u002Fzenodo.23545079",{"doi":326,"openalex_id":328,"authors":329,"venue":240,"cited_by_count":95,"oa_url":291,"card":332,"direction":336,"ingested_from":102},"W7223114002",[330],{"name":331,"orcid":9},"Kadiatou Barry",{"tldr":333,"method":334,"finding":335,"direction":336,"opportunity":337},"推导卫星云掩膜缺口下反事实解释可靠性的理论界，并评估其对合作社储粮决策的影响。","基于Lipschitz连续性假设与掩膜扰动分析推导可靠性界，并用2000个设计单","缺口比例上升使可靠性非线性下降；ICC=0.64表明需合作社层级聚合以保持解释稳定。","农业遥感与作物表型","可研究不同作物区云掩膜统计特性下的自适应解释暂停阈值与聚合策略。","2026-10-22T11:20:39.240791Z",{"id":340,"title":341,"url":342,"summary":343,"summary_zh":344,"content":9,"source_name":345,"source_url":342,"published_at":346,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":114,"score_detail":347,"sources":350,"tags":352,"primary_topic":354,"secondary_topics":355,"classification":356,"search_phrases":381,"slug":384,"view_count":95,"doi":385,"paper":386,"created_at":414},5080,"Explainable deep learning for Mango leaf disease recognition using AlexNet and grad-CAM","https:\u002F\u002Fdoi.org\u002F10.1177\u002F1088467x261495708","Early and accurate detection of plant diseases is essential for improving crop productivity and supporting sustainable agricultural practices. Mango (Mangifera indica), one of the most economically important tropical fruit crops worldwide, is highly susceptible to several leaf diseases that can significantly reduce yield and fruit quality. This study presents an explainable deep learning framework for automated ten-class mango leaf disease classification using a publicly available dataset containing approximately 5000 images distributed across Anthracnose, Bacterial Canker, Cutting Weevil, Dag Disease, Die Back, Galls, Healthy, Powdery Mildew, Red Rust, and Sooty Mould classes. Unlike conventional AlexNet-based approaches, the proposed framework integrates Bayesian hyperparameter optimization, two-stage fine-tuning, test-time augmentation (TTA), and Grad-CAM-based explainability to improve both predictive performance and interpretability. Experimental results demonstrate that the proposed framework achieves a final test accuracy of 99.2% using eight-view test-time augmentation (TTA), with corresponding precision, recall, and F1-score of approximately 99.1%. The model also achieves micro-average and macro-average ROC–AUC scores of 1.00 on the curated test set, indicating strong discriminative capability across the considered disease categories. Comparative evaluations against MobileNetV2, DenseNet201, ConvNeXt, and LeafNet demonstrate competitive or superior classification performance while maintaining moderate computational complexity and architectural simplicity. Furthermore, Grad-CAM visualizations qualitatively highlight biologically relevant disease symptoms, including lesions, necrotic regions, rust spots, fungal growth, and tissue discoloration, thereby enhancing model interpretability. Although field-scale validation and quantitative explainability assessment remain subjects for future investigation, the proposed framework provides an accurate, computationally efficient, and interpretable approach for automated mango leaf disease diagnosis with strong potential for smart agriculture and precision farming applications.","植物病害的早期准确检测对于提高作物生产力和支持可持续农业实践至关重要。芒果（Mangifera indica）是全球最具经济重要性的热带水果作物之一，极易感染多种叶部病害，这些病害可显著降低产量和果实品质。本研究提出了一种可解释的深度学习框架，用于自动化十类芒果叶部病害分类，所使用的公开数据集包含约5000张图像，分布于炭疽病（Anthracnose）、细菌性溃疡病（Bacterial Canker）、切叶象甲（Cutting Weevil）、枯枝病（Dag Disease）、回枯病（Die Back）、虫瘿（Galls）、健康（Healthy）、白粉病（Powdery Mildew）、红锈病（Red Rust）和煤烟病（Sooty Mould）十个类别。与传统的基于AlexNet的方法不同，所提出的框架集成了贝叶斯超参数优化、两阶段微调、测试时增强（TTA）以及基于Grad-CAM的可解释性，以同时提升预测性能和可解释性。实验结果表明，所提出的框架在使用八视图测试时增强（TTA）时达到了99.2%的最终测试准确率，相应的精确率、召回率和F1分数约为99.1%。该模型在精选测试集上还取得了1.00的微平均和宏平均ROC–AUC分数，表明在所考虑的病害类别间具有较强的判别能力。与MobileNetV2、DenseNet201、ConvNeXt和LeafNet的对比评估表明，该框架在保持适中计算复杂度和架构简洁性的同时，具有竞争性或更优的分类性能。此外，Grad-CAM可视化定性地突出了具有生物学相关性的病害症状，包括病斑、坏死区域、锈斑、真菌生长和组织变色，从而增强了模型的可解释性。尽管田间规模验证和定量可解释性评估仍有待未来研究，但所提出的框架为自动化芒果叶部病害诊断提供了一种准确、计算高效且可解释的方法，在智慧农业和精准农业应用中具有强大潜力。","Intelligent Data Analysis","2026-10-15T00:00:00Z",{"impact":19,"substance":348,"depth":117,"authority":118,"freshness":20,"relevant":21,"comment":349},21,"将贝叶斯调参、两阶段微调与Grad-CAM结合用于十类芒果叶病识别，准确率99.2%，方法新颖且可解释性强，对智慧农业植保应用有参考价值。",[351],{"name":345,"url":342},[26,27,29,123,353],"芒果病害",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[],{"status":39,"locked":14,"version":40,"method":41,"evidence":357},{"candidates":358,"agricultural_evidence":45},[359,365,370,376],{"code":32,"score":19,"supported":45,"title":360,"abstract":362,"tags":364},[361],"deep learning",[363,361],"深度学习",[27,29],{"code":75,"score":141,"supported":45,"title":366,"abstract":367,"tags":369},[],[26,123,368],"smart agriculture",[26,123],{"code":371,"score":69,"supported":45,"title":372,"abstract":373,"tags":375},"B2",[],[374],"芒果",[],{"code":133,"score":69,"supported":45,"title":377,"abstract":378,"tags":380},[],[379],"可持续农业",[],[382,383],"芒果叶片病害 AlexNet","Grad-CAM 芒果病害识别","芒果叶片病害AlexNet-5080","10.1177\u002F1088467x261495708",{"doi":385,"openalex_id":387,"authors":388,"venue":345,"cited_by_count":95,"oa_url":9,"card":409,"direction":232,"ingested_from":102},"W7222328160",[389,392,395,397,400,403,406],{"name":390,"orcid":391},"Anay Ghosh","https:\u002F\u002Forcid.org\u002F0009-0006-6237-7905",{"name":393,"orcid":394},"Tawfikur Rahman","https:\u002F\u002Forcid.org\u002F0000-0002-0456-5143",{"name":396,"orcid":9},"Md Moniruzzaman",{"name":398,"orcid":399},"Nibedita Deb","https:\u002F\u002Forcid.org\u002F0000-0002-1579-1138",{"name":401,"orcid":402},"Hanan Abdullah Mengash","https:\u002F\u002Forcid.org\u002F0000-0002-4103-2434",{"name":404,"orcid":405},"Noorlindawaty Md Jizat","https:\u002F\u002Forcid.org\u002F0000-0001-7618-2283",{"name":407,"orcid":408},"Samir Salem Al‐Bawri","https:\u002F\u002Forcid.org\u002F0000-0003-2852-575X",{"tldr":410,"method":411,"finding":412,"direction":100,"opportunity":413},"提出可解释深度学习框架，用AlexNet结合贝叶斯调参、微调和Grad-CAM实现芒果叶病十分类。","AlexNet+贝叶斯超参优化、两阶段微调、八视图TTA、Grad-CAM，约5","测试准确率达99.2%，ROC-AUC为1.00，Grad-CAM能定位病斑等生物相关症状。","可延伸至田间复杂背景验证、可解释性定量评估及轻量化边缘部署研究。","2026-10-16T23:30:08.303796Z",{"id":416,"title":417,"url":418,"summary":419,"summary_zh":420,"content":9,"source_name":421,"source_url":418,"published_at":346,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":422,"score_detail":423,"sources":425,"tags":427,"primary_topic":430,"secondary_topics":431,"classification":432,"search_phrases":462,"slug":465,"view_count":21,"doi":466,"paper":467,"created_at":482},5076,"Trustworthy Explainable AI for Lettuce Growth Stage Prediction Using Sensor Data in Smart Agriculture","https:\u002F\u002Fdoi.org\u002F10.20944\u002Fpreprints202610.1452.v1","Background In precision agriculture, especially in hydroponic systems where environmental management affects both production and resource efficiency, interpretable modeling of crop phenological stages is crucial. Typical visualization-based approaches are generally expensive and ecologically dependent. To classify and visualize lettuce growth stages (Seedling, Vegetative Growth, Heading, Bolting and Senescence) using temperature, humidity, total dissolved solids (TDS) and pH, this study proposed XGB-StageVision, a sensor-powered machine learning setup combining supervised and unsupervised learning techniques. Using XG Boost, LightGBM, CatBoost, ExtraTrees, Random Forest and LSTM under an 80\u002F20 train-test split, a dataset of 3,169 unique observations from 70 hydroponically grown lettuce plants was evaluated. Methods and Results XGB-StageVision achieved 99.86% test accuracy, a ROC AUC of 1.00, an MCC of 0.9982 and perfect precision, recall, F1 scores and specificity across all classes, performing competitively with the best baseline models, including LightGBM, which achieved statistically comparable cross validation performance. The unsupervised t-SNE visualization showed clearly distinct structural clusters across growth stages, while SHAP revealed temperature as the strongest predictor (≈75% contribution), followed by TDS. Novel Contributions SHAP, LIME and t-SNE provided global, local and structural interpretability and transparent perspectives on environmental influences regarding plant development. Statistical analysis and 10-fold cross-validation were used to further test the model’s robustness, reliability and generalizability. Conclusion XGB-StageVision provides an IoT-ready, interpretable and imaging-free system for monitoring lettuce growth stages in real time. The framework combines supervised predictive modeling with unsupervised structural visualization to ensure transparency, sustainability and high performance, supporting resource-efficient, data-driven strategic planning in contemporary hydroponic agriculture to lower monitoring costs, optimize inputs and improve yield consistency.","背景 在精准农业中，尤其是在环境管理同时影响产量和资源效率的水培系统中，作物物候阶段的可解释建模至关重要。典型的基于可视化的方法通常成本高昂且依赖生态条件。为了利用温度、湿度、总溶解固体（TDS）和pH对生菜生长阶段（幼苗期、营养生长期、结球期、抽薹期和衰老期）进行分类和可视化，本研究提出了XGB-StageVision，一种由传感器驱动的机器学习方案，结合了监督学习和无监督学习技术。在80\u002F20的训练-测试划分下，使用XGBoost、LightGBM、CatBoost、ExtraTrees、随机森林和LSTM，对来自70株水培生菜的3，169个独特观测值数据集进行了评估。方法与结果 XGB-StageVision达到了99.86%的测试准确率、1.00的ROC AUC、0.9982的MCC，并在所有类别上实现了完美的精确率、召回率、F1分数和特异性，与最佳基线模型表现相当，包括LightGBM，其交叉验证性能在统计上具有可比性。无监督t-SNE可视化显示了各生长阶段之间明显不同的结构聚类，而SHAP揭示温度是最强的预测因子（约75%贡献），其次是TDS。新颖贡献 SHAP、LIME和t-SNE提供了全局、局部和结构可解释性，以及关于环境对植物发育影响的可透明视角。使用统计分析和10折交叉验证进一步检验了模型的稳健性、可靠性和泛化性。结论 XGB-StageVision提供了一种物联网就绪、可解释且无需成像的系统，用于实时监测生菜生长阶段。该框架将监督预测建模与无监督结构可视化相结合，以确保透明度、可持续性和高性能，支持当代水培农业中资源高效、数据驱动的战略规划，以降低监测成本、优化投入并提高产量一致性。","Preprints.org",64,{"impact":244,"substance":116,"depth":117,"authority":54,"freshness":20,"relevant":21,"comment":424},"预印本论文，方法新颖、数据规模可观且强调可解释性，但尚未经同行评审，属细分领域技术进展。",[426],{"name":421,"url":418},[26,27,29,428,429],"水培生菜","传感器监测",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[],{"status":39,"locked":14,"version":40,"method":41,"evidence":433},{"candidates":434,"agricultural_evidence":45},[435,441,445,450,456],{"code":75,"score":436,"supported":45,"title":437,"abstract":438,"tags":440},15,[368],[123,439],"precision agriculture",[26],{"code":32,"score":17,"supported":45,"title":442,"abstract":443,"tags":444},[],[49,47],[27,29],{"code":182,"score":54,"supported":45,"title":446,"abstract":447,"tags":449},[],[185,448],"IoT",[],{"code":451,"score":69,"supported":45,"title":452,"abstract":453,"tags":455},"B1",[],[454],"生菜",[],{"code":457,"score":69,"supported":45,"title":458,"abstract":459,"tags":461},"B7",[],[460],"hydroponic",[],[463,464],"XGB-StageVision 生菜 生长阶段","水培生菜 传感器 生长预测","XGB-StageVision生菜生长阶段-5076","10.20944\u002Fpreprints202610.1452.v1",{"doi":466,"openalex_id":468,"authors":469,"venue":421,"cited_by_count":95,"oa_url":418,"card":477,"direction":232,"ingested_from":102},"W7222080602",[470,472,475],{"name":471,"orcid":9},"Muhammad Majid",{"name":473,"orcid":474},"Attia Ghafoor","https:\u002F\u002Forcid.org\u002F0009-0000-3828-6479",{"name":476,"orcid":9},"Amna Munir",{"tldr":478,"method":479,"finding":480,"direction":232,"opportunity":481},"提出XGB-StageVision，用传感器数据与可解释AI预测水培生菜生长阶段。","XGBoost等模型结合SHAP、LIME、t-SNE，基于3169条温湿度TD","测试准确率99.86%，温度贡献约75%，t-SNE显示各生长阶段聚类清晰。","可探索多作物、多环境迁移与低成本传感器部署下的可解释模型泛化能力。","2026-10-16T23:30:08.020788Z",{"id":484,"title":485,"url":486,"summary":487,"summary_zh":488,"content":9,"source_name":489,"source_url":486,"published_at":490,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":491,"score_detail":492,"sources":496,"tags":498,"primary_topic":499,"secondary_topics":500,"classification":501,"search_phrases":524,"slug":527,"view_count":95,"doi":528,"paper":529,"created_at":545},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":18,"substance":493,"depth":18,"authority":494,"freshness":20,"relevant":21,"comment":495},23,14,"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。",[497],{"name":489,"url":486},[26,27,145,28,374],{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[],{"status":39,"locked":14,"version":40,"method":41,"evidence":502},{"candidates":503,"agricultural_evidence":45},[504,510,516,520],{"code":32,"score":505,"supported":45,"title":506,"abstract":507,"tags":509},19,[361],[363,145,361,147,508],"YOLO",[27,145],{"code":371,"score":511,"supported":45,"title":512,"abstract":513,"tags":515},7,[],[514,374],"果树",[374],{"code":75,"score":21,"supported":14,"title":517,"abstract":518,"tags":519},[],[],[26],{"code":80,"score":21,"supported":14,"title":521,"abstract":522,"tags":523},[],[],[28],[525,526],"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":528,"openalex_id":530,"authors":531,"venue":489,"cited_by_count":95,"oa_url":539,"card":540,"direction":232,"ingested_from":102},"W7221685456",[532,534,537],{"name":533,"orcid":9},"Víctor Yoel Huamán Asencio",{"name":535,"orcid":536},"Heber I. Mejía-Cabrera","https:\u002F\u002Forcid.org\u002F0000-0002-0007-0928",{"name":538,"orcid":9},"Juan Carlos Arcila Diaz","https:\u002F\u002Fwww.frontiersin.org\u002Fjournals\u002Fplant-science\u002Farticles\u002F10.3389\u002Ffpls.2026.1943711\u002Fpdf",{"tldr":541,"method":542,"finding":543,"direction":100,"opportunity":544},"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"]