[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-463":3,"related-463":82},{"id":4,"title":5,"url":6,"summary":7,"summary_zh":8,"content":9,"source_name":10,"source_url":8,"published_at":11,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":15,"score_detail":16,"sources":22,"tags":24,"primary_topic":29,"secondary_topics":35,"classification":36,"search_phrases":77,"slug":80,"view_count":50,"doi":8,"paper":8,"created_at":81},463,"Adaptive Generalized Regressive Deep Convolutional Reinforcement Learning for Crop Yield Prediction in Smart Precision Farming","https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41598-026-66501-5","本文提出自适应广义回归深度卷积强化学习(AGR-DCRL)模型,用于智慧精确农业作物产量预测。模型由输入层、卷积层、池化层和密集层组成,IoT技术采集天气、土壤、农药使用数据。在卷积层,采用加权局部相似性插补和广义Tietjen-Moore检验进行缺失值处理和异常值检测;在池化层,采用Camargo自适应多样性指数进行特征选择;在密集层,通过多项式逻辑回归进行产量预测,采用softmax激活函数生成多类预测结果。实验结果表明,AGR-DCRL模型相比传统深度学习方法准确率提高4%,误差率降低69%,预测时间快20%。原文标题:Adaptive generalized regressive deep convolutional reinforcement learning for crop yield prediction in smart precision farming,作者:Preethi P、Devadas R M,适用于大规模产量预测数据集。",null,"*   Article\n*   [Open access](https:\u002F\u002Fwww.springernature.com\u002Fgp\u002Fopen-science\u002Fabout\u002Fthe-fundamentals-of-open-access-and-open-research)\n*   Published: 12 August 2026\n\n*   [Preethi Preethi](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41598-026-66501-5#auth-Preethi-Preethi-Aff1)[1](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41598-026-66501-5#Aff1)&\n*   [Raghavendra M. Devadas](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41598-026-66501-5#auth-Raghavendra_M_-Devadas-Aff1)[1](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41598-026-66501-5#Aff1)\n\n[_Scientific Reports_](https:\u002F\u002Fwww.nature.com\u002Fsrep) (2026) [Cite this article](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41598-026-66501-5#citeas)\n\nWe are providing an unedited version of this manuscript to give early access to its findings. Before final publication, the manuscript will undergo further editing. Please note there may be errors present which affect the content, and all legal disclaimers apply.\n\n## Abstract\n\nAccurate crop yield assessment is essential for improving agricultural productivity and resource management. However, yield prediction is challenging due to multiple influencing factors such as soil properties, weather conditions, crop practices, pests, and diseases. While deep learning techniques have improved prediction capabilities, achieving high accuracy with low error and reduced computation time remains a key concern for large-scale datasets. To address this, an Adaptive Generalized Regressive Deep Convolutional Reinforcement Learning (AGR-DCRL) model is proposed to enhance prediction accuracy for smart farming. The model consists of input, convolution, pooling and dense layers. Data is first collected in data harvesting using IoT technologies that monitor weather, soil conditions, and pesticide usage. During data augmentation, Adaptive proximity sampling process is utilized in input layer to create new data samples. In the convolution layer, preprocessing is performed by using weighted local similarity-based imputation and Generalized Tietjen–Moore test. The missing values are handled and outlier’s are detected. Feature selection is applied in pooling layer using Camargo’s adaptive diversity index to select the most relevant features and remove irrelevant features for reducing the dimensionality. Crop yield prediction is performed at dense layer for analyzing both extracted features and data samples via polytomous logistic regression. The output layer employs a softmax activation function to create multi-class prediction results. Then, the error rate is measured with each prediction outcome and rewards. The Q-values are iteratively updated based on rewards until the model converges. Lastly, the accurate crop yield prediction is obtained with higher accuracy and lesser error. Experimental assessment of the proposed technique is implanted in Python using Smart Farming Sensor Data for Yield Prediction dataset with several metrics. Experimental results demonstrate that the AGR-DCRL model achieves higher accuracy by 4%, lesser error rates by 69%, and faster prediction time by 20% compared to conventional deep learning approaches.\n\n## Funding\n\nOpen access funding provided by Manipal Academy of Higher Education, Manipal. The authors affirm that this research work did not receive any funding to carry out the work.\n\n## Author information\n\n### Authors and Affiliations\n\n1.   Manipal Institute of Technology Bengaluru, Manipal Academy of Higher Education, Manipal, India\n\nPreethi Preethi&Raghavendra M. Devadas\n\nAuthors\n\n1.   Preethi Preethi\n2.   Raghavendra M. Devadas\n\n### Corresponding author\n\nCorrespondence to [Raghavendra M. Devadas](mailto:raghavendra.devadas@manipal.edu).\n\n## Ethics declarations\n\n### Competing interests\n\nThe authors declare no competing interests.\n\n## Additional information\n\n### Publisher’s note\n\nSpringer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.\n\n## Rights and permissions\n\n**Open Access** This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit [http:\u002F\u002Fcreativecommons.org\u002Flicenses\u002Fby\u002F4.0\u002F](http:\u002F\u002Fcreativecommons.org\u002Flicenses\u002Fby\u002F4.0\u002F).\n\n[Reprints and permissions](https:\u002F\u002Fs100.copyright.com\u002FAppDispatchServlet?title=Adaptive%20generalized%20regressive%20deep%20convolutional%20reinforcement%20learning%20for%20crop%20yield%20prediction%20in%20smart%20precision%20farming&author=Preethi%20Preethi%20et%20al&contentID=10.1038%2Fs41598-026-66501-5&copyright=The%20Author%28s%29&publication=2045-2322&publicationDate=2026-08-12&publisherName=SpringerNature&orderBeanReset=true&oa=CC%20BY)\n\n## About this article\n\n[![Image 1: Check for updates. Verify currency and authenticity via CrossMark](blob:http:\u002F\u002Flocalhost\u002Fdf43c82f6cb6dcd2a87db38f317bf9f2)](https:\u002F\u002Fcrossmark.crossref.org\u002Fdialog\u002F?doi=10.1038\u002Fs41598-026-66501-5)\n\n### Cite this article\n\nPreethi, P., Devadas, R.M. Adaptive generalized regressive deep convolutional reinforcement learning for crop yield prediction in smart precision farming. _Sci Rep_ (2026). https:\u002F\u002Fdoi.org\u002F10.1038\u002Fs41598-026-66501-5\n\n[Download citation](https:\u002F\u002Fcitation-needed.springer.com\u002Fv2\u002Freferences\u002F10.1038\u002Fs41598-026-66501-5?format=refman&flavour=citation)\n\n*   Received: 30 March 2026\n\n*   Accepted: 07 August 2026\n\n*   Published: 12 August 2026\n\n*   DOI: https:\u002F\u002Fdoi.org\u002F10.1038\u002Fs41598-026-66501-5\n\n### Keywords","Nature Scientific Reports 2026-08-12","2026-08-12T00:00:00Z","论文",10,false,78,{"impact":17,"substance":18,"depth":19,"authority":17,"freshness":13,"relevant":20,"comment":21},15,20,18,1,"提出AGR-DCRL新模型提升作物产量预测精度，方法新颖且实验数据详实，对智慧农业有实质贡献。",[23],{"name":10,"url":6},[25,26,27,28],"产量预测","深度学习","精准农业","强化学习",{"code":30,"slug":31,"name":32,"path":33,"parent_code":34},"I2","agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai","I",[],{"status":37,"locked":14,"version":38,"method":39,"evidence":40},"classified","2026-10-22.2","rules",{"candidates":41,"agricultural_evidence":44},[42,48,55,61,67,72],{"code":30,"score":43,"supported":44,"title":45,"abstract":46,"tags":47},5,true,[],[26],[28,26],{"code":49,"score":50,"supported":44,"title":51,"abstract":52,"tags":54},"F4",3,[],[53],"农药",[],{"code":56,"score":50,"supported":44,"title":57,"abstract":58,"tags":60},"H2",[],[59],"土壤",[],{"code":62,"score":50,"supported":44,"title":63,"abstract":64,"tags":66},"I5",[],[65],"IoT",[],{"code":68,"score":20,"supported":14,"title":69,"abstract":70,"tags":71},"I1",[],[],[27],{"code":73,"score":20,"supported":14,"title":74,"abstract":75,"tags":76},"I3",[],[],[25],[78,79],"产量预测 强化学习 深度学习 精准农业","产量预测 强化学习","产量预测强化学习深度学习精准农业-463","2026-08-12T23:57:24.202258Z",{"total":83,"page":20,"page_size":83,"items":84},6,[85,158,231,294,387,442],{"id":86,"title":87,"url":88,"summary":89,"summary_zh":90,"content":8,"source_name":91,"source_url":88,"published_at":92,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":93,"score_detail":94,"sources":99,"tags":101,"primary_topic":105,"secondary_topics":106,"classification":107,"search_phrases":137,"slug":140,"view_count":141,"doi":142,"paper":143,"created_at":157},4912,"Deep Learning Approaches for Plant Disease Detection: A Review of Convolutional Neural Networks and Emerging Techniques","https:\u002F\u002Fdoi.org\u002F10.56201\u002Fijcsmt.vol.11.no12.2025.pg220.231","Plant diseases pose a significant threat to global food security, crop yield, and agricultural sustainability. Traditional disease diagnosis methods, relying on expert visual inspection and laboratory analyses, are often time-consuming, labor-intensive, and costly, limiting their scalability. Recent advances in artificial intelligence, particularly deep learning, offer automated and accurate solutions for plant disease detection. This review synthesizes current research on deep learning–based approaches, focusing on convolutional neural networks (CNNs), hybrid CNN-transformer models, attention mechanisms, and generative data augmentation techniques. CNNs have demonstrated strong capabilities in extracting hierarchical features from leaf images, capturing complex patterns such as lesion shapes, color variations, and textures, thereby outperforming conventional machine learning and rule-based methods. Hybrid architectures and transformer-based models enhance feature extraction, particularly for subtle or visually similar disease symptoms. Data augmentation strategies, including generative adversarial networks (GANs), help address class imbalance and improve model generalization. Despite high reported accuracies, challenges remain, including limited dataset diversity, environmental variability, overfitting, lack of explainable AI, and computational constraints for mobile deployment. Future research should prioritize explainable models, multi-crop and multi-disease detection, lightweight architectures suitable for field applications, and integration with precision agriculture systems. Overall, deep learning–based plant disease detection presents a transformative approach for early diagnosis, timely intervention, and sustainable agricultural practices, offering the potential to reduce crop losses, enhance productivity, and contribute to global food security.","植物病害对全球粮食安全、作物产量和农业可持续性构成重大威胁。传统的病害诊断方法依赖专家视觉检查和实验室分析，往往耗时、费力且成本高昂，限制了其可扩展性。人工智能，特别是深度学习的最新进展，为植物病害检测提供了自动化且准确的解决方案。本文综述了基于深度学习方法的最新研究，重点关注卷积神经网络（CNN）、CNN-Transformer混合模型、注意力机制和生成式数据增强技术。CNN在从叶片图像中提取层次化特征方面展现出强大能力，能够捕捉病变形状、颜色变化和纹理等复杂模式，从而优于传统机器学习和基于规则的方法。混合架构和基于Transformer的模型增强了特征提取能力，尤其针对细微或视觉上相似的病害症状。数据增强策略，包括生成对抗网络（GAN），有助于解决类别不平衡问题并提高模型泛化能力。尽管已有研究报告了较高的准确率，但仍存在诸多挑战，包括数据集多样性有限、环境变异性、过拟合、缺乏可解释的人工智能以及移动端部署的计算约束。未来研究应优先关注可解释模型、多作物和多病害检测、适用于田间应用的轻量级架构以及与精准农业系统的集成。总体而言，基于深度学习的植物病害检测为早期诊断、及时干预和可持续农业实践提供了一种变革性方法，具有减少作物损失、提高生产力并促进全球粮食安全的潜力。","INTERNATIONAL JOURNAL OF COMPUTER SCIENCE AND MATHEMATICAL THEORY E-ISSN","2026-10-12T00:00:00Z",75,{"impact":19,"substance":18,"depth":95,"authority":96,"freshness":97,"relevant":20,"comment":98},17,12,8,"系统综述深度学习植物病害检测的方法进展与挑战，学术价值较高但偏理论，对产业落地有参考意义。",[100],{"name":91,"url":88},[102,103,26,27,104],"智慧农业","农业人工智能","植物病害检测",{"code":30,"slug":31,"name":32,"path":33,"parent_code":34},[],{"status":37,"locked":14,"version":38,"method":39,"evidence":108},{"candidates":109,"agricultural_evidence":44},[110,119,124,131],{"code":30,"score":111,"supported":44,"title":112,"abstract":114,"tags":118},19,[113],"deep learning",[26,115,116,113,117],"机器学习","machine learning","artificial intelligence",[103,26],{"code":68,"score":97,"supported":44,"title":120,"abstract":121,"tags":123},[],[27,122],"precision agriculture",[102,27],{"code":125,"score":83,"supported":44,"title":126,"abstract":127,"tags":130},"A8",[],[128,129],"粮食安全","food security",[],{"code":132,"score":50,"supported":44,"title":133,"abstract":134,"tags":136},"H8",[],[135],"可持续农业",[],[138,139],"植物病害检测 深度学习","卷积神经网络 作物病害识别","植物病害检测深度学习-4912",2,"10.56201\u002Fijcsmt.vol.11.no12.2025.pg220.231",{"doi":142,"openalex_id":144,"authors":145,"venue":91,"cited_by_count":148,"oa_url":149,"card":150,"direction":154,"ingested_from":156},"W7221637766",[146],{"name":147,"orcid":8},"NUHU ABDULLAHI",0,"https:\u002F\u002Fiiardjournals.org\u002Fget\u002FIJCSMT\u002FVOL. 11 NO. 12 2025\u002FDeep Learning Approaches for Plant 220-231.pdf",{"tldr":151,"method":152,"finding":153,"direction":154,"opportunity":155},"综述深度学习在植物病害检测中的应用，涵盖CNN、混合Transformer、注意力机制与生成式数据增","综述CNN、CNN-Transformer混合模型、注意力机制及GAN数据增强在","深度学习检测精度高，但仍受数据集多样性、环境变异、可解释性和移动部署算力限制。","农业人工智能与决策模型","可探索轻量化可解释模型、多作物多病害检测及与精准农业系统集成，解决田间部署难题。","openalex","2026-10-14T23:30:54.687367Z",{"id":159,"title":160,"url":161,"summary":162,"summary_zh":163,"content":8,"source_name":164,"source_url":161,"published_at":92,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":165,"score_detail":166,"sources":171,"tags":173,"primary_topic":175,"secondary_topics":181,"classification":183,"search_phrases":213,"slug":216,"view_count":148,"doi":217,"paper":218,"created_at":230},4801,"CROPINTEL DATASET: A COMPREHENSIVE AGROENVIRONMENTAL RESOURCE FOR CROP CLASSIFICATION, IRRIGATION PLANNING, AND YIELD ANALYSIS","https:\u002F\u002Fdoi.org\u002F10.30546\u002F09090.2025.210.005","The growing demand for sustainable agricultural intensification necessitates advanced tools for optimizing crop selection and resource management.This study presents an integrated analytical framework using machine learning to derive agro-environmental intelligence from the comprehensive CropIntel dataset.We evaluated a suite of over fifteen machine learning algorithms for three critical precision agriculture tasks: crop classification, seasonal irrigation requirement prediction, and yield potential forecasting.The results demonstrate the exceptional capability of tree-based models in this domain.Notably, Decision Tree, Bagging, and Gaussian Naïve Bayes classifiers achieved near-perfect accuracy (Acc≈99.7%) in identifying suitable crops based on climatic and edaphic conditions.For regression tasks, LightGBM and Histogram Gradient Boosting models proved most effective, explaining approximately 84% of the variance in irrigation needs (R²≈0.84)and about 70% of the variance in yield potential (R²≈0.696).These findings underscore the potential of machine learning to create powerful decision support systems that can guide crop selection, optimize water allocation, and provide reliable yield forecasts.This research contributes to a holistic, data-driven methodology that can enhance the efficiency and sustainability of agricultural practices.","对可持续农业集约化的需求日益增长，亟需先进工具来优化作物选择与资源管理。本研究提出一个集成的分析框架，利用机器学习从综合性的CropIntel数据集中提取农业环境智能信息。我们针对三个关键精准农业任务评估了超过十五种机器学习算法：作物分类、季节性灌溉需求预测和产量潜力预报。结果表明，基于树的模型在该领域具有卓越能力。值得注意的是，决策树、装袋法和高斯朴素贝叶斯分类器在基于气候和土壤条件识别适宜作物方面达到了近乎完美的准确率（Acc≈99.7%）。在回归任务中，LightGBM和直方图梯度提升模型最为有效，分别解释了约84%的灌溉需求方差（R²≈0.84）和约70%的产量潜力方差（R²≈0.696）。这些发现凸显了机器学习在构建强大决策支持系统方面的潜力，可指导作物选择、优化水资源分配并提供可靠的产量预报。本研究为一种整体性、数据驱动的方法论做出了贡献，有助于提升农业实践的效率与可持续性。","JOURNAL OF BAKU ENGINEERING UNIVERSITY- MATHEMATICS AND COMPUTER SCIENCE",72,{"impact":96,"substance":167,"depth":95,"authority":168,"freshness":169,"relevant":20,"comment":170},21,13,9,"构建CropIntel农业环境数据集并系统评测15种以上机器学习算法，在作物分类、灌溉需求与产量预测上给出可复现结论，方法扎实但属细分领域学术进展，产业影响有限。",[172],{"name":164,"url":161},[102,103,25,27,174],"灌溉决策",{"code":176,"slug":177,"name":178,"path":179,"parent_code":180},"H3","irrigation","水资源与灌溉","\u002Ftopic\u002Fenvironment\u002Firrigation","H",[182],{"code":30,"slug":31,"name":32,"path":33,"parent_code":34},{"status":37,"locked":14,"version":38,"method":39,"evidence":184},{"candidates":185,"agricultural_evidence":44},[186,192,197,201,205,209],{"code":176,"score":19,"supported":44,"title":187,"abstract":188,"tags":191},[177],[189,190,177],"灌溉","水资源",[174],{"code":30,"score":13,"supported":44,"title":193,"abstract":194,"tags":196},[],[115,116,195],"decision support",[103],{"code":68,"score":97,"supported":44,"title":198,"abstract":199,"tags":200},[],[27,122],[102,27],{"code":56,"score":50,"supported":44,"title":202,"abstract":203,"tags":204},[],[59],[],{"code":132,"score":50,"supported":44,"title":206,"abstract":207,"tags":208},[],[135],[],{"code":73,"score":20,"supported":14,"title":210,"abstract":211,"tags":212},[],[],[25],[214,215],"CropIntel 数据集 作物分类","机器学习 灌溉需求 预测","CropIntel数据集作物分类-4801","10.30546\u002F09090.2025.210.005",{"doi":217,"openalex_id":219,"authors":220,"venue":164,"cited_by_count":148,"oa_url":161,"card":224,"direction":229,"ingested_from":156},"W7221521521",[221],{"name":222,"orcid":223},"Artughrul Gayibov","https:\u002F\u002Forcid.org\u002F0009-0009-7349-0286",{"tldr":225,"method":226,"finding":227,"direction":154,"opportunity":228},"基于CropIntel数据集，用15+种机器学习算法实现作物分类、灌溉需求预测与产量潜力预测。","CropIntel农业环境数据集，对比15+种机器学习算法，含树模型与贝叶斯分类","树模型分类准确率近99.7%，LightGBM等回归模型对灌溉需求R²约0.84、产量约0.70。","可探索多任务联合建模与模型可解释性，并将预测结果嵌入实时灌溉决策系统。","智慧农业 \u002F 农业物联网","2026-10-13T23:30:08.688669Z",{"id":232,"title":233,"url":234,"summary":235,"summary_zh":236,"content":8,"source_name":237,"source_url":234,"published_at":238,"category":12,"cover_url":8,"hotness":239,"is_selected":14,"score":240,"score_detail":241,"sources":244,"tags":248,"primary_topic":250,"secondary_topics":256,"classification":258,"search_phrases":275,"slug":278,"view_count":20,"doi":279,"paper":280,"created_at":293},4241,"Detection of Apple Plant Diseases Using Leaf Images through Convolutional Neural Network","https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23142591","Plant diseases have major social, ecological and economic impact on agriculture and frequently results in large financial losses for farmers. Therefore early and precise disease detection is crucial for efficient crop management. The severity of leaf diseases and their direct impact on yield quality have been highlighted in recent studies. Both digital cameras and flatbed scanners can be used to take leaf images but they have certain drawbacks. Although scanners offer a more secure and regulated method of acquiring images they are comparatively slow and may cause edge shadows. This study uses a hybrid deep learning approach to identify diseases in apple plant leaves. To improve classification performance; the proposed approach combines the advantages of Residual Networks (ResNet), Genetic Algorithms (GA) and Convolutional Neural Networks (CNN). CNNs are used to extract features and classify diseases whereas ResNet enhances the learning efficiency of deeper networks. The feature set is optimized using a genetic algorithm which improves accuracy and lowers computational complexity. The model also calculates the percentage of the leaf area that is impacted. According to experimental results the proposed method detects leaf diseases with high accuracy of percentage and low computational cost making it a dependable and effective precision agriculture solution. The proposed model exhibited strong classification efficiency across various leaf conditions that pertaining an accuracy of 0.97 for healthy leaves, 0.994 for apple scab, 0.89 for rust, and 0.91 for powdery mildew.","植物病害对农业产生重大的社会、生态和经济影响，并经常给农民带来巨大的经济损失。因此，早期且精确的病害检测对于高效的作物管理至关重要。近期研究强调了叶部病害的严重程度及其对产量品质的直接影响。数码相机和平板扫描仪均可用于拍摄叶片图像，但它们各有一定缺陷。尽管扫描仪提供了更安全、更可控的图像获取方式，但其速度相对较慢，且可能产生边缘阴影。本研究采用混合深度学习方法识别苹果植株叶部病害。为提高分类性能，所提方法结合了残差网络（ResNet）、遗传算法（GA）和卷积神经网络（CNN）的优势。CNN用于提取特征并分类病害，而ResNet增强了更深网络的学习效率。利用遗传算法对特征集进行优化，从而提高准确率并降低计算复杂度。该模型还计算了受影响的叶面积百分比。实验结果表明，所提方法以高准确率和低计算成本检测叶部病害，使其成为可靠且有效的精准农业解决方案。所提模型在多种叶片条件下均表现出较强的分类效率，健康叶片准确率为0.97，苹果黑星病为0.994，锈病为0.89，白粉病为0.91。","Zenodo (CERN European Organization for Nuclear Research)","2026-10-04T00:00:00Z",25,67,{"impact":96,"substance":19,"depth":242,"authority":96,"freshness":169,"relevant":20,"comment":243},16,"基于ResNet+GA+CNN的苹果叶病识别研究，准确率数据具体，方法有增量，但属实验室论文，产业影响有限。",[245,246],{"name":237,"url":234},{"name":237,"url":247},"https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23142592",[102,103,26,27,249],"苹果病害",{"code":251,"slug":252,"name":253,"path":254,"parent_code":255},"B2","fruit","果树","\u002Ftopic\u002Fhorticulture\u002Ffruit","B",[257],{"code":30,"slug":31,"name":32,"path":33,"parent_code":34},{"status":37,"locked":14,"version":38,"method":39,"evidence":259},{"candidates":260,"agricultural_evidence":44},[261,267,271],{"code":251,"score":17,"supported":44,"title":262,"abstract":264,"tags":266},[263],"apple",[265,263],"苹果",[249],{"code":68,"score":97,"supported":44,"title":268,"abstract":269,"tags":270},[],[27,122],[102,27],{"code":30,"score":97,"supported":44,"title":272,"abstract":273,"tags":274},[],[26,113],[103,26],[276,277],"苹果叶片病害 卷积神经网络","苹果锈病 白粉病 识别","苹果叶片病害卷积神经网络-4241","10.5281\u002Fzenodo.23142591",{"doi":279,"openalex_id":281,"authors":282,"venue":237,"cited_by_count":148,"oa_url":234,"card":287,"direction":292,"ingested_from":156},"W7219637192",[283,285],{"name":284,"orcid":8},"Mohd Arshad Raza",{"name":286,"orcid":8},"Kiran Pandey",{"tldr":288,"method":289,"finding":290,"direction":154,"opportunity":291},"提出ResNet+GA+CNN混合模型，用苹果叶片图像检测病害并计算病斑面积占比。","ResNet提取深层特征，遗传算法优化特征集，CNN分类，苹果叶片图像数据。","健康叶、黑星病、锈病、白粉病准确率分别达0.97、0.994、0.89、0.91，计算成本低。","可探索轻量化模型在田间移动端实时检测，并融合多病害严重度分级与产量损失预测。","数字乡村与农业信息化","2026-10-05T23:30:30.830366Z",{"id":295,"title":296,"url":297,"summary":298,"summary_zh":299,"content":8,"source_name":300,"source_url":297,"published_at":301,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":302,"score_detail":303,"sources":307,"tags":309,"primary_topic":312,"secondary_topics":313,"classification":314,"search_phrases":355,"slug":358,"view_count":20,"doi":359,"paper":360,"created_at":386},3522,"An agronomically informed framework for temporally transferable yield prediction","https:\u002F\u002Fdoi.org\u002F10.1007\u002Fs11119-026-10454-2","Abstract Purpose Yield maps are essential for agricultural decision-making, yet existing methods are often crop- or machine-specific, limiting their broader applicability. Methods This study introduces an agronomically informed framework for temporally transferable yield prediction that combines optimal phenological moment (OPM) selection with a spatial yield redistribution strategy: density-based yield mapping (DBYM). The framework relies on vegetation indices derived from satellite imagery acquired at physiologically meaningful crop stages and on independently known field-scale production values to guide spatial yield allocation. The method was evaluated across 10 growing seasons involving coffee, sugarcane, and wheat under both manual and mechanized production systems in commercial fields located in Brazil and Australia. In addition to DBYM, a simple linear regression baseline using the same OPM-derived inputs was evaluated to investigate the contribution of agronomically informed temporal input selection itself. Results Results demonstrated that OPM-based inputs alone already provided strong predictive capability across seasons, while DBYM improved the spatial coherence of predictions, particularly under sparse calibration scenarios. The index of agreement ranged from 0.73 to 0.94, while Moran’s bivariate index was significant ( p \u003C 0.01) for all DBYM predictions, confirming spatial consistency between observed and predicted yield patterns. RMSE values were comparable to those reported in machine learning and deep learning yield prediction studies, ranging from 0.21 to 0.28 Mg ha⁻¹ for coffee, 5.43–6.22 Mg ha⁻¹ for sugarcane, 0.15–0.26 Mg ha⁻¹ for wheat in Brazil, and 0.44–0.61 Mg ha⁻¹ for wheat in Australia. Conclusion The results suggest that predictive performance can be affected by the agronomic relevance of acquisition timing. By combining agronomically informed temporal inputs with a flexible spatial redistribution framework, DBYM enables the extension of yield map time series across seasons and provides a scalable and resource-efficient alternative for precision agriculture applications in fields lacking dedicated yield monitoring systems.","摘要 目的 产量图对农业决策至关重要，但现有方法往往针对特定作物或机器，限制了其更广泛的适用性。方法 本研究引入了一个基于农学知识的时间可迁移产量预测框架，该框架将最佳物候时刻（OPM）选择与空间产量再分配策略——基于密度的产量制图（DBYM）相结合。该框架依赖于在具有生理意义的作物阶段获取的卫星影像所衍生的植被指数，以及独立已知的田块尺度产量值来指导空间产量分配。该方法在巴西和澳大利亚的商业田块中进行了评估，涵盖10个生长季，涉及咖啡、甘蔗和小麦在人工与机械化生产系统下的种植。除DBYM外，还评估了使用相同OPM衍生输入的简单线性回归基线，以探究基于农学知识的时间输入选择本身的贡献。结果 结果表明，仅基于OPM的输入已在各生长季提供了较强的预测能力，而DBYM改善了预测的空间一致性，尤其是在校准样本稀疏的情景下。一致性指数范围为0.73至0.94，同时所有DBYM预测的Moran双变量指数均显著（p \u003C 0.01），证实了观测产量模式与预测产量模式之间的空间一致性。RMSE值与机器学习和深度学习产量预测研究中报道的值相当，巴西咖啡为0.21至0.28 Mg ha⁻¹，甘蔗为5.43至6.22 Mg ha⁻¹，小麦为0.15至0.26 Mg ha⁻¹，澳大利亚小麦为0.44至0.61 Mg ha⁻¹。结论 结果表明，预测性能可能受到获取时机农学相关性的影响。通过将基于农学知识的时间输入与灵活的空间再分配框架相结合，DBYM能够将产量图时间序列扩展至多个生长季，并为缺乏专用产量监测系统的田块提供了一种可扩展且资源高效的精准农业应用替代方案。","Precision Agriculture","2026-09-25T00:00:00Z",80,{"impact":19,"substance":304,"depth":19,"authority":305,"freshness":97,"relevant":20,"comment":306},22,14,"提出农学信息驱动的跨季可迁移产量预测框架，多作物多国验证，方法新颖且实用，值得进入每日精选。",[308],{"name":300,"url":297},[102,25,27,310,311],"遥感估产","咖啡",{"code":30,"slug":31,"name":32,"path":33,"parent_code":34},[],{"status":37,"locked":14,"version":38,"method":39,"evidence":315},{"candidates":316,"agricultural_evidence":44},[317,321,325,332,339,345,351],{"code":30,"score":169,"supported":44,"title":318,"abstract":319,"tags":320},[],[26,115,116,113],[],{"code":68,"score":97,"supported":44,"title":322,"abstract":323,"tags":324},[],[27,122],[102,27],{"code":326,"score":83,"supported":44,"title":327,"abstract":328,"tags":331},"A2",[],[329,330],"小麦","wheat",[],{"code":333,"score":83,"supported":44,"title":334,"abstract":335,"tags":338},"A6",[],[336,337],"甘蔗","sugarcane",[],{"code":340,"score":83,"supported":44,"title":341,"abstract":342,"tags":344},"B3",[],[311,343],"coffee",[],{"code":346,"score":50,"supported":44,"title":347,"abstract":348,"tags":350},"F5",[],[349],"机械化",[],{"code":73,"score":141,"supported":14,"title":352,"abstract":353,"tags":354},[],[],[25,310],[356,357],"遥感 产量预测 咖啡 甘蔗 小麦","DBYM 产量制图 精准农业","遥感产量预测咖啡甘蔗小麦-3522","10.1007\u002Fs11119-026-10454-2",{"doi":359,"openalex_id":361,"authors":362,"venue":300,"cited_by_count":148,"oa_url":297,"card":380,"direction":154,"ingested_from":156},"W7214356320",[363,366,369,372,374,377],{"name":364,"orcid":365},"Ricardo Canal Filho","https:\u002F\u002Forcid.org\u002F0000-0001-8043-8325",{"name":367,"orcid":368},"José Paulo Molin","https:\u002F\u002Forcid.org\u002F0000-0001-7250-3780",{"name":370,"orcid":371},"Patrick Filippi","https:\u002F\u002Forcid.org\u002F0000-0003-3573-084X",{"name":373,"orcid":8},"Eudocio Rafael Otavio da Silva",{"name":375,"orcid":376},"Luiz Gustavo de Góes Sterle","https:\u002F\u002Forcid.org\u002F0000-0001-9748-3923",{"name":378,"orcid":379},"Marcelo Chan Fu Wei","https:\u002F\u002Forcid.org\u002F0000-0002-8242-8435",{"tldr":381,"method":382,"finding":383,"direction":384,"opportunity":385},"提出农学知情框架，结合最优物候期与密度产量制图，实现跨季节可迁移的产量预测。","利用卫星植被指数与最优物候期选择，结合密度产量制图DBYM，在巴西和澳大利亚多作","仅用最优物候期输入即可跨季节强预测，DBYM提升空间一致性，精度与机器学习方法相当。","农业遥感与作物表型","可探索将农学知情时间选择与深度时空模型结合，提升无产量监测系统田块的跨区域迁移能力。","2026-09-26T23:30:02.989455Z",{"id":388,"title":389,"url":390,"summary":391,"summary_zh":392,"content":8,"source_name":393,"source_url":390,"published_at":394,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":240,"score_detail":395,"sources":397,"tags":399,"primary_topic":401,"secondary_topics":402,"classification":403,"search_phrases":420,"slug":423,"view_count":50,"doi":424,"paper":425,"created_at":441},2963,"A comprehensive review of deep learning methods for weed classification in precision agriculture","https:\u002F\u002Fdoi.org\u002F10.1007\u002Fs44163-026-01916-7","Weeds today are among the factors contributing to low agricultural productivity. As the world’s population continues to grow, there is an urgent need to meet global food demand. Nigeria currently lacks sufficient crop production to feed its growing population, and weeds are among the core contributors to poor agricultural yield. This study conducts a comprehensive review of Deep Learning (DL) approaches to weed classification in precision agriculture, covering literature from 2018 to 2025, was carried out. We employed a mix of quantitative and qualitative methods in the course of this review paper. Our data source is centred on Scopus-indexed papers, published with Sensors, Electronics, and Agriculture in MDPI as well as IEEE, Thomson Reuters, and Springer. The study systematically reviewed and analysed machine learning (ML), DL, and instance segmentation techniques to identify the key technological and environmental barriers, such as data limitations, class imbalance, environmental variability, and model scalability issues that affect the effectiveness and efficiency of these models when deployed in real time. These findings show that while weed management models like the YOLO variants, ResNet, and Vision Transformers achieved high accuracy in training and testing, they are associated with several challenges in their real world-deployment, such as occlusion, small object detection, and environmental adaptability. Overall, this research provides recommended solutions to enhance model robustness, scalability, and efficiency. It further provides a summary of the current state and future directions for AI-driven weed management.","杂草是当前导致农业生产力低下的因素之一。随着世界人口持续增长，满足全球粮食需求已成为迫切任务。尼日利亚目前的作物产量不足以养活其不断增长的人口，而杂草是导致农业产量低下的核心因素之一。本研究对精准农业中基于深度学习（Deep Learning，DL）的杂草分类方法进行了全面综述，涵盖2018年至2025年的文献。在综述过程中，我们采用了定量与定性相结合的方法。数据来源集中于Scopus索引论文，这些论文发表于MDPI旗下的Sensors、Electronics和Agriculture，以及IEEE、Thomson Reuters和Springer。本研究系统综述并分析了机器学习（Machine Learning，ML）、深度学习及实例分割技术，以识别影响这些模型实时部署效果与效率的关键技术和环境障碍，如数据局限性、类别不平衡、环境变异性及模型可扩展性问题。研究结果表明，尽管YOLO系列、ResNet和视觉Transformer（Vision Transformer）等杂草管理模型在训练和测试中达到了较高精度，但在实际部署中仍面临诸多挑战，如遮挡、小目标检测和环境适应性等问题。总体而言，本研究提出了增强模型鲁棒性、可扩展性和效率的推荐解决方案，并进一步总结了人工智能驱动杂草管理的现状与未来方向。","Discover Artificial Intelligence","2026-09-18T00:00:00Z",{"impact":96,"substance":19,"depth":242,"authority":168,"freshness":97,"relevant":20,"comment":396},"系统综述2018—2025年深度学习杂草分类方法，指出遮挡、小目标与环境适应性等落地瓶颈，对农业AI研究有参考价值，但属综述类论文、非突破性成果。",[398],{"name":393,"url":390},[102,103,26,400,27],"杂草识别",{"code":30,"slug":31,"name":32,"path":33,"parent_code":34},[],{"status":37,"locked":14,"version":38,"method":39,"evidence":404},{"candidates":405,"agricultural_evidence":44},[406,412,416],{"code":30,"score":111,"supported":44,"title":407,"abstract":408,"tags":411},[113],[409,26,115,116,113,410],"目标检测","YOLO",[103,26],{"code":68,"score":242,"supported":44,"title":413,"abstract":414,"tags":415},[122],[27,122],[102,27],{"code":49,"score":20,"supported":14,"title":417,"abstract":418,"tags":419},[],[],[400],[421,422],"深度学习 杂草分类 精准农业","YOLO 杂草识别 模型部署","深度学习杂草分类精准农业-2963","10.1007\u002Fs44163-026-01916-7",{"doi":424,"openalex_id":426,"authors":427,"venue":393,"cited_by_count":148,"oa_url":390,"card":436,"direction":154,"ingested_from":156},"W7213558223",[428,430,432,434],{"name":429,"orcid":8},"Njoku Camillus Ekene",{"name":431,"orcid":8},"Francis A. Okoye",{"name":433,"orcid":8},"Ebere Uzoka Chidi",{"name":435,"orcid":8},"OGBU MARY NNENNA",{"tldr":437,"method":438,"finding":439,"direction":154,"opportunity":440},"综述2018-2025年深度学习杂草分类方法，分析技术瓶颈并给出改进建议。","混合定量定性法，基于Scopus及MDPI、IEEE等文献，分析ML、DL与实例","YOLO、ResNet、ViT等精度高，但实际部署受遮挡、小目标与环境适应性限制。","可研究轻量化、跨域自适应模型，解决小目标与遮挡下的实时杂草识别难题。","2026-09-19T23:30:56.875661Z",{"id":443,"title":444,"url":445,"summary":446,"summary_zh":447,"content":8,"source_name":448,"source_url":445,"published_at":394,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":449,"score_detail":450,"sources":452,"tags":454,"primary_topic":457,"secondary_topics":458,"classification":463,"search_phrases":486,"slug":489,"view_count":20,"doi":490,"paper":491,"created_at":511},2942,"Deep learning-driven multisource remote sensing image fusion: Advances, challenges, and future directions","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.engappai.2026.116319","Multisource remote sensing image fusion has become an important solution to a long-standing limitation in Earth observation: individual sensors rarely provide high spatial detail, rich spectral information, reliable structural sensitivity, and frequent temporal coverage at the same time. This review examines how deep learning and artificial intelligence are being used to integrate multispectral, hyperspectral, panchromatic, optical, and synthetic aperture radar imagery for more reliable interpretation of complex ground scenes. It provides a technical synthesis of convolutional neural networks, autoencoders, generative adversarial networks, transformer architectures, diffusion models, and hybrid model driven approaches, with attention to their fusion mechanisms, reconstruction behavior, computational demand, and suitability for operational use. Applications include land cover mapping, precision agriculture, environmental monitoring, urban analysis, disaster assessment, and defense related interpretation. Rather than treating each fusion task separately, this review connects sensor heterogeneity, spatial and spectral resolution trade offs, radiometric correction, geometric correction, registration, noise reduction, and fusion level design within a single framework. The analysis indicates that convolutional models remain effective for stable local detail recovery, adversarial learning can improve visual sharpness but may introduce spectral distortion, transformer models better capture long range spatial and spectral relationships, and diffusion models offer refined reconstruction at greater computational cost. The review further identifies open challenges involving misregistration, spectral bias, limited labeled data, weak generalization across sensors, high memory requirements, and limited interpretability. Future progress should prioritize sensor aware learning, self supervised training, uncertainty aware evaluation, lightweight deployment, and application oriented benchmarks to improve reliability in operational Earth observation.","多源遥感图像融合已成为解决地球观测领域一个长期局限的重要方案：单一传感器很少能够同时提供高空间细节、丰富光谱信息、可靠的结构敏感性以及频繁的时间覆盖。本文综述了如何利用深度学习和人工智能整合多光谱、高光谱、全色、光学和合成孔径雷达（synthetic aperture radar, SAR）影像，以更可靠地解译复杂地表场景。文章对卷积神经网络、自编码器、生成对抗网络、Transformer架构、扩散模型以及混合模型驱动方法进行了技术综合，重点关注其融合机制、重建行为、计算需求以及业务化适用性。应用领域包括土地覆盖制图、精准农业、环境监测、城市分析、灾害评估和国防相关解译。本文并非将每种融合任务分开处理，而是在一个统一框架内将传感器异质性、空间与光谱分辨率权衡、辐射校正、几何校正、配准、降噪和融合层级设计联系起来。分析表明，卷积模型在稳定的局部细节恢复方面仍然有效，对抗学习可以提升视觉锐度但可能引入光谱失真，Transformer模型能更好地捕捉长程空间与光谱关系，而扩散模型以更高的计算成本提供精细重建。本文进一步指出了涉及配准误差、光谱偏差、标注数据有限、跨传感器泛化能力弱、高内存需求以及可解释性有限等开放挑战。未来的进展应优先关注传感器感知学习、自监督训练、不确定性感知评估、轻量化部署以及面向应用的基准测试，以提高业务化地球观测的可靠性。","Engineering Applications of Artificial Intelligence",82,{"impact":19,"substance":304,"depth":111,"authority":305,"freshness":169,"relevant":20,"comment":451},"发表于核心期刊的综述，系统梳理深度学习多源遥感融合的方法、应用与挑战，对农业遥感与精准农业有直接参考价值，时效性强，值得进入每日精选。",[453],{"name":448,"url":445},[103,26,27,455,456],"遥感","多源数据融合",{"code":30,"slug":31,"name":32,"path":33,"parent_code":34},[459],{"code":73,"slug":460,"name":461,"path":462,"parent_code":34},"remote-sensing","遥感与农情监测","\u002Ftopic\u002Fdigital-agriculture\u002Fremote-sensing",{"status":37,"locked":14,"version":38,"method":39,"evidence":464},{"candidates":465,"agricultural_evidence":44},[466,470,477,482],{"code":30,"score":111,"supported":44,"title":467,"abstract":468,"tags":469},[113],[26,113,117],[103,26],{"code":73,"score":19,"supported":44,"title":471,"abstract":473,"tags":476},[472],"remote sensing",[455,472,474,475],"multispectral","hyperspectral",[455],{"code":68,"score":478,"supported":44,"title":479,"abstract":480,"tags":481},7,[],[27,122],[27],{"code":62,"score":20,"supported":14,"title":483,"abstract":484,"tags":485},[],[],[456],[487,488],"多源遥感 图像融合 深度学习","农业人工智能 多源数据融合 深度学习 精准农业","多源遥感图像融合深度学习-2942","10.1016\u002Fj.engappai.2026.116319",{"doi":490,"openalex_id":492,"authors":493,"venue":448,"cited_by_count":148,"oa_url":445,"card":506,"direction":384,"ingested_from":156},"W7213544281",[494,497,500,503],{"name":495,"orcid":496},"Shahid Karim","https:\u002F\u002Forcid.org\u002F0000-0001-9986-5052",{"name":498,"orcid":499},"Akeel Qadir","https:\u002F\u002Forcid.org\u002F0000-0003-0358-6505",{"name":501,"orcid":502},"Asif Ali Laghari","https:\u002F\u002Forcid.org\u002F0000-0001-5831-5943",{"name":504,"orcid":505},"Irfana Bibi","https:\u002F\u002Forcid.org\u002F0000-0003-2794-504X",{"tldr":507,"method":508,"finding":509,"direction":384,"opportunity":510},"综述深度学习多源遥感图像融合方法、挑战与未来方向。","综述CNN、GAN、Transformer、扩散模型等融合机制与重建行为。","CNN擅局部细节，GAN易谱失真，Transformer长程关系强，扩散模型精度高但算力大。","面向农业的轻量、自监督、不确定性感知融合与基准数据集构建。","2026-09-19T23:30:32.767384Z"]