[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-2802":3,"related-2802":100},{"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":71,"slug":74,"view_count":75,"doi":76,"paper":77,"created_at":99},2802,"Advanced olive leaf area prediction using machine learning methods","https:\u002F\u002Fdoi.org\u002F10.1038\u002Fs41598-026-71390-9","Abstract Accurate leaf area estimation is essential for understanding olive tree physiology, productivity, and stress adaptationThis study developed and evaluated machine learning models for non-destructive olive leaf area prediction using linear measurements (length and width) from 30 diverse cultivars at the Tarom Olive Research Station, Iran. Six machine learning algorithms, Artificial Neural Network (ANN), Support Vector Regression (SVR), Random Forest, Decision Tree, AdaBoost, and XGBoost were optimized using Bayesian optimization, Genetic Algorithm (GA), and Particle Swarm Optimization (PSO). Results demonstrated that PSO consistently outperformed other optimization methods across most models. The ANN model optimized with PSO achieved the highest prediction accuracy (R 2 = 0.9828, RMSE = 0.3009 cm 2 ). External validation using eight additional cultivars confirmed model generalizability, with the universal ANN model maintaining R 2 > 0.98. This study provides a robust, non-destructive methodology for olive leaf area estimation applicable across diverse cultivars, offering practical implications for precision agriculture, phenotyping, and orchard management under changing climatic conditions.","摘要 准确的叶面积估算对于理解油橄榄树的生理特性、生产力及逆境适应性至关重要。本研究在伊朗塔罗姆油橄榄研究站，利用来自30个不同品种的线性测量数据（长度和宽度），开发并评估了用于无损油橄榄叶面积预测的机器学习模型。采用贝叶斯优化、遗传算法（GA）和粒子群优化（PSO）对六种机器学习算法——人工神经网络（ANN）、支持向量回归（SVR）、随机森林、决策树、AdaBoost和XGBoost——进行了优化。结果表明，在大多数模型中，PSO始终优于其他优化方法。经PSO优化后的ANN模型取得了最高的预测精度（R² = 0.9828，RMSE = 0.3009 cm²）。利用另外八个品种进行的外部验证证实了模型的泛化能力，通用ANN模型保持R² > 0.98。本研究为适用于不同品种的油橄榄叶面积估算提供了一种稳健的无损方法，为气候变化条件下的精准农业、表型分析和果园管理提供了实际应用价值。",null,"Scientific Reports","2026-09-17T00:00:00Z","论文",10,false,73,{"impact":17,"substance":18,"depth":19,"authority":20,"freshness":13,"relevant":21,"comment":22},12,20,17,14,1,"基于30个品种的机器学习叶片面积无损预测研究，方法新颖、验证充分，对精准农业与表型分析有实用价值，但属细分领域技术进展，影响范围有限。",[24],{"name":10,"url":6},[26,27,28,29,30],"智慧农业","农业人工智能","机器学习","精准农业","表型分析",{"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},[44,51,58,65],{"code":32,"score":45,"supported":46,"title":47,"abstract":49,"tags":50},16,true,[48],"machine learning",[28,48],[27,28],{"code":52,"score":53,"supported":46,"title":54,"abstract":55,"tags":57},"I1",8,[],[29,56],"precision agriculture",[26,29],{"code":59,"score":60,"supported":46,"title":61,"abstract":62,"tags":64},"B2",3,[],[63],"orchard",[],{"code":66,"score":60,"supported":46,"title":67,"abstract":68,"tags":70},"H7",[],[69],"气候变化",[],[72,73],"农业人工智能 智慧农业 机器学习 精准农业","农业人工智能 智慧农业","农业人工智能智慧农业机器学习精准农业-2802",4,"10.1038\u002Fs41598-026-71390-9",{"doi":76,"openalex_id":78,"authors":79,"venue":10,"cited_by_count":90,"oa_url":6,"card":91,"direction":97,"ingested_from":98},"W7213449017",[80,83,85,87],{"name":81,"orcid":82},"Ahmad Reza Dadras","https:\u002F\u002Forcid.org\u002F0000-0001-8591-5813",{"name":84,"orcid":9},"Hossein Sabouri",{"name":86,"orcid":9},"Ali Tanhaei",{"name":88,"orcid":89},"Sayed Javad Sajadi","https:\u002F\u002Forcid.org\u002F0000-0002-6555-080X",0,{"tldr":92,"method":93,"finding":94,"direction":95,"opportunity":96},"用机器学习基于叶长宽非破坏性预测30个橄榄品种叶面积，PSO优化ANN精度最高。","30个品种叶长宽数据，六种ML算法结合贝叶斯、GA、PSO优化。","PSO优化ANN最优（R²=0.9828），外部8品种验证R²>0.98，通用性好。","农业遥感与作物表型","可拓展至多物种、多环境及无人机\u002F手机图像自动测量，构建通用叶面积表型平台。","农业人工智能与决策模型","openalex","2026-09-17T23:30:59.169744Z",{"total":101,"page":21,"page_size":101,"items":102},6,[103,185,250,307,368,404],{"id":104,"title":105,"url":106,"summary":107,"summary_zh":108,"content":9,"source_name":109,"source_url":106,"published_at":110,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":111,"score_detail":112,"sources":115,"tags":117,"primary_topic":119,"secondary_topics":120,"classification":121,"search_phrases":142,"slug":145,"view_count":90,"doi":146,"paper":147,"created_at":184},5306,"A growing challenge: stakeholder perceptions of AI and ML applications in U.S. agriculture","https:\u002F\u002Fdoi.org\u002F10.1007\u002Fs11119-026-10459-x","Abstract Context The global population is expected to exceed ten billion by 2050 and food demand is projected to increase by up to 100% in the same timeframe. Agricultural management must adapt to meet the increasing global demands. Artificial intelligence (AI) and machine learning (ML) technologies can help the agricultural community respond to these needs. Aims An online survey and follow-up interviews were conducted to examine patterns of use, perceptions, and barriers among early adopting stakeholders regarding AI and ML. Methods The online survey with agricultural stakeholders (n=129) included questions regarding the participants’ views on incorporating AI and ML techniques into farming practices and individual follow-up interviews with agricultural stakeholders (n=10) were conducted to better understand their concerns toward technology use and adoption of AI-enabled tools and technologies. Key results Using a mixed-methods approach, four challenges to establishing AI-\u002FML-enabled technology in agriculture were identified. Namely, challenges related to (1) systems of systems (SoS), (2) sub-seasonal to seasonal weather (S2S) forecasts, (3) precision soil management, and (4) policy and economics. Each challenge was discussed in the context of recent literature. Conclusion Concerns were voiced clustered around four broad research themes. This brief communication brings the perspective of agricultural stakeholders to the discussion of the potential of AI\u002FML-enabled technology and the direction of their development in the agricultural sector. Implications Data collected in 2021, establish an important baseline for the early phase of AI\u002FML use in U.S. agriculture, enabling comparison of future perceptions and developments in the sector. Impact This brief communication describes a survey and follow-up interviews conducted with agricultural stakeholders on factors influencing their willingness and reluctance to incorporate AI and ML techniques into farming practices. The findings, summarized and discussed as four key challenges for AI and ML in the agricultural sector, have the potential to streamline research activities.","摘要\n\n背景 预计到2050年全球人口将超过100亿，同期粮食需求预计将增加多达100%。农业管理必须作出调整以满足不断增长的全球需求。人工智能（AI）和机器学习（ML）技术可帮助农业界应对这些需求。\n\n目的 通过开展在线调查和后续访谈，考察早期采用相关技术的利益相关者在AI和ML方面的使用模式、认知及障碍。\n\n方法 针对农业利益相关者（n=129）的在线调查包含有关参与者对将AI和ML技术纳入农业实践看法的问题，并对农业利益相关者（n=10）进行个体后续访谈，以更好地了解他们对技术使用及采用AI赋能工具和技术的顾虑。\n\n主要结果 采用混合方法，识别出在农业中建立AI\u002FML赋能技术面临的四项挑战，即与以下方面相关的挑战：（1）系统之系统（SoS），（2）次季节至季节天气（S2S）预报，（3）精准土壤管理，以及（4）政策与经济。每项挑战均结合近期文献进行了讨论。\n\n结论 所表达的顾虑集中于四个广泛的研究主题。本简报将农业利益相关者的视角引入关于AI\u002FML赋能技术潜力及其在农业部门发展方向讨论中。\n\n意义 2021年收集的数据为美国农业中AI\u002FML应用的早期阶段建立了重要基线，有助于比较该部门未来的认知和发展。\n\n影响 本简报描述了针对农业利益相关者开展的一项调查和后续访谈，内容涉及影响其愿意和不愿将AI和ML技术纳入农业实践的因素。研究结果总结并讨论为农业部门AI和ML的四项关键挑战，有望优化研究活动。","Precision Agriculture","2026-10-18T00:00:00Z",77,{"impact":113,"substance":18,"depth":19,"authority":20,"freshness":53,"relevant":21,"comment":114},18,"基于129份问卷与10次访谈的混合方法研究，系统梳理美国农业AI\u002FML应用的四大挑战，为早期采用者认知提供基线数据，对智慧农业研究有参考价值。",[116],{"name":109,"url":106},[26,27,28,29,118],"农户认知",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[],{"status":39,"locked":14,"version":40,"method":41,"evidence":122},{"candidates":123,"agricultural_evidence":46},[124,130,137],{"code":32,"score":125,"supported":46,"title":126,"abstract":127,"tags":129},11,[],[28,48,128],"artificial intelligence",[27,28],{"code":131,"score":101,"supported":46,"title":132,"abstract":133,"tags":136},"H2",[],[134,135],"土壤","soil",[],{"code":52,"score":138,"supported":14,"title":139,"abstract":140,"tags":141},2,[],[],[26,29],[143,144],"美国农业 AI ML 调查","农业人工智能 应用障碍","美国农业AIML调查-5306","10.1007\u002Fs11119-026-10459-x",{"doi":146,"openalex_id":148,"authors":149,"venue":109,"cited_by_count":90,"oa_url":106,"card":179,"direction":97,"ingested_from":98},"W7222866367",[150,153,156,159,162,165,168,171,173,176],{"name":151,"orcid":152},"Jennifer Koch","https:\u002F\u002Forcid.org\u002F0000-0002-7067-2705",{"name":154,"orcid":155},"Andrew Fox","https:\u002F\u002Forcid.org\u002F0000-0001-6719-8902",{"name":157,"orcid":158},"Richa Bhattarai","https:\u002F\u002Forcid.org\u002F0000-0003-3827-7761",{"name":160,"orcid":161},"Hank C. Jenkins‐Smith","https:\u002F\u002Forcid.org\u002F0000-0002-5204-2726",{"name":163,"orcid":164},"Jonathan McFadden","https:\u002F\u002Forcid.org\u002F0000-0003-3762-1428",{"name":166,"orcid":167},"Audrey Reinert","https:\u002F\u002Forcid.org\u002F0000-0001-8951-702X",{"name":169,"orcid":170},"Juan Sesmero","https:\u002F\u002Forcid.org\u002F0000-0002-2669-4536",{"name":172,"orcid":9},"M. Z. Walkup",{"name":174,"orcid":175},"Wolfgang Jentner","https:\u002F\u002Forcid.org\u002F0000-0003-1045-6020",{"name":177,"orcid":178},"David S. Ebert","https:\u002F\u002Forcid.org\u002F0000-0001-6177-1296",{"tldr":180,"method":181,"finding":182,"direction":97,"opportunity":183},"通过问卷与访谈调查美国农业利益相关者对AI\u002FML应用的认知与障碍。","在线问卷129人加10人深度访谈的混合方法。","识别出系统之系统、次季节天气预报、精准土壤管理和政策经济四大挑战。","可针对四类障碍开展分区域、分作物类型的采纳意愿追踪与干预研究。","2026-10-19T23:30:02.345893Z",{"id":186,"title":187,"url":188,"summary":189,"summary_zh":190,"content":9,"source_name":191,"source_url":188,"published_at":192,"category":12,"cover_url":9,"hotness":193,"is_selected":14,"score":194,"score_detail":195,"sources":200,"tags":204,"primary_topic":206,"secondary_topics":207,"classification":208,"search_phrases":233,"slug":236,"view_count":138,"doi":237,"paper":238,"created_at":249},3271,"A Systematic Study of Supervised and Ensemble Learning Approaches for Crop Selection in Smart Agriculture","https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.22896343","Selecting the crop best matched to a field’s soil and climate is one of the highest-leverage decisions in agriculture, and one that farmers have traditionally made by intuition or inherited practice. Soil pH together with nitrogen, phosphorus and potassium concentration, and the local temperature, humidity and rainfall regime, jointly determine which crop will flourish and which will fail, and the relationship between those variables and crop performance is neither linear nor independent. Machine learning has therefore become the dominant approach to automated crop recommendation. This paper reviews the field across four technique families — classical supervised learning, ensemble and boosting methods, deep learning and metaheuristic hybrids, and IoT and deployment-oriented systems — and compares twenty-four representative studies published between 2016 and 2026 in terms of method, data source, reported accuracy, advantage and limitation. A generic seven-stage recommendation pipeline is presented and each family is situated within it. The comparison shows that reported accuracy on the standard nutrient-and-climate benchmark has converged in a narrow band between roughly 98 and 99.5 per cent, that boosting and ensemble methods occupy the upper part of that band, and that further gains on the benchmark are no longer the binding constraint on the field. The gaps that remain open are instead the absence of socio-economic and market variables from the decision, the lack of region-specific and long-horizon environmental validation, dataset narrowness and geographic bias, limited interpretability, and the accessibility of these systems to small and resource-poor farmers. These are consolidated into a set of research directions for future work.","选择与田块土壤和气候最匹配的作物是农业中杠杆效应最高的决策之一，而农民传统上依靠直觉或世代相传的经验来做出这一决策。土壤pH值以及氮、磷、钾浓度，加上当地的气温、湿度和降雨状况，共同决定了哪种作物能够茁壮成长、哪种会歉收，而这些变量与作物表现之间的关系既非线性也非相互独立。因此，机器学习已成为自动化作物推荐的主流方法。本文从四个技术族系——经典监督学习、集成与提升方法、深度学习与元启发式混合方法，以及物联网与面向部署的系统——对该领域进行了综述，并从方法、数据来源、报告精度、优势和局限性方面比较了2016年至2026年间发表的二十四项代表性研究。本文提出了一个通用的七阶段推荐流程，并将每个技术族系置于该流程中加以定位。比较结果表明，在标准养分与气候基准上的报告精度已收敛于约98%至99.5%的狭窄区间内，提升与集成方法占据该区间的上端，而在该基准上进一步提升已不再是该领域的约束瓶颈。真正尚未填补的空白在于：决策中缺乏社会经济和市场变量，缺少针对特定区域和长期环境验证，数据集狭窄且存在地理偏差，可解释性有限，以及这些系统对小型和资源匮乏农户的可及性不足。这些空白被归纳为未来工作的一系列研究方向。","Zenodo (CERN European Organization for Nuclear Research)","2026-09-22T00:00:00Z",25,78,{"impact":45,"substance":196,"depth":113,"authority":197,"freshness":198,"relevant":21,"comment":199},22,13,9,"系统综述24项研究并指出基准精度已趋饱和，真正瓶颈转向社会经济变量与可解释性，对智慧农业选种方向有参考价值。",[201,202],{"name":191,"url":188},{"name":191,"url":203},"https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.22896344",[26,27,28,29,205],"作物推荐",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[],{"status":39,"locked":14,"version":40,"method":41,"evidence":209},{"candidates":210,"agricultural_evidence":46},[211,217,222,226],{"code":32,"score":125,"supported":46,"title":212,"abstract":213,"tags":216},[],[214,28,48,215],"深度学习","deep learning",[27,28],{"code":52,"score":13,"supported":46,"title":218,"abstract":220,"tags":221},[219],"smart agriculture",[],[26,29],{"code":131,"score":101,"supported":46,"title":223,"abstract":224,"tags":225},[],[134,135],[],{"code":227,"score":101,"supported":46,"title":228,"abstract":229,"tags":232},"I5",[],[230,231],"物联网","IoT",[],[234,235],"农业人工智能 作物推荐 智慧农业 机器学习","农业人工智能 作物推荐","农业人工智能作物推荐智慧农业机器学习-3271","10.5281\u002Fzenodo.22896343",{"doi":237,"openalex_id":239,"authors":240,"venue":191,"cited_by_count":90,"oa_url":188,"card":243,"direction":248,"ingested_from":98},"W7214002748",[241],{"name":242,"orcid":9},"Prof. Nagendra Patel Sahil Verma",{"tldr":244,"method":245,"finding":246,"direction":97,"opportunity":247},"系统综述2016-2026年24项作物推荐研究，比较四类机器学习方法并指出基准精度已趋饱和。","综述监督学习、集成提升、深度学习与元启发式、物联网部署四类方法及七阶段流程。","标准基准精度收敛于98%-99.5%，提升集成法最优，但精度已非领域瓶颈。","将社会经济与市场变量、区域长期环境验证及可解释性纳入作物推荐，服务小农户。","智慧农业 \u002F 农业物联网","2026-09-23T23:30:09.369987Z",{"id":251,"title":252,"url":253,"summary":254,"summary_zh":255,"content":9,"source_name":10,"source_url":253,"published_at":256,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":257,"score_detail":258,"sources":261,"tags":263,"primary_topic":265,"secondary_topics":266,"classification":267,"search_phrases":289,"slug":292,"view_count":21,"doi":293,"paper":294,"created_at":306},2338,"An optimized machine learning approach for reliable agronomic parameter prediction in precision farming systems","https:\u002F\u002Fdoi.org\u002F10.1038\u002Fs41598-026-71592-1","Optimized machine learning models are very important for improving predictive performance in precision agriculture because they let us analyze soil and environmental data in a data-driven way. But conventional predictive methods often can’t be used to make generalizations because agronomic data is often very variable, has nonlinear interactions, and is very different from one another. The study presents an enhanced machine learning-based predictive framework for estimating agricultural parameters utilizing structured numerical soil and environmental datasets. The framework combines systematic data preprocessing, feature selection, and hyperparameter optimization to make models more stable and reliable. The model performance was evaluated using five-fold cross-validation and standard regression metrics, including the Coefficient of Determination (R 2 ), Root Mean Square Error (RMSE), and Mean Absolute Error (MAE). The average R 2 is 0.696 ± 0.148, the RMSE is 5.160 ± 1.571, and the MAE is 4.636 ± 1.545, which shows that the model can make accurate predictions across all validation folds. Further the classification of the growth stages of spinach is done with an accuracy of 84.63% and with the precision of 85% using the proposed Hybrid ensemble model. The results show that the proposed framework works well with nonlinear agricultural data and could be used for data-driven decisions in precision farming systems.","优化的机器学习模型对于提升精准农业中的预测性能至关重要，因为它使我们能够以数据驱动的方式分析土壤和环境数据。然而，传统的预测方法往往无法用于泛化，因为农艺数据通常变异性很大、具有非线性交互作用，且彼此之间差异显著。本研究提出了一种基于增强机器学习的预测框架，利用结构化数值土壤和环境数据集来估算农业参数。该框架结合了系统化的数据预处理、特征选择和超参数优化，使模型更加稳定可靠。采用五折交叉验证和标准回归指标评估模型性能，包括决定系数（R²）、均方根误差（RMSE）和平均绝对误差（MAE）。平均R²为0.696 ± 0.148，RMSE为5.160 ± 1.571，MAE为4.636 ± 1.545，表明该模型在所有验证折中均能做出准确预测。此外，利用所提出的混合集成模型对菠菜生长阶段进行分类，准确率达到84.63%，精确率达到85%。结果表明，所提出的框架能够很好地处理非线性农业数据，可用于精准农业系统中的数据驱动决策。","2026-09-12T00:00:00Z",67,{"impact":17,"substance":113,"depth":259,"authority":20,"freshness":53,"relevant":21,"comment":260},15,"该论文提出融合预处理、特征选择与超参数优化的机器学习框架，在土壤环境数据上取得R²约0.70的预测表现并以84.63%准确率识别菠菜生长期，方法扎实但属常规模型优化，产业影响有限，可作为智慧农业技术参考。",[262],{"name":10,"url":253},[26,27,28,29,264],"土壤数据",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[],{"status":39,"locked":14,"version":40,"method":41,"evidence":268},{"candidates":269,"agricultural_evidence":46},[270,274,278,283],{"code":32,"score":45,"supported":46,"title":271,"abstract":272,"tags":273},[48],[28,48],[27,28],{"code":52,"score":53,"supported":46,"title":275,"abstract":276,"tags":277},[],[29,56],[26,29],{"code":131,"score":279,"supported":46,"title":280,"abstract":281,"tags":282},7,[],[134,135],[264],{"code":284,"score":60,"supported":46,"title":285,"abstract":286,"tags":288},"I6",[],[287],"agricultural data",[],[290,291],"农业人工智能 土壤数据 智慧农业 机器学习","农业人工智能 土壤数据","农业人工智能土壤数据智慧农业机器学习-2338","10.1038\u002Fs41598-026-71592-1",{"doi":293,"openalex_id":295,"authors":296,"venue":10,"cited_by_count":90,"oa_url":253,"card":301,"direction":97,"ingested_from":98},"W7212396250",[297,299],{"name":298,"orcid":9},"T. Suba",{"name":300,"orcid":9},"K. Lakshmi Joshitha",{"tldr":302,"method":303,"finding":304,"direction":97,"opportunity":305},"提出优化机器学习框架，用土壤环境数据预测农艺参数并分类菠菜生长阶段。","数据预处理、特征选择、超参数优化，五折交叉验证与混合集成模型。","回归平均R²为0.696，菠菜生长阶段分类准确率达84.63%。","可探索跨作物、跨区域迁移学习与可解释性，提升非线性农艺数据泛化能力。","2026-09-13T23:30:43.564799Z",{"id":308,"title":309,"url":310,"summary":311,"summary_zh":312,"content":9,"source_name":313,"source_url":310,"published_at":314,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":315,"score_detail":316,"sources":318,"tags":320,"primary_topic":322,"secondary_topics":323,"classification":324,"search_phrases":348,"slug":350,"view_count":60,"doi":351,"paper":352,"created_at":367},2179,"Comparative evaluation of classical machine learning and deep learning models for early weed detection in precision agriculture","https:\u002F\u002Fdoi.org\u002F10.1007\u002Fs44163-026-02195-y","Unless treated and controlled early, weeds considerably impede the growth of crops as they compete over vital factors (resources) like nutrients, water, and light. To overcome this problem, an automated weed species early detection and classification system using machine learning-based image analysis is suggested. The framework integrates classical machine learning algorithms Support Vector Machines (SVM), Random Forests and k-Nearest Neighbors (k-NN), using handcrafted features like texture, shape, and color with deep learning models, Convolutional Neural Networks (CNNs) which automatically learn discriminative features in the data. The study comparatively evaluates classical ML and deep learning models. In order to test the experiments, a publicly available Sugar Beet dataset was used, as well as a custom maize seedling dataset. To enhance generalization and robustness, some preprocessing steps were performed, including normalization, background removal, and augmentation. Common metrics used to assess the performance of the models are F1-score and Intersection over Union (IoU). The experimental results show deep learning models outperforming the traditional machine learning methods, especially CNNs The CNN model also exhibited a classification accuracy of 98.7%. These results demonstrate the promise of deep learning to quickly, reliably, and in large scale detect weeds in the early stages of precision agriculture. This study highlights the feasibility of machine learning systems in proactive management of weed as well as in supporting healthy crop growth at the early stages of development.","除非在早期进行处理和控制，否则杂草会争夺养分、水分和光照等关键因素（资源），从而严重阻碍作物生长。为解决这一问题，提出了一种基于机器学习的图像分析自动化杂草物种早期检测与分类系统。该框架整合了经典机器学习算法——支持向量机（SVM）、随机森林和k近邻（k-NN），利用纹理、形状和颜色等手工特征，并结合深度学习模型——卷积神经网络（CNN），后者可自动学习数据中的判别性特征。本研究对经典机器学习和深度学习模型进行了对比评估。为验证实验，使用了公开的甜菜数据集以及自建的玉米幼苗数据集。为增强泛化性和鲁棒性，执行了一些预处理步骤，包括归一化、背景去除和数据增强。用于评估模型性能的常用指标为F1分数和交并比（IoU）。实验结果表明，深度学习模型优于传统机器学习方法，尤其是CNN。CNN模型的分类准确率达到了98.7%。这些结果证明了深度学习在精准农业中快速、可靠且大规模地早期检测杂草的潜力。本研究凸显了机器学习系统在杂草主动管理以及支持作物早期健康生长方面的可行性。","Discover Artificial Intelligence","2026-09-10T00:00:00Z",72,{"impact":45,"substance":18,"depth":45,"authority":17,"freshness":53,"relevant":21,"comment":317},"对比经典机器学习与深度学习在作物早期杂草检测中的表现，CNN 分类准确率达 98.7%，方法扎实、结论明确，对精准农业智能除草有参考价值。",[319],{"name":313,"url":310},[26,27,28,321,29],"杂草识别",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[],{"status":39,"locked":14,"version":40,"method":41,"evidence":325},{"candidates":326,"agricultural_evidence":46},[327,332,336,343],{"code":32,"score":328,"supported":46,"title":329,"abstract":330,"tags":331},27,[48,215],[214,28,48,215],[27,28],{"code":52,"score":45,"supported":46,"title":333,"abstract":334,"tags":335},[56],[29,56],[26,29],{"code":337,"score":101,"supported":46,"title":338,"abstract":339,"tags":342},"A3",[],[340,341],"玉米","maize",[],{"code":344,"score":21,"supported":14,"title":345,"abstract":346,"tags":347},"F4",[],[],[321],[349,73],"农业人工智能 智慧农业 机器学习 杂草识别","农业人工智能智慧农业机器学习杂草识别-2179","10.1007\u002Fs44163-026-02195-y",{"doi":351,"openalex_id":353,"authors":354,"venue":313,"cited_by_count":90,"oa_url":310,"card":362,"direction":97,"ingested_from":98},"W7212111551",[355,358,360],{"name":356,"orcid":357},"Rajeev Kumar","https:\u002F\u002Forcid.org\u002F0000-0001-8414-3778",{"name":359,"orcid":9},"P. K. Singh",{"name":361,"orcid":9},"Rohit Kumar Tiwari",{"tldr":363,"method":364,"finding":365,"direction":97,"opportunity":366},"对比经典机器学习与深度学习模型在精准农业早期杂草检测中的性能。","用SVM、随机森林、k-NN及CNN，基于甜菜和玉米幼苗图像数据集。","CNN分类准确率达98.7%，显著优于传统机器学习方法。","可探索轻量化模型在田间边缘设备实时检测杂草的部署与跨物种泛化能力。","2026-09-11T23:30:52.642134Z",{"id":369,"title":370,"url":371,"summary":372,"summary_zh":9,"content":373,"source_name":374,"source_url":9,"published_at":375,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":15,"score_detail":376,"sources":378,"tags":380,"primary_topic":382,"secondary_topics":383,"classification":384,"search_phrases":401,"slug":402,"view_count":60,"doi":9,"paper":9,"created_at":403},1195,"A Taxonomy of Machine Learning for UAV-Enabled Precision Agriculture: A Structured Survey (面向无人机赋能精准农业的机器学习方法分类:结构化综述)","https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249","MDPI AgriEngineering 2026, 8(6), 249。精准农业越来越依赖应用于无人机(UAV)采集的高分辨率数据的机器学习来支持作物监测、胁迫检测和产量预测。本综述提出了无人机赋能精准农业机器学习方法的结构化综述,在一个由感知模态、数据类型、模型家族和分析任务四个维度定义的统一分类法中组织100多项同行评议研究。该分类法能够在RGB、多光谱、高光谱、LiDAR和IoT数据源之间进行系统比较,并在经典机器学习、深度学习、混合顺序模型和新兴基于Transformer的架构之间进行系统比较。分析建模选择如何与数据特征相互作用以影响鲁棒性、跨环境泛化、计算效率和UAV及边缘平台上的部署可行性。识别新兴研究方向,包括数据高效学习、表示级多模态融合、域自适应、轻量级架构和不确定性感知决策支持。","Logical Operator Operator\n\nSearch Text\n\nSearch Type\n\n_add\\_circle\\_outline_\n\n_remove\\_circle\\_outline_\n\n[![Image 1: agriengineering-logo](https:\u002F\u002Fpub.mdpi-res.com\u002Fimg\u002Fjournals\u002Fagriengineering-logo.png?9e66d6162a2bcb52)](https:\u002F\u002Fwww.mdpi.com\u002Fjournal\u002Fagriengineering)\n\n## Article Menu\n\nFont Type:\n\n_Arial_ _Georgia_ _Verdana_\n\nFont Size:\n\nAa Aa Aa\n\nLine Spacing:\n\n__ __ __\n\nColumn Width:\n\n__ __ __\n\nBackground:\n\nOpen Access Review\n\nby \nWan D. Bae\n\n 1,*[](mailto:baew@seattleu.edu)[![Image 2: ORCID](https:\u002F\u002Fpub.mdpi-res.com\u002Fimg\u002Fdesign\u002Forcid.png?0465bc3812adeb52?1788181901)](https:\u002F\u002Forcid.org\u002F0000-0002-4611-5546), \nShayma Alkobaisi\n\n 2[](mailto:shayma.alkobaisi@uaeu.ac.ae)[![Image 3: ORCID](https:\u002F\u002Fpub.mdpi-res.com\u002Fimg\u002Fdesign\u002Forcid.png?0465bc3812adeb52?1788181901)](https:\u002F\u002Forcid.org\u002F0000-0003-4237-7976), \nMuhammad Farhan Safdar\n\n 3[](mailto:muhammad.safdar@pw.edu.pl)[![Image 4: ORCID](https:\u002F\u002Fpub.mdpi-res.com\u002Fimg\u002Fdesign\u002Forcid.png?0465bc3812adeb52?1788181901)](https:\u002F\u002Forcid.org\u002F0000-0001-7818-4805) and \nPrachitee Chouhan\n\n 1[](mailto:pchouhan@seattleu.edu)\n\n1\n\nDepartment of Computer Science, Seattle University, Seattle, WA 98122, USA\n\n2\n\nCollege of Information Technology, United Arab Emirates University, Sheik Khalifa Bin Zayed Street, Al Ain P.O. Box 15551, Abu Dhabi, United Arab Emirates\n\n3\n\nFaculty of Electronics and Information Technology, Warsaw University of Technology, Pl. Politechniki 1, 00-661 Warsaw, Poland\n\n*\n\nAuthor to whom correspondence should be addressed.\n\nSubmission received: 21 February 2026 \u002F Revised: 4 June 2026 \u002F Accepted: 5 June 2026 \u002F Published: 18 June 2026\n\n## Abstract\n\nPrecision agriculture increasingly relies on machine learning applied to high-resolution data acquired by unmanned aerial vehicles (UAVs) to support crop monitoring, stress detection, and yield forecasting. This survey presents a structured review of machine learning methods for UAV-enabled precision agriculture and organizes over 100 peer-reviewed studies within a unified four-dimensional taxonomy defined by sensing modality, data type, model family, and analytical task. The taxonomy enables systematic comparison across RGB, multispectral, hyperspectral, LiDAR, and IoT data sources and across classical machine learning, deep learning, hybrid sequential models, and emerging transformer-based architectures. We analyze how modeling choices interact with data characteristics to influence robustness, cross-environment generalization, computational efficiency, and deployment feasibility on UAV and edge platforms. Recurring challenges include limited labeled data, domain shift across seasons and fields, multimodal heterogeneity, occlusion, and real-time processing constraints. We identify emerging research directions, including data-efficient learning, representation-level multimodal fusion, domain adaptation, lightweight architectures for embedded deployment, and uncertainty aware decision support. By formalizing the landscape through a unified taxonomy, this survey provides a foundation for designing scalable, robust, and deployable machine learning systems for next-generation precision agriculture.\n\n[![Image 5](https:\u002F\u002Fpub.mdpi-res.com\u002Fagriengineering\u002Fagriengineering-08-00249\u002Farticle_deploy\u002Fhtml\u002Fimages\u002Fagriengineering-08-00249-ag-550.jpg?1782112396)](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#)\n\nGraphical Abstract\n\n## 1. Introduction\n\nPrecision agriculture (PA) increasingly depends on unmanned aerial vehicles (UAVs) and data-driven modeling systems to enable high-resolution crop monitoring, stress detection, and yield forecasting. Advances in sensing platforms, including RGB, multispectral, hyperspectral, thermal, and LiDAR (Light Detection and Ranging) sensors, combined with machine learning and deep learning techniques, have transformed how agricultural data are collected, processed, and interpreted. These technologies support scalable data-driven decision making leading to improved productivity, resource efficiency, and environmental sustainability.\n\nBeyond serving as an application domain, PA presents a set of characteristics that actively shape the development of artificial intelligence (AI) methods. Agricultural environments expose AI systems to strong non-stationarity across seasons, cultivars, and management practices; extreme intra class variability caused by growth stages and phenology; and severe data imbalance between healthy and stressed conditions. These properties challenge assumptions commonly made in benchmark-driven computer vision and time-series modeling, where data distributions are often static and labels are abundant.\n\nExisting AI approaches in PA are often limited in scope, focusing on specific sensing modalities, individual crops, or isolated analytical tasks such as segmentation, pest and disease detection, fruit and bloom count, and yield prediction. Many surveys emphasize either model architectures or sensing technologies but do not systematically connect data characteristics, model design, and downstream agricultural applications. As a result, it remains difficult to compare approaches across studies or to understand how methodological choices interact with real-world deployment constraints in UAV-enabled precision agriculture. In view of these limitations, the following research questions were formulated:\n\n*   How do different AI model families (e.g., convolutional neural networks (CNNs), transformers, recurrent neural networks (RNNs), and classical models) perform across UAV-based sensing modalities and agricultural tasks?\n\n*   What factors limit the reliability of AI-based predictions under varying real-world agricultural conditions?\n\n*   How can UAV-based AI systems be translated into actionable insights for crop monitoring and interventional management?\n\nUnlike previous surveys that primarily organize the literature around individual sensing platforms, crop types, or model architectures, this review introduces a unified four-dimensional taxonomy that jointly connects sensing modality, data type, model family, and analytical task within a single analytical framework. This structure enables systematic comparison across studies that would otherwise remain disconnected, including segmentation, detection, counting, and yield prediction pipelines operating under different sensing and deployment conditions. The proposed taxonomy also provides practical guidance for designing scalable and operationally feasible UAV-enabled agricultural analytics systems.\n\nThe key contributions of this survey are as follows:\n\n*   We propose a unified taxonomy for AI-driven precision agriculture that integrates four complementary dimensions: sensing modality, data type, model family, and analytical task, enabling systematic cross-study comparison.\n\n*   We identify structural challenges including data scarcity, annotation cost, domain shift, limited cross-modal integration, and deployment constraints that limit robustness and scalability.\n\n*   We outline emerging research directions, including data-efficient learning, domain adaptation, representation-level multimodal fusion, synthetic data generation, and lightweight architectures for real-time deployment.\n\nThe literature reviewed in this survey was collected primarily from IEEE Xplore, ACM Digital Library, Scopus, Web of Science, ScienceDirect, and MDPI databases using combinations of keywords related to UAVs, precision agriculture, machine learning, deep learning, crop monitoring, disease detection, segmentation, fruit counting, and yield prediction. The survey focuses primarily on peer-reviewed journal and conference papers published between 2015 and 2025, with emphasis on studies involving UAV-enabled sensing and AI-driven analytical pipelines. Papers were selected based on their relevance to at least one of the four taxonomy dimensions introduced in this survey: sensing modality, data type, model family, or analytical task. More than 100 studies were analyzed across segmentation, pest and disease detection, bloom and fruit counting, and yield prediction tasks, spanning RGB, multispectral, hyperspectral, thermal, LiDAR, satellite, and IoT-based sensing modalities.\n\nThe remainder of this paper is organized as follows. [Section 2](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#sec2-agriengineering-08-00249) provides background on precision agriculture, discusses key challenges, and introduces our taxonomy. [Section 3](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#sec3-agriengineering-08-00249) reviews sensing modalities, data-acquisition strategies, and data-preparation techniques. [Section 4](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#sec4-agriengineering-08-00249) summarizes segmentation methods. [Section 5](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#sec5-agriengineering-08-00249) reviews pest and disease detection. [Section 6](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#sec6-agriengineering-08-00249) covers bloom detection, fruit counting, and yield prediction. [Section 8](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#sec8-agriengineering-08-00249) synthesizes cross-cutting challenges and emerging research directions, and [Section 9](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#sec9-agriengineering-08-00249) concludes the survey.\n\n## 2. Background, and Taxonomy\n\n### 2.1. Background and Research Gap\n\nThe evolution of information and communication technologies, remote sensing, and internet of things (IoT) has dramatically expanded the range of data available for precision agriculture [[1](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#B1-agriengineering-08-00249),[2](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#B2-agriengineering-08-00249)]. UAVs and ground-based sensors provide high-resolution measurements of crop and soil conditions, while satellite platforms offer broad spatial and temporal coverage. Although these technologies enable data-driven decision making, they also introduce several challenges that directly impact the design and deployment of deep learning models.\n\nData acquisition is central to both deep learning (DL) models and IoT systems in agriculture. Sensors, UAVs, and other devices collect diverse data on pests, soil properties, canopy structure, crop yields, and environmental variables such as temperature, light, and humidity [[2](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#B2-agriengineering-08-00249),[3](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#B3-agriengineering-08-00249)]. Studies on smart agriculture systems [[4](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#B4-agriengineering-08-00249),[5](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#B5-agriengineering-08-00249),[6](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#B6-agriengineering-08-00249)] emphasize the importance of reliable sensing infrastructures and communication networks for real-time monitoring and control.\n\nHigh-quality annotation is equally critical. Accurate labels enable supervised DL models to learn robust representations for tasks such as pest detection, disease diagnosis, and crop health monitoring [[7](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#B7-agriengineering-08-00249),[8](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#B8-agriengineering-08-00249)]. For UAV imagery, the annotation process is often labor intensive and requires domain expertise, particularly when dealing with subtle symptoms or occluded plant parts.\n\nSeveral works highlight the difficulties of vegetation stress detection using UAVs, where models must capture both intra field variability and temporal dynamics [[9](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#B9-agriengineering-08-00249),[10](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#B10-agriengineering-08-00249)]. Large-scale, well-annotated datasets are rarely available, and models trained on small datasets tend to overfit or fail to generalize to new fields. To mitigate these issues, researchers employ data augmentation and regularization strategies such as geometric transformations and dropout layers [[11](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#B11-agriengineering-08-00249)]. The use of high-resolution imagery and appropriate preprocessing steps is particularly important for building reliable prediction models [[12](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#B12-agriengineering-08-00249)].\n\nData quality issues remain a recurring challenge. Agricultural data vary widely across crop types, regions, and seasons, leading to covariate shifts that degrade model performance. Incremental or phased adoption of advanced sensing and annotation technologies has been proposed as a practical strategy to improve data quality over time and support the sustainable deployment of PA systems [[13](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#B13-agriengineering-08-00249)].\n\nIn summary, the following research gap continues to shape the development of AI-based solutions in UAV-enabled precision agriculture:\n\n*   Data variability and generalization: Although UAV sensing enables high-resolution monitoring, significant variability across environments, crop varieties, and management practices limits the generalizability of machine learning (ML) models.\n\n*   Data quality and limited annotated datasets: Many deep learning approaches rely on large, consistently annotated datasets, which remain scarce and costly to obtain, particularly for fine-grained agricultural tasks.\n\n*   Operational deployment and scalability: Practical deployment requires the robust integration of sensing platforms, data infrastructures, and analytical models, which is still evolving in many precision agriculture systems.\n\nThis preliminary information about the research gap motivates the methodological choices discussed in subsequent sections and underscores the need for a structured taxonomy that connects sensing modalities, data types, model families, and analytical tasks.\n\n### 2.2. Taxonomy of UAV-Based Sensing and Machine Learning in Precision Agriculture\n\nThe diversity of sensing platforms, data types, model architectures, and agricultural applications makes it challenging to compare existing approaches in a systematic manner. To provide a unifying perspective, we introduce a taxonomy that organizes UAV-based sensing and machine learning methods along four complementary dimensions: (1) sensing modality, (2) data type, (3) model family, and (4) analytical task. This taxonomy links upstream sensing and data preparation steps in [Section 3](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#sec3-agriengineering-08-00249) with downstream segmentation, detection, counting, and prediction methods discussed in later sections. It also provides the structural foundation for the survey, with each subsequent section mapped explicitly to one or more of these four dimensions.\n\nSeveral recent surveys review deep learning applications in agriculture or remote sensing independently; however, they typically organize the literature along a single dimension, such as model architecture, crop type, or sensing platform. In contrast, the taxonomy proposed in this survey integrates four complementary dimensions including sensing modality, data type, model family, and analytical task into a unified framework. This multidimensional structure enables systematic comparison across studies that would otherwise appear disconnected, such as UAV-based object detection and satellite-driven yield forecasting.\n\nUnlike model centric surveys that emphasize architectural trends, or remote sensing reviews that focus primarily on sensor characteristics, our taxonomy explicitly links upstream sensing infrastructure and data preparation choices to downstream learning models and decision support objectives. This integrated perspective reveals underexplored combinations of sensing modalities and model families and provides a structured foundation for analyzing and designing scalable, end-to-end data-driven systems in precision agriculture.\n\nCompared to previous reviews that focus primarily on algorithms or crop-specific case studies, this survey integrates sensing infrastructure, data structure, modeling approaches, and analytical objectives within a unified engineering framework. By bridging sensing technologies and machine learning methodologies, the proposed taxonomy supports the design and evaluation of deployable, robust, and scalable agricultural systems.\n\n1.\nSensing modality: AI-enabled precision agriculture relies on heterogeneous sensing platforms that vary widely in spatial, spectral, and temporal resolution. UAV-mounted RGB, multispectral, hyperspectral, thermal, and LiDAR sensors provide high spatial detail for canopy structure, plant health assessment, and fine-scale monitoring [[3](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#B3-agriengineering-08-00249),[14](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#B14-agriengineering-08-00249)]. Satellite platforms extend temporal and regional coverage, offering multi-temporal vegetation indices and spectral diagnostics used extensively for crop health and yield modeling [[2](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#B2-agriengineering-08-00249)]. Ground-based IoT sensors, including soil moisture probes, weather stations, and in-field cameras provide micro-environmental and crop-level observations that complement airborne and satellite sensing [[15](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#B15-agriengineering-08-00249)]. Together, these modalities form the multi-resolution sensing backbone of modern agricultural analytics.\n\n2.\nData type: The raw data produced by these sensing modalities differ in structure, dimensionality, and temporal properties. We categorize data types into five groups: (i) single frame 2D imagery, (ii) multi-temporal image sequences used for phenological monitoring and yield prediction [[16](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#B16-agriengineering-08-00249)], (iii) 3D point clouds derived from LiDAR or structure from motion photogrammetry [[14](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#B14-agriengineering-08-00249)], (iv) tabular sensor and environmental streams from IoT systems, and (v) fused multimodal datasets combining imagery, spectral indices, and ground measurements [[17](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#B17-agriengineering-08-00249)]. Each data type imposes different requirements on preprocessing pipelines, feature extraction, and model selection.\n\n3.\nModel family: A wide range of machine learning and deep learning models have been applied to agricultural imagery and sensor data. Classical methods such as support vector machines (SVMs), random forests, and fuzzy rule-based systems remain competitive when datasets are small or interpretability is required. Deep learning models dominate imagery-driven tasks, including convolutional neural networks (CNNs) [[18](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#B18-agriengineering-08-00249)], residual networks (ResNet) [[19](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#B19-agriengineering-08-00249)], encoder–decoder architectures such as U-Net [[20](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#B20-agriengineering-08-00249)], region-based detectors including Faster region-CNN (Faster R-CNN) and Mask Region-CNN (Mask R-CNN) [[21](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#B21-agriengineering-08-00249),[22](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#B22-agriengineering-08-00249)], and one-stage detectors such as you only look once (YOLO) variants [[23](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#B23-agriengineering-08-00249),[24](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#B24-agriengineering-08-00249)]. Sequential models (long short-term memory (LSTM) [[25](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#B25-agriengineering-08-00249)], Convolutional LSTM (ConvLSTM) [[26](https:\u002F\u002Fwww.mdpi.com\u002F2624-7402\u002F8\u002F6\u002F249#B26-agriengineering-08-00249)]) and hybrid architectures (CNN–SVM, CNN–LSTM, and stacking ensembles) further extend these capabilities. Recent advances in attention-based and transformer architectures further expand this model family dimension, offering alternatives to convolutional and recurrent networks for modeling long-range spatial and temporal dependencies in agricultural data.\n\n4.\nAnalytical task: The fi","AgriEngineering 2026, 8(6), 249","2026-08-25T08:00:00Z",{"impact":113,"substance":196,"depth":113,"authority":197,"freshness":138,"relevant":21,"comment":377},"结构化综述系统梳理无人机精准农业机器学习方法，提出四维分类法，对领域研究有重要参考价值。",[379],{"name":374,"url":371},[26,381,27,28,29],"无人机",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[],{"status":39,"locked":14,"version":40,"method":41,"evidence":385},{"candidates":386,"agricultural_evidence":46},[387,392,397],{"code":32,"score":388,"supported":46,"title":389,"abstract":390,"tags":391},24,[28,48],[214,28],[27,28],{"code":52,"score":393,"supported":46,"title":394,"abstract":395,"tags":396},21,[29,56],[29],[26,29],{"code":227,"score":60,"supported":46,"title":398,"abstract":399,"tags":400},[],[231],[],[72,73],"农业人工智能智慧农业机器学习精准农业-1195","2026-09-01T00:03:40.394459Z",{"id":405,"title":406,"url":407,"summary":408,"summary_zh":409,"content":9,"source_name":410,"source_url":407,"published_at":411,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":412,"score_detail":413,"sources":415,"tags":417,"primary_topic":419,"secondary_topics":420,"classification":431,"search_phrases":477,"slug":480,"view_count":90,"doi":481,"paper":482,"created_at":500},5553,"Artificial Intelligence–Driven Precision Agriculture Systems: Engineering Frameworks for Climate-Resilient Food Production","https:\u002F\u002Fdoi.org\u002F10.53982\u002Fajerd.2026.01.14-j","Climate change, soil issues, water scarcity, and the increasing need for food worldwide mean we need smart farming systems that can manage resources flexibly. Old farming methods used the same inputs and simple rules for decisions. AI can help us manage resources better by adapting and optimizing them as we deal with complex environmental factors. This paper looks closely at AI-powered precision agriculture systems designed to make food production more resilient to changing weather. The authors reviewed research from 2021 to 2026 in agricultural engineering, computer engineering, and environmental systems. The authors examined how machine learning, deep learning, reinforcement learning, computer vision, robotics, and IoT sensors are used for predicting crop yields, spotting diseases, improving irrigation, managing nutrients, and automating field tasks. Our analysis shows that hybrid deep learning models (CNN-LSTM) are better than single models for predicting crop yields over space and time. Reinforcement learning algorithms are also better at adjusting irrigation cycles when weather is unpredictable. Using edge computing for system design reduces delays and improves real-time decision-making compared to traditional cloud-based systems. However, there are still big challenges with understanding how AI models work, having enough diverse data, ensuring cybersecurity, setting common standards, and making these systems affordable, especially in developing countries. To tackle these problems, we suggest a framework that combines AI analysis, distributed sensor networks, adaptive controls, and climate risk assessments. The authors provided critical insights into essential research gaps and outline a technological roadmap towards achieving sustainable, resilient and intelligent agricultural systems.","气候变化、土壤问题、水资源短缺以及全球日益增长的粮食需求，意味着我们需要能够灵活管理资源的智能农业系统。传统农业方法对相同的投入采用固定模式，并依赖简单规则进行决策。人工智能可以帮助我们更好地管理资源，通过适应和优化资源利用来应对复杂的环境因素。本文深入探讨了人工智能驱动的精准农业系统，旨在使粮食生产对不断变化的天气更具韧性。作者综述了2021年至2026年间农业工程、计算机工程和环境系统领域的研究。作者考察了机器学习、深度学习、强化学习、计算机视觉、机器人技术及物联网传感器如何被用于作物产量预测、病害识别、灌溉优化、养分管理和田间作业自动化。我们的分析表明，混合深度学习模型（CNN-LSTM）在时空作物产量预测方面优于单一模型。在天气不可预测的情况下，强化学习算法在调整灌溉周期方面也表现更优。与传统基于云的系统相比，采用边缘计算进行系统设计可降低延迟并改善实时决策能力。然而，在理解人工智能模型的工作原理、获取足够多样化的数据、确保网络安全、制定通用标准以及使这些系统具备可负担性方面，仍存在重大挑战，尤其是在发展中国家。为应对这些问题，我们提出了一个融合人工智能分析、分布式传感器网络、自适应控制和气候风险评估的框架。作者对关键研究空白提供了重要见解，并勾勒了实现可持续、韧性和智能农业系统的技术路线图。","ABUAD Journal of Engineering Research and Development (AJERD)","2026-10-21T00:00:00Z",80,{"impact":113,"substance":196,"depth":113,"authority":197,"freshness":198,"relevant":21,"comment":414},"系统综述AI驱动精准农业的工程框架与技术路线，方法梳理扎实、结论有参考价值，但属综述类论文，产业落地影响有限。",[416],{"name":410,"url":407},[26,27,230,29,418],"气候韧性",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[421,425],{"code":227,"slug":422,"name":423,"path":424,"parent_code":36},"agricultural-iot","物联网传感与边缘计算","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-iot",{"code":426,"slug":427,"name":428,"path":429,"parent_code":430},"H3","irrigation","水资源与灌溉","\u002Ftopic\u002Fenvironment\u002Firrigation","H",{"status":39,"locked":14,"version":40,"method":41,"evidence":432},{"candidates":433,"agricultural_evidence":46},[434,440,444,450,456,461,465,471],{"code":32,"score":113,"supported":46,"title":435,"abstract":436,"tags":439},[128],[214,28,437,48,215,438],"计算机视觉","computer vision",[27],{"code":52,"score":45,"supported":46,"title":441,"abstract":442,"tags":443},[56],[29,56],[26,29],{"code":227,"score":13,"supported":46,"title":445,"abstract":446,"tags":449},[],[230,447,231,448],"边缘计算","edge computing",[230],{"code":426,"score":198,"supported":46,"title":451,"abstract":452,"tags":455},[],[453,454,427],"灌溉","水资源",[],{"code":66,"score":279,"supported":46,"title":457,"abstract":458,"tags":460},[],[69,459],"climate change",[418],{"code":131,"score":101,"supported":46,"title":462,"abstract":463,"tags":464},[],[134,135],[],{"code":466,"score":60,"supported":46,"title":467,"abstract":468,"tags":470},"A7",[],[469],"间作",[],{"code":472,"score":60,"supported":46,"title":473,"abstract":474,"tags":476},"F3",[],[475],"养分管理",[],[478,479],"AI 精准农业 气候韧性","CNN-LSTM 作物产量预测","AI精准农业气候韧性-5553","10.53982\u002Fajerd.2026.01.14-j",{"doi":481,"openalex_id":483,"authors":484,"venue":410,"cited_by_count":90,"oa_url":407,"card":495,"direction":248,"ingested_from":98},"W7223121456",[485,488,490,492],{"name":486,"orcid":487},"Olumayowa J. Iyanda","https:\u002F\u002Forcid.org\u002F0000-0002-5340-9884",{"name":489,"orcid":9},"Oluwadare OYEBODE",{"name":491,"orcid":9},"Iretiayo Adelaiye",{"name":493,"orcid":494},"Ayomide Afolabi","https:\u002F\u002Forcid.org\u002F0000-0002-5171-886X",{"tldr":496,"method":497,"finding":498,"direction":97,"opportunity":499},"综述2021-2026年AI驱动精准农业系统，提出气候韧性粮食生产工程框架。","综述机器学习、深度学习、强化学习、计算机视觉、机器人与物联网在精准农业中的应用。","CNN-LSTM混合模型和强化学习在产量预测与灌溉调度上优于单一方法，边缘计算提升实时性。","可研究边缘计算与强化学习结合的轻量级灌溉决策系统，并探索发展中国家低成本部署与标准化路径。","2026-10-22T11:20:18.151376Z"]