[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-3536":3,"related-3536":54},{"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":24,"tags":26,"search_phrases":32,"slug":35,"view_count":36,"doi":37,"paper":38,"created_at":53},3536,"Functional photonic and optoelectronic materials and devices for climate-resilient smart agriculture: a review","https:\u002F\u002Fdoi.org\u002F10.3389\u002Ffmats.2026.1943334","The variation in the temperature, precipitation, and extreme events has rendered climate change very threatening to crop production and the overall world food security. The prospects of sustainable and climate resilient agricultural production is brought up with the use of photonics and optoelectronics coupled with smart farming technology. The paper offers an overall foundation of smart agricultural systems that are assisted by photonics and optoelectronics that combine the state-of-the-art optical sensors, Internet of Things (IoT) communications, and machine learning algorithms in the crop and soil monitoring. Due to optical sensing technologies including fiber optic sensors, LiDAR, fluorescence spectroscopy, and hyperspectral imaging, optical sensors can also be used in high resolution and non-destructive measurement of physiological properties of plants, nutrient status, water stress and disease incidence. With the help of optoelectronic devices, it is possible to carry out the accurate signal processing, data collection, automated control of irrigation, fertigation and microclimate regulating systems. When sensor-based information is incorporated with any decision support systems the quality of early stress detection is improved, wastage of resources is minimized and the effect on the environment is minimized. Multispectral and temporal data based on machine learning models enhance forecasting of climate risk, diseases, and pest outbreaks forecasting, and crop yield forecasting. The system will aid in making the contemporary crop production systems more sustainable and resilient in the long-term due to the fact that it will provide scalable and flexible solutions across the agroecological regions.","温度、降水及极端事件的变化使气候变化对作物生产和全球粮食安全构成严重威胁。光子学与光电子学结合智慧农业技术，为实现可持续且气候韧性的农业生产带来了前景。本文提供了由光子学和光电子学辅助的智慧农业系统的总体基础，该系统将先进的光学传感器、物联网（IoT）通信和机器学习算法结合应用于作物与土壤监测。借助光纤传感器、激光雷达（LiDAR）、荧光光谱和高光谱成像等光学传感技术，光学传感器还可用于高分辨率、非破坏性地测量植物生理特性、养分状况、水分胁迫和病害发生情况。借助光电子器件，可以实现精确的信号处理、数据采集以及灌溉、施肥和微气候调节系统的自动控制。当基于传感器的信息与决策支持系统相结合时，早期胁迫检测的质量得以提高，资源浪费降至最低，对环境的影响也降至最小。基于机器学习模型的多光谱和时间序列数据增强了气候风险、病虫害暴发预测以及作物产量预测的能力。该系统将有助于使当代作物生产系统在长期内更具可持续性和韧性，因为它将为各农业生态区域提供可扩展且灵活的解决方案。",null,"Frontiers in Materials","2026-09-25T00:00:00Z","论文",10,false,76,{"impact":17,"substance":18,"depth":19,"authority":20,"freshness":21,"relevant":22,"comment":23},18,20,17,13,8,1,"系统综述光子与光电子技术在气候韧性智慧农业中的应用，方法覆盖全面、结论可靠，对农业信息化领域有较高参考价值。",[25],{"name":10,"url":6},[27,28,29,30,31],"智慧农业","农业物联网","机器学习","农业传感器","光谱遥感",[33,34],"光子学 光电材料 智慧农业","光纤传感 高光谱成像 作物监测","光子学光电材料智慧农业-3536",0,"10.3389\u002Ffmats.2026.1943334",{"doi":37,"openalex_id":39,"authors":40,"venue":10,"cited_by_count":36,"oa_url":6,"card":46,"direction":50,"ingested_from":52},"W7214293156",[41,43],{"name":42,"orcid":9},"Karthika Vishnu Priya Kathula",{"name":44,"orcid":45},"Murugesan Mohana Keerthi","https:\u002F\u002Forcid.org\u002F0000-0002-2018-0200",{"tldr":47,"method":48,"finding":49,"direction":50,"opportunity":51},"综述光子与光电子材料器件结合物联网和机器学习，支撑气候韧性智慧农业的作物与土壤监测。","综述光纤传感、LiDAR、荧光光谱、高光谱成像与IoT、机器学习融合方案。","光学传感与光电器件可实现无损高分辨监测，提升早期胁迫检测并减少资源浪费。","智慧农业 \u002F 农业物联网","可探索低成本光子传感器与轻量ML模型在田间边缘端的集成及跨生态区泛化验证。","openalex","2026-09-26T23:30:09.644423Z",{"total":55,"page":22,"page_size":55,"items":56},6,[57,104,136,170,204,258],{"id":58,"title":59,"url":60,"summary":61,"summary_zh":62,"content":9,"source_name":63,"source_url":60,"published_at":64,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":65,"score_detail":66,"sources":70,"tags":72,"search_phrases":75,"slug":78,"view_count":36,"doi":79,"paper":80,"created_at":103},2803,"Volatile Fingerprinting Empowers Salinity Monitoring in Peppermint Using MOS Sensors and Feature-Optimized Machine Learning","https:\u002F\u002Fdoi.org\u002F10.3390\u002Felectronics15184233","Early detection of salinity stress is essential for precision agriculture, particularly in scalable, resource-constrained monitoring systems. This study presents a portable sensing module integrating a low-cost metal oxide semiconductor (MOS) sensor array with potential application for edge deployment to detect salinity stress in peppermint. Salinity significantly reduced plant biomass, confirming physiological stress induction. Volatile organic compound (VOC) fingerprints were collected over eleven consecutive days in a controlled enclosure. Sensor signals underwent outlier filtering, normalization, and smoothing, while treatment discrimination was verified using the Kruskal–Wallis test. Thirty-three machine learning models were evaluated using a 75:25 train–test split with five-fold cross-validation. Wide neural network models achieved the highest predictive performance, exceeding 98% test accuracy and a 97% macro F1 score. Feature adequacy analysis showed that six sensors captured the dominant variance required for reliable classification. Considering computational constraints, a bilayered neural network using only six features maintained over 97% accuracy with a memory footprint of 0.008 MB while remaining Pareto optimal. These findings support the feasibility of a compact, computationally efficient, and edge-compatible VOC sensing framework for salinity stress detection in precision agriculture and intelligent crop monitoring.","盐胁迫的早期检测对精准农业至关重要，尤其是在可扩展、资源受限的监测系统中。本研究提出了一种便携式传感模块，将低成本金属氧化物半导体（MOS）传感器阵列集成其中，具备边缘部署的应用潜力，用于检测薄荷中的盐胁迫。盐胁迫显著降低了植物生物量，证实了生理胁迫的诱导作用。在受控密闭环境中连续十一天采集了挥发性有机化合物（VOC）指纹图谱。对传感器信号进行了异常值过滤、归一化和平滑处理，并使用Kruskal–Wallis检验验证了处理组间的区分度。采用75:25的训练-测试划分和五折交叉验证评估了三十三种机器学习模型。宽神经网络模型取得了最高的预测性能，测试准确率超过98%，宏F1分数达到97%。特征充分性分析表明，六个传感器即可捕获可靠分类所需的主要方差。考虑到计算约束，仅使用六个特征的双层神经网络在保持超过97%准确率的同时，内存占用仅为0.008 MB，且仍处于帕累托最优。这些发现支持了一种紧凑、计算高效且兼容边缘计算的VOC传感框架用于精准农业和智能作物监测中盐胁迫检测的可行性。","Electronics","2026-09-17T00:00:00Z",78,{"impact":67,"substance":68,"depth":17,"authority":20,"freshness":13,"relevant":22,"comment":69},15,22,"低成本MOS传感器阵列结合特征优化机器学习实现薄荷盐胁迫早期无损检测，方法新颖、数据扎实，对边缘部署式作物监测有参考价值。",[71],{"name":63,"url":60},[27,29,73,30,74],"精准农业","盐胁迫监测",[76,77],"农业传感器 盐胁迫监测 智慧农业 机器学习","农业传感器 盐胁迫监测","农业传感器盐胁迫监测智慧农业机器学习-2803","10.3390\u002Felectronics15184233",{"doi":79,"openalex_id":81,"authors":82,"venue":63,"cited_by_count":36,"oa_url":60,"card":97,"direction":102,"ingested_from":52},"W7213452763",[83,86,89,91,94],{"name":84,"orcid":85},"Ahmad Ali","https:\u002F\u002Forcid.org\u002F0000-0001-5530-7374",{"name":87,"orcid":88},"Vinie Lee Silva Alvarado","https:\u002F\u002Forcid.org\u002F0009-0000-5857-3248",{"name":90,"orcid":9},"Arman Heydari",{"name":92,"orcid":93},"Sandra Sendra","https:\u002F\u002Forcid.org\u002F0000-0001-9556-9088",{"name":95,"orcid":96},"Jaime Lloret","https:\u002F\u002Forcid.org\u002F0000-0002-0862-0533",{"tldr":98,"method":99,"finding":100,"direction":50,"opportunity":101},"用低成本MOS传感器阵列采集薄荷VOC指纹，结合特征优化机器学习实现盐胁迫检测。","11天VOC指纹采集，33种机器学习模型，五折交叉验证，特征充分性分析。","宽神经网络准确率超98%，仅用6个特征的双层网络保持97%以上且内存仅0.008MB。","可探索多作物VOC指纹迁移学习与田间边缘设备长期稳定性验证。","农业人工智能与决策模型","2026-09-17T23:30:59.249847Z",{"id":105,"title":106,"url":107,"summary":108,"summary_zh":109,"content":9,"source_name":110,"source_url":107,"published_at":111,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":112,"score_detail":113,"sources":116,"tags":118,"search_phrases":120,"slug":123,"view_count":36,"doi":124,"paper":125,"created_at":135},2652,"ARTIFICIAL INTELLIGENCE IN HIGHER EDUCATION: TRANSFORMING TEACHING, LEARNING, AND STUDENT ENGAGEMENT","https:\u002F\u002Fdoi.org\u002F10.65725\u002Fijhlt\u002F1\u002F2\u002F001","Efficient irrigation management is essential for sustainable agriculture, particularly in the context of increasing freshwater scarcity and the growing imperative to optimize crop productivity. Conventional irrigation practices rely predominantly on fixed time schedules or manual field assessments, which frequently induce over-irrigation, root-zone nutrient leaching, under-irrigation water stress, and substantial resource inefficiency. This paper proposes an Intelligent Irrigation Management System that integrates Internet of Things (IoT) sensing architectures, multi-parameter environmental telemetry, and supervised machine learning (ML) algorithms to facilitate dynamic, data-driven, and automated irrigation control. The proposed system continuously acquires real-time field data—including soil moisture, ambient temperature, relative humidity, soil temperature, and rainfall—via deployed sensor nodes managed by an ESP32 microcontroller pipeline. The telemetry stream is transmitted through low-power communication channels to a centralized processing engine, where a Random Forest classification model evaluates multidimensional soil-environmental interactions to predict immediate irrigation requirements. The predicted states feed into an automated actuation layer that directly modulates a solenoid-valve and water-pump relay, forming a closed-loop feedback pipeline. Evaluated against traditional threshold-based and schedule-driven approaches, the proposed IoT-ML framework demonstrates superior operational responsiveness, minimizes unnecessary water application, and offers a robust, scalable architectural template for modern precision agriculture.","高效灌溉管理对可持续农业至关重要，尤其是在淡水日益稀缺、优化作物生产力需求不断增长的背景下。传统灌溉实践主要依赖固定时间表或人工田间评估，这常常导致过度灌溉、根区养分淋失、灌溉不足引起的水分胁迫以及严重的资源低效。本文提出了一种智能灌溉管理系统，该系统集成了物联网（IoT）感知架构、多参数环境遥测以及监督式机器学习（ML）算法，以实现动态、数据驱动和自动化的灌溉控制。所提出的系统通过由ESP32微控制器管道管理的部署传感器节点，持续采集实时田间数据——包括土壤湿度、环境温度、相对湿度、土壤温度和降雨量。遥测数据流通过低功耗通信信道传输至集中处理引擎，其中随机森林分类模型评估多维土壤-环境相互作用，以预测即时灌溉需求。预测状态输入自动执行层，直接调节电磁阀和水泵继电器，形成闭环反馈管道。与传统基于阈值和时间表驱动的方法相比，所提出的IoT-ML框架展现出更优的运行响应能力，最大限度地减少了不必要的灌溉用水，并为现代精准农业提供了一种稳健、可扩展的架构模板。","INTERNATIONAL JOURNAL OF HUMANITIES AND LEARNING TECHNOLOGY INNOVATION (IJHLT)","2026-09-15T00:00:00Z",72,{"impact":114,"substance":17,"depth":19,"authority":20,"freshness":21,"relevant":22,"comment":115},16,"论文提出IoT与随机森林融合的闭环智能灌溉系统，方法完整、数据驱动，对节水农业有参考价值，但标题与摘要主题不符需核实。",[117],{"name":110,"url":107},[27,28,29,119,73],"智能灌溉",[121,122],"农业物联网 智慧农业 智能灌溉 机器学习","农业物联网 智慧农业","农业物联网智慧农业智能灌溉机器学习-2652","10.65725\u002Fijhlt\u002F1\u002F2\u002F001",{"doi":124,"openalex_id":126,"authors":127,"venue":110,"cited_by_count":36,"oa_url":9,"card":130,"direction":50,"ingested_from":52},"W7213277724",[128],{"name":129,"orcid":9},"M. Rathamani",{"tldr":131,"method":132,"finding":133,"direction":50,"opportunity":134},"提出融合物联网传感与随机森林的智能灌溉系统，实现数据驱动的自动灌溉控制。","ESP32传感器节点采集土壤温湿度等数据，随机森林分类预测灌溉需求。","相比传统定时或阈值方法，该系统响应更优、减少不必要灌溉，可扩展性强。","可探索多模态数据融合与边缘智能，提升灌溉决策的实时性与泛化能力。","2026-09-16T23:30:16.194086Z",{"id":137,"title":138,"url":139,"summary":140,"summary_zh":141,"content":9,"source_name":142,"source_url":139,"published_at":143,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":65,"score_detail":144,"sources":147,"tags":149,"search_phrases":152,"slug":155,"view_count":36,"doi":156,"paper":157,"created_at":169},2149,"Smart biosensing ecosystems: Integrating nanotechnology, chemometrics, and machine learning for precision analytical applications","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.nxmate.2026.103476","The growing complexity of food, environmental, and biological matrices has intensified demand for analytical technologies that are rapid, selective, and deployable outside centralized laboratories, exposing the limitations of conventional chromatographic and single-analyte biosensing approaches. This review introduces the concept of a Smart Biosensing Ecosystem, a framework in which biorecognition elements (enzymes, antibodies, nucleic acids, aptamers, whole cells), functional nanomaterials, and electrochemical or optical transducers are integrated with chemometric preprocessing, machine-learning-based interpretation, IoT connectivity, and cloud-enabled decision support into a single analytical pipeline. Within this framework, the review examines smart biosensor architecture, biorecognition mechanisms, enzyme-based catalytic sensing, electrochemical and optical transduction physics, and a comparative evaluation of graphene, MXenes, carbon nanotubes, quantum dots, and metal-based nanomaterials for signal amplification and multiplexing. It further surveys chemometric tools (PCA, PLS, LDA) and machine-learning algorithms (SVM, Random Forest, ANN, CNN, gradient boosting, ensemble learning) for model development and validation, alongside wearable, multiplexed, and hybrid electrochemical–optical platforms applied to food-quality, environmental, and biomedical analysis. Persistent challenges, including biofouling, batch-to-batch reproducibility, scalable manufacturing, regulatory translation, explainable AI, and data governance, are critically discussed as barriers limiting laboratory-to-market translation. The review concludes that continued convergence of nanotechnology, intelligent analytics, and connected infrastructure is poised to transform biosensors from isolated transduction devices into autonomous, self-calibrating, and predictive analytical ecosystems with substantial translational potential across precision food, environmental, and biomedical monitoring.","食品、环境和生物基质的日益复杂，加剧了对快速、选择性且可在中心实验室之外部署的分析技术的需求，也暴露了传统色谱方法和单分析物生物传感方法的局限性。本综述引入了“智能生物传感生态系统”的概念，该框架将生物识别元件（酶、抗体、核酸、适配体、全细胞）、功能纳米材料以及电化学或光学换能器，与化学计量学预处理、基于机器学习的解析、物联网连接和云端决策支持整合为单一分析流程。在该框架内，本综述考察了智能生物传感器架构、生物识别机制、基于酶的催化传感、电化学与光学换能物理，并对石墨烯、MXenes、碳纳米管、量子点和金属基纳米材料在信号放大与多重检测方面进行了比较评估。文章进一步综述了用于模型开发与验证的化学计量学工具（PCA、PLS、LDA）和机器学习算法（SVM、随机森林、ANN、CNN、梯度提升、集成学习），以及应用于食品质量、环境和生物医学分析的可穿戴、多重和混合电化学–光学平台。文章批判性地讨论了持续存在的挑战，包括生物污损、批次间重现性、可规模化制造、法规转化、可解释人工智能和数据治理，这些是限制实验室到市场转化的障碍。综述得出结论：纳米技术、智能分析和互联基础设施的持续融合，有望将生物传感器从孤立的换能设备转变为自主、自校准和预测性分析生态系统，在精准食品、环境和生物医学监测方面具有巨大的转化潜力。","Next Materials","2026-09-09T00:00:00Z",{"impact":17,"substance":18,"depth":17,"authority":145,"freshness":21,"relevant":22,"comment":146},14,"综述提出智能生物传感生态系统框架，整合纳米材料、化学计量学与机器学习，对农产品质量与食品安全精准检测有较强参考价值，但属实验室前沿综述，产业落地尚远。",[148],{"name":142,"url":139},[27,29,150,151,30],"食品安全","纳米材料",[153,154],"农业传感器 智慧农业 机器学习 纳米材料","农业传感器 智慧农业","农业传感器智慧农业机器学习纳米材料-2149","10.1016\u002Fj.nxmate.2026.103476",{"doi":156,"openalex_id":158,"authors":159,"venue":142,"cited_by_count":36,"oa_url":163,"card":164,"direction":50,"ingested_from":52},"W7212013524",[160],{"name":161,"orcid":162},"Kushagra Sharma","https:\u002F\u002Forcid.org\u002F0000-0003-1126-8697","https:\u002F\u002Fwww.sciencedirect.com\u002Fscience\u002Farticle\u002Fpii\u002FS2949822826018939\u002Fpdf",{"tldr":165,"method":166,"finding":167,"direction":102,"opportunity":168},"综述提出智能生物传感生态系统，融合纳米材料、化学计量学、机器学习与物联网，实现精准分析。","综述生物识别、纳米材料、电化学\u002F光学传感、PCA\u002FPLS及SVM\u002FCNN等机器学","纳米技术、智能分析与互联基础设施融合，可将生物传感器变为自主预测性分析生态系统。","可探索面向农业食品链的智能生物传感生态系统，解决生物污损、可解释AI与数据治理等落地瓶颈。","2026-09-11T23:30:12.070495Z",{"id":171,"title":172,"url":173,"summary":174,"summary_zh":9,"content":9,"source_name":175,"source_url":173,"published_at":176,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":177,"score_detail":178,"sources":182,"tags":184,"search_phrases":186,"slug":187,"view_count":22,"doi":188,"paper":189,"created_at":203},1258,"Integration of IoT and Sensor Technologies for Sustainable Smart Irrigation Systems: Trends, Challenges, and Future Directions","https:\u002F\u002Fdoi.org\u002F10.31272\u002Fajece.35","Efficient water management is a persistent challenge in modern agriculture, especially in arid and semi-arid regions. The development and adoption of advanced soil sensor technologies are essential for optimising water use and supporting sustainable, high-yield agricultural systems. This systematic review compares primary soil sensor methodologies and synthesises findings from 156 peer-reviewed studies published between 2015 and 2025. The review assesses dielectric techniques, including time-domain reflectometry (TDR) and frequency-domain reflectometry (FDR), as well as capacitive sensors, tensiometers, neutron probes, and instruments for measuring salinity and pH. Meta-analytical results show that TDR systems achieve high accuracy (pooled RMSE: 0.016 m³\u002Fm³) but are associated with significant costs ($500–$1200 per unit). In contrast, low-cost capacitive sensors can provide acceptable accuracy (RMSE: 0.045 m³\u002Fm³) when rigorously calibrated. LoRaWAN (Long Range Wide Area Network) is identified as the most effective communication protocol for agricultural Internet of Things (IoT) applications, with transmission ranges of 2–15 km and battery lifespans of 2–10 years. The use of machine learning methods improves irrigation scheduling accuracy by 15–30%, and artificial intelligence (AI) systems can achieve water savings of 35–50%. Key research gaps include the absence of standardised calibration procedures, as 67% of studies lack comprehensive validation, limited long-term stability assessments, with only 23% reporting data beyond six months, and insufficient investigation of multi-sensor data fusion. The review concludes by identifying future research priorities to support the development of robust and scalable innovative irrigation systems for sustainable agriculture.","Academic Journal of Electrical and Computer Engineering","2026-08-30T00:00:00Z",81,{"impact":68,"substance":179,"depth":18,"authority":20,"freshness":180,"relevant":22,"comment":181},24,2,"系统综述整合156项研究，对比多种土壤传感器与通信协议，数据详实，对智慧灌溉技术发展有重要参考价值。",[183],{"name":175,"url":173},[27,28,29,119,185],"传感器",[121,122],"农业物联网智慧农业智能灌溉机器学习-1258","10.31272\u002Fajece.35",{"doi":188,"openalex_id":190,"authors":191,"venue":175,"cited_by_count":36,"oa_url":173,"card":198,"direction":50,"ingested_from":52},"W7204712150",[192,194,196],{"name":193,"orcid":9},"Omar Talib Khazraji",{"name":195,"orcid":9},"Marwan J. Hussein",{"name":197,"orcid":9},"Ahmed M. Almawla",{"tldr":199,"method":200,"finding":201,"direction":50,"opportunity":202},"系统综述2015-2025年土壤传感器与IoT在智能灌溉中的应用，评估技术、成本与性能，指出研究空白","系统综述156篇研究，比较TDR、FDR、电容等传感器，结合LoRaWAN和机器","TDR精度高但贵，电容传感器性价比好；LoRaWAN最优；AI节水35-50%；缺乏标准化校准和长期","研究空白：多传感器数据融合、标准化校准协议、长期稳定性评估，可开发低成本高精度融合系统。","2026-09-01T04:03:11.465487Z",{"id":205,"title":206,"url":207,"summary":208,"summary_zh":209,"content":9,"source_name":210,"source_url":207,"published_at":11,"category":12,"cover_url":9,"hotness":13,"is_selected":211,"score":212,"score_detail":213,"sources":218,"tags":220,"search_phrases":224,"slug":227,"view_count":36,"doi":228,"paper":229,"created_at":257},3548,"Cybersecurity and Privacy in AI-Enabled Agricultural IoT Ecosystems: A Systematic Review of Threats, Safeguards, and Resilience Gaps","https:\u002F\u002Fdoi.org\u002F10.3390\u002Fa19100827","Agricultural Internet of Things (IoT) ecosystems increasingly connect sensors, drones, edge devices, and cloud platforms to support precision farming, yet cybersecurity, privacy, and the real-world readiness of proposed safeguards remain fragmented across the literature. This study systematically reviewed cybersecurity threats, privacy concerns, AI-driven and traditional safeguards, and evidence gaps in agricultural IoT research published between 2015 and 2025. Following the Kitchenham and Charters methodology, 103 studies were selected from 2535 records retrieved across five databases. STRIDE and LINDDUN were retrospectively applied as complementary frameworks for threat and privacy classification. Because the coding scheme was multi-label, reliability was assessed at the category level using presence\u002Fabsence decisions on a 20-study sample and observed agreement ranged from 75% to 95% for STRIDE and 95% to 100% for LINDDUN, with interpretable Cohen’s κ values ranging from 0.348 to 0.794 and 0.875 to 1.000, respectively. All included studies also underwent quality appraisal and a supplementary ecological-validity assessment. Denial-of-service, tampering, and spoofing were the most frequently reported threats, concentrated at the device, network, and cloud layers, while the edge layer remained underexamined. AI- and machine-learning-based intrusion detection and privacy-preserving methods such as federated learning emerged as prominent safeguards, but adversarial manipulation of agricultural AI models received limited attention. Privacy research remained oriented toward confidentiality, with 90.3% of studies referencing no applicable regulatory framework. Most importantly, only 8 of 103 studies (7.8%) received a High ecological-validity rating, showing how rarely the evidence base is grounded in real agricultural field conditions. The review identifies field-grounded evaluation, adversarially robust AI, privacy governance, and cyber resilience as priorities for future agricultural IoT security research.","农业物联网（IoT）生态系统日益将传感器、无人机、边缘设备和云平台连接起来，以支持精准农业，然而网络安全、隐私以及所提出保障措施的现实适用性在文献中仍呈现碎片化状态。本研究系统综述了2015年至2025年间发表的农业物联网研究中的网络安全威胁、隐私问题、人工智能驱动及传统保障措施以及证据缺口。遵循Kitchenham和Charters方法论，从五个数据库检索到的2535条记录中筛选出103项研究。STRIDE和LINDDUN被回溯性应用为威胁与隐私分类的互补框架。由于编码方案为多标签，可靠性在类别层面通过20项研究样本的存在\u002F缺失判定进行评估，STRIDE的观察一致率为75%至95%，LINDDUN为95%至100%，可解释的Cohen's κ值分别为0.348至0.794和0.875至1.000。所有纳入研究还接受了质量评价和补充性生态效度评估。拒绝服务、篡改和欺骗是报告最频繁的威胁，集中在设备层、网络层和云层，而边缘层仍未被充分考察。基于人工智能和机器学习的入侵检测以及联邦学习等隐私保护方法成为突出的保障措施，但农业人工智能模型的对抗性操纵受到的关注有限。隐私研究仍以保密性为导向，90.3%的研究未引用任何适用的监管框架。最重要的是，103项研究中仅有8项（7.8%）获得高生态效度评级，表明证据基础鲜有扎根于真实农业田间条件。本综述将田间实证评估、对抗鲁棒人工智能、隐私治理和网络韧性确定为未来农业物联网安全研究的优先事项。","Algorithms",true,82,{"impact":17,"substance":214,"depth":215,"authority":20,"freshness":216,"relevant":22,"comment":217},23,19,9,"系统综述103项研究，揭示农业物联网安全证据多脱离田间实际，对智慧农业安全研究有较高参考价值。",[219],{"name":210,"url":207},[27,221,28,222,223],"农业人工智能","隐私保护","数据安全",[225,226],"农业物联网 网络安全","农业AI 隐私保护 联邦学习","农业物联网网络安全-3548","10.3390\u002Fa19100827",{"doi":228,"openalex_id":230,"authors":231,"venue":210,"cited_by_count":36,"oa_url":207,"card":252,"direction":50,"ingested_from":52},"W7214403401",[232,235,238,241,244,247,249],{"name":233,"orcid":234},"Emmanuel Kojo Gyamfi","https:\u002F\u002Forcid.org\u002F0009-0002-0441-2830",{"name":236,"orcid":237},"Jess Kropczynski","https:\u002F\u002Forcid.org\u002F0000-0002-7458-6003",{"name":239,"orcid":240},"Jacques Bou Abdo","https:\u002F\u002Forcid.org\u002F0000-0002-3482-9154",{"name":242,"orcid":243},"Joseph Samuel Johnson","https:\u002F\u002Forcid.org\u002F0000-0003-2555-8142",{"name":245,"orcid":246},"Mustapha Awinsongya Yakubu","https:\u002F\u002Forcid.org\u002F0009-0005-6623-0858",{"name":248,"orcid":9},"Anthony Tsetse",{"name":250,"orcid":251},"Gertrude Kaneah Abagale","https:\u002F\u002Forcid.org\u002F0009-0009-1502-3188",{"tldr":253,"method":254,"finding":255,"direction":50,"opportunity":256},"系统综述2015-2025年农业物联网的网络安全、隐私威胁与防护措施及证据缺口。","Kitchenham系统综述法，筛选103项研究，用STRIDE与LINDDUN","拒绝服务、篡改、欺骗威胁最多，边缘层研究不足，仅7.8%研究具高生态效度。","农业AI模型的对抗鲁棒性、边缘层安全、隐私治理与真实田间条件下的韧性评估是明显空白。","2026-09-26T23:30:15.415998Z",{"id":259,"title":260,"url":261,"summary":262,"summary_zh":263,"content":9,"source_name":264,"source_url":261,"published_at":265,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":266,"score_detail":267,"sources":269,"tags":271,"search_phrases":274,"slug":277,"view_count":36,"doi":278,"paper":279,"created_at":302},3516,"Approaches to forecast soil nutrient dynamics for precision agriculture and sustainable fertiliser management: A review","https:\u002F\u002Fdoi.org\u002F10.14719\u002Fpst.16160","Predictive modelling of soil nutrient dynamics is an essential tool for promoting sustainable agricultural practices and environmentally responsible farming methods. The statistical and machine learning techniques used to forecast the availability and dynamics of soil nutrients are summarised in this review. The core frameworks for measuring spatio-temporal nutritional variability are established by traditional statistical approaches such as time-series models autoregressive integrated moving average (ARIMA), seasonal autoregressive integrated moving average (SARIMA), multivariate techniques (Principal component analysis (PCA) and factor analysis) and geostatistical tools (kriging). By capturing intricate nonlinear interactions within heterogeneous agroecosystems, machine learning techniques like random forest, support vector machines and ensemble approaches (XGBoost, LightGBM and AdaBoost) provide higher prediction accuracy. Forecasting capabilities are further enhanced by hybrid frameworks [Autoregressive integrated moving average with exogenous variables–artificial neural network. (ARIMAX-ANN)] and deep learning architectures (Convolutional neural network (CNN), long short-term memory (LSTM), ANN). With R2 values above 0.93 and notable decreases in prediction errors, ensemble approaches routinely perform better than traditional linear models. Nevertheless, persistent challenges include data quality limitations, spatial sampling constraints, insufficient environmental covariates and reduced model transferability across diverse pedoclimatic regions. Integrating high-resolution soil properties, climatic variables, terrain attributes and spectral information with advanced modelling architectures remains crucial for enhancing predictive reliability, ultimately supporting precision nutrient management, improved fertiliser efficiency and environmentally responsible agricultural systems.","土壤养分动态的预测建模是推动可持续农业实践和环境友好型耕作方法的重要工具。本综述总结了用于预测土壤养分有效性及其动态变化的统计与机器学习技术。传统统计方法，如时间序列模型自回归积分滑动平均模型（ARIMA）、季节性自回归积分滑动平均模型（SARIMA）、多变量技术（主成分分析（PCA）和因子分析）以及地统计工具（克里金法），为量化养分的时空变异性奠定了核心框架。通过捕捉异质性农业生态系统中复杂的非线性相互作用，随机森林、支持向量机和集成方法（XGBoost、LightGBM和AdaBoost）等机器学习技术可实现更高的预测精度。混合框架[含外生变量的自回归积分滑动平均模型–人工神经网络（ARIMAX-ANN）]和深度学习架构（卷积神经网络（CNN）、长短期记忆网络（LSTM）、人工神经网络（ANN））进一步增强了预测能力。集成方法的R²值超过0.93，且预测误差显著降低，其表现通常优于传统线性模型。然而，持续存在的挑战包括数据质量限制、空间采样约束、环境协变量不足以及模型在不同土壤气候区域间可迁移性降低等问题。将高分辨率土壤属性、气候变量、地形属性和光谱信息与先进建模架构相结合，对于提高预测可靠性仍然至关重要，最终可为精准养分管理、提高肥料利用效率以及环境友好型农业系统提供支撑。","Plant Science Today","2026-09-24T00:00:00Z",79,{"impact":17,"substance":68,"depth":17,"authority":20,"freshness":21,"relevant":22,"comment":268},"系统综述土壤养分动态预测的统计与机器学习方法，方法体系完整、结论有量化支撑，对精准施肥与农业信息化有参考价值，但属综述类论文，产业级影响有限。",[270],{"name":264,"url":261},[27,272,29,73,273],"变量施肥","土壤养分",[275,276],"土壤养分 预测模型 精准农业","机器学习 施肥管理 可持续农业","土壤养分预测模型精准农业-3516","10.14719\u002Fpst.16160",{"doi":278,"openalex_id":280,"authors":281,"venue":264,"cited_by_count":36,"oa_url":261,"card":297,"direction":102,"ingested_from":52},"W7214167059",[282,285,288,291,294],{"name":283,"orcid":284},"R Rathna","https:\u002F\u002Forcid.org\u002F0009-0004-7797-2673",{"name":286,"orcid":287},"B Sivasankari","https:\u002F\u002Forcid.org\u002F0000-0001-9921-8170",{"name":289,"orcid":290},"R. Gangai Selvi","https:\u002F\u002Forcid.org\u002F0000-0002-4475-2293",{"name":292,"orcid":293},"J Prabhakaran","https:\u002F\u002Forcid.org\u002F0000-0001-7339-175X",{"name":295,"orcid":296},"K. G. Sabarinathan","https:\u002F\u002Forcid.org\u002F0000-0002-8659-6479",{"tldr":298,"method":299,"finding":300,"direction":102,"opportunity":301},"综述土壤养分动态预测的统计与机器学习方法，比较精度与局限。","综述ARIMA、地统计、随机森林、XGBoost、CNN\u002FLSTM及混合模型。","集成与深度学习模型精度更高（R²>0.93），但数据质量与跨区迁移性仍是瓶颈。","可研究多源遥感与气候数据融合的迁移学习模型，提升跨区域养分预测泛化能力。","2026-09-25T23:30:54.950445Z"]