[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-4576":3,"related-4576":99},{"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,"primary_topic":32,"secondary_topics":38,"classification":39,"search_phrases":72,"slug":75,"view_count":22,"doi":76,"paper":77,"created_at":98},4576,"Compressing Semantic Segmentation Models for Agricultural IoT: A Systematic Study of Pruning Trade-Offs for Crop–Weed Segmentation","https:\u002F\u002Fdoi.org\u002F10.3390\u002Fjimaging12100495","Semantic segmentation of crops and weeds is an important component of selective weeding systems, but accurate segmentation networks can impose substantial computational and memory demands on resource-constrained agricultural platforms. This paper presents a systematic study of the accuracy–efficiency trade-offs involved in compressing semantic segmentation networks for agricultural edge computing. Six architectures spanning conventional CNN encoder–decoders, atrous and pyramid-pooling decoders, a mobile-oriented CNN, and two transformer scales are evaluated on the Crop\u002FWeed Field Image Dataset (CWFID), with three random seeds per architecture. The most favourable compact model is subsequently evaluated under unstructured global magnitude pruning and dependency-aware structured channel pruning, using one-shot and iterative prune-and-fine-tune schedules at sparsity levels from 10% to 80%, with three seeds per configuration. Among the dense baselines, U-Net with a MobileNetV3-Large encoder matches the accuracy of the larger SegFormer-B2 model (88.85% versus 88.67% mIoU, a difference of 0.18% across three seeds, p = 0.58) with 3.7× fewer parameters than SegFormer-B2. Iterative unstructured pruning reaches 80% weight sparsity while retaining 89.04% mIoU; however, because the dense tensor dimensions are unchanged, this sparsity does not reduce the parameter count, GFLOPs, or model size on conventional hardware. Structured channel pruning, which does reduce all three, remains viable only up to 30% sparsity, reducing GFLOPs by 2.1× at a cost of 4.21 mIoU points and collapsing beyond 40%. Without fine-tuning, structured pruning causes severe degradation even at 10% channel sparsity. The results show that architecture selection has a major influence on the accuracy–efficiency trade-off, while the effects of pruning depend strongly on the pruning structure and fine-tuning strategy. Crop segmentation consistently degrades before the aggregate mIoU indicates severe failure, highlighting the need for class-wise evaluation when compressing crop–weed segmentation models.","作物与杂草的语义分割是选择性除草系统的重要组成部分，但精确的分割网络会对资源受限的农业平台造成大量的计算和内存需求。本文针对农业边缘计算中语义分割网络压缩所涉及的精度-效率权衡进行了系统性研究。在Crop\u002FWeed Field Image Dataset（CWFID）数据集上评估了六种架构，涵盖传统CNN编码器-解码器、空洞卷积与金字塔池化解码器、面向移动端的CNN以及两种规模的Transformer，每种架构使用三个随机种子。随后对最优紧凑模型在非结构化全局幅度剪枝和依赖感知结构化通道剪枝下进行评估，采用一次性及迭代式剪枝-微调方案，稀疏度从10%到80%，每种配置使用三个种子。在稠密基线中，采用MobileNetV3-Large编码器的U-Net在精度上匹配了更大的SegFormer-B2模型（88.85%对88.67% mIoU，三个种子间差异为0.18%，p = 0.58），而参数量比SegFormer-B2少3.7倍。迭代式非结构化剪枝在保持89.04% mIoU的同时达到80%权重稀疏度；然而，由于稠密张量维度不变，该稀疏度在常规硬件上并不减少参数量、GFLOPs或模型大小。结构化通道剪枝虽能同时降低这三者，但仅在30%稀疏度以内可行，此时GFLOPs降低2.1倍，代价是损失4.21个mIoU点，超过40%后性能崩溃。在不进行微调的情况下，结构化剪枝即使在10%通道稀疏度下也会导致严重退化。结果表明，架构选择对精度-效率权衡具有重大影响，而剪枝的效果则强烈依赖于剪枝结构和微调策略。作物分割始终在总体mIoU显示严重失败之前就已退化，凸显了在压缩作物-杂草分割模型时进行逐类评估的必要性。",null,"Journal of Imaging","2026-10-08T00:00:00Z","论文",10,false,78,{"impact":17,"substance":18,"depth":19,"authority":20,"freshness":21,"relevant":22,"comment":23},16,22,18,13,9,1,"系统评估语义分割模型剪枝精度-效率权衡，含多架构多稀疏度对比数据，对农业边缘部署有实质参考价值。",[25],{"name":10,"url":6},[27,28,29,30,31],"智慧农业","农业人工智能","农业物联网","杂草识别","模型轻量化",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},"I5","agricultural-iot","物联网传感与边缘计算","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-iot","I",[],{"status":40,"locked":14,"version":41,"method":42,"evidence":43},"classified","2026-10-22.2","rules",{"candidates":44,"agricultural_evidence":47},[45,54,61,67],{"code":33,"score":46,"supported":47,"title":48,"abstract":50,"tags":53},15,true,[49],"IoT",[51,52],"边缘计算","edge computing",[29],{"code":55,"score":56,"supported":47,"title":57,"abstract":58,"tags":60},"F4",4,[],[59],"除草",[30],{"code":62,"score":63,"supported":14,"title":64,"abstract":65,"tags":66},"I2",2,[],[],[28,31],{"code":68,"score":22,"supported":14,"title":69,"abstract":70,"tags":71},"I1",[],[],[27],[73,74],"CWFID 作物 杂草 分割","MobileNetV3 U-Net 剪枝","CWFID作物杂草分割-4576","10.3390\u002Fjimaging12100495",{"doi":76,"openalex_id":78,"authors":79,"venue":10,"cited_by_count":89,"oa_url":6,"card":90,"direction":96,"ingested_from":97},"W7221103527",[80,83,86],{"name":81,"orcid":82},"Flora Amato","https:\u002F\u002Forcid.org\u002F0000-0002-5128-5558",{"name":84,"orcid":85},"Rajib Chandra Ghosh","https:\u002F\u002Forcid.org\u002F0009-0001-1137-3465",{"name":87,"orcid":88},"Semanto Mondal","https:\u002F\u002Forcid.org\u002F0009-0000-2306-4478",0,{"tldr":91,"method":92,"finding":93,"direction":94,"opportunity":95},"系统评估六种语义分割模型在农业IoT上的压缩精度-效率权衡，重点分析剪枝策略。","在CWFID数据集上对比CNN与Transformer模型，采用非结构化与结构化","MobileNetV3-Large U-Net精度接近SegFormer-B2且参数少3.7倍；结构","农业人工智能与决策模型","需开发类别感知的压缩评估与自适应剪枝方法，以在资源受限下保持作物分割精度。","智慧农业 \u002F 农业物联网","openalex","2026-10-10T23:30:15.016722Z",{"total":100,"page":22,"page_size":100,"items":101},6,[102,189,257,356,415,470],{"id":103,"title":104,"url":105,"summary":106,"summary_zh":107,"content":9,"source_name":108,"source_url":105,"published_at":109,"category":12,"cover_url":9,"hotness":110,"is_selected":14,"score":111,"score_detail":112,"sources":116,"tags":120,"primary_topic":123,"secondary_topics":129,"classification":131,"search_phrases":166,"slug":169,"view_count":89,"doi":170,"paper":171,"created_at":188},5474,"Real-World Impact: Transforming Agriculture with Technology A Multi-Stakeholder Review of Smart Monitoring, Precision Irrigation, and Digital Value-Chain Integration","https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23510500","There's a deliberate and significant transition from input to information. This article combines peer-reviewed evidence and public-sector data on six interrelated technology clusters that are fundamentally changing on-farm and off-farm decision making artificial intelligence (AI) based Farm Management Platforms, Internet of Things (IoT) smart irrigation, aerial and satellite crop monitoring, blockchain-enabled supply-chain digitisation, sustainability and carbon-tracking instruments, and farm data analytics dashboards. In the literature examined, precision tools have been correlated with 15% to 30% gains in crop yields, 30% to 55% of water saved during irrigation, and 10% to 20% of operational costs reduced, all with a range of variability based on the crop, the geographical region, and the size of the farm. It examines the mechanisms of these effects on the one hand and the differential rates of benefit, on the other, between large commercial farms and smallholders, and the policy and infrastructural conditions that drive actual adoption. It also highlights areas of current evidence which are lacking, such as limited long horizon field trials, poor interoperability standards, and lack of social outcome data. The article ends with a discussion on how to turn pilots into lasting, system-level change in food production, both from the government and enterprise side, as well as from the perspective of small and medium producers.","从投入要素到信息的转变是有意为之且意义重大的。本文结合经同行评审的证据与公共部门数据，聚焦六个相互关联、正在从根本上改变农场内外决策方式的技术集群：基于人工智能（AI）的农场管理平台、物联网（IoT）智能灌溉、航空与卫星作物监测、区块链赋能的供应链数字化、可持续性与碳追踪工具，以及农场数据分析仪表盘。在所考察的文献中，精准工具与作物产量提升15%至30%、灌溉用水节约30%至55%、运营成本降低10%至20%相关联，且这些效果因作物、地理区域和农场规模而存在不同程度的差异。本文一方面考察这些效应的作用机制，另一方面分析大型商业农场与小农户之间受益程度的差异，以及驱动实际采用的政策与基础设施条件。文章还指出了当前证据的薄弱之处，如缺乏长期田间试验、互操作性标准不完善以及社会成效数据缺失。文章最后讨论了如何从政府与企业两侧，以及中小生产者视角出发，将试点项目转化为粮食生产中持久、系统层面的变革。","Zenodo (CERN European Organization for Nuclear Research)","2026-11-25T00:00:00Z",25,73,{"impact":18,"substance":113,"depth":114,"authority":20,"freshness":89,"relevant":22,"comment":115},21,17,"多利益相关方综述整合六类技术集群的量化效益与采纳差异，信息增量足、产业参考价值高，但发布于一个月后，时效性不足，暂不宜进入每日精选。",[117,118],{"name":108,"url":105},{"name":108,"url":119},"https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23527988",[27,28,29,121,122],"区块链溯源","精准灌溉",{"code":124,"slug":125,"name":126,"path":127,"parent_code":128},"H3","irrigation","水资源与灌溉","\u002Ftopic\u002Fenvironment\u002Firrigation","H",[130],{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},{"status":40,"locked":14,"version":41,"method":42,"evidence":132},{"candidates":133,"agricultural_evidence":47},[134,139,145,150,157,162],{"code":124,"score":46,"supported":47,"title":135,"abstract":136,"tags":138},[125],[137,125],"灌溉",[122],{"code":33,"score":13,"supported":47,"title":140,"abstract":141,"tags":144},[],[142,143,49],"物联网","internet of things",[29],{"code":62,"score":56,"supported":47,"title":146,"abstract":147,"tags":149},[],[148],"artificial intelligence",[28],{"code":151,"score":152,"supported":47,"title":153,"abstract":154,"tags":156},"G2",3,[],[155],"小农户",[],{"code":158,"score":22,"supported":14,"title":159,"abstract":160,"tags":161},"E4",[],[],[121],{"code":68,"score":22,"supported":14,"title":163,"abstract":164,"tags":165},[],[],[27],[167,168],"Zenodo 智慧农业 多利益相关方综述","精准灌溉 物联网 区块链 供应链","Zenodo智慧农业多利益相关方综述-5474","10.5281\u002Fzenodo.23510500",{"doi":170,"openalex_id":172,"authors":173,"venue":108,"cited_by_count":89,"oa_url":105,"card":182,"direction":96,"ingested_from":97},"W7223034914",[174,177,179],{"name":175,"orcid":176},"A. Shaji George","https:\u002F\u002Forcid.org\u002F0000-0002-8677-3682",{"name":178,"orcid":9},"Dr.T.Baskar",{"name":180,"orcid":181},"P. Balaji Srikaanth","https:\u002F\u002Forcid.org\u002F0000-0003-4717-7367",{"tldr":183,"method":184,"finding":185,"direction":186,"opportunity":187},"综述六大农业数字技术集群对农场决策的变革及实际效益与采纳差异。","综合同行评议证据与公共部门数据，多利益相关方综述六类技术。","精准技术可增产15-30%、节水30-55%、降本10-20%，但小农获益不均。","数字乡村与农业信息化","缺乏长期田间试验、互操作标准与社会效益数据，小农采纳机制是研究空白。","2026-10-21T23:30:08.684798Z",{"id":190,"title":191,"url":192,"summary":193,"summary_zh":194,"content":9,"source_name":195,"source_url":192,"published_at":196,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":197,"score_detail":198,"sources":201,"tags":203,"primary_topic":206,"secondary_topics":207,"classification":212,"search_phrases":231,"slug":234,"view_count":89,"doi":235,"paper":236,"created_at":256},5376,"Agri-Synergy: A Hardware–Software Co-Designed Hybrid IoT System for Sensor Interface Standardization and LLM-Driven Interactive Decision Support","https:\u002F\u002Fdoi.org\u002F10.3390\u002Fagriengineering8100446","This paper presents Agri-Synergy, a hybrid Internet of Things (IoT) system featuring a hardware–software co-design to tackle sensor data heterogeneity and the barriers to effective AI-driven decision support in smart agriculture. Unlike software-based data harmonization, we propose a universal sensor printed circuit board (PCB) with RS485\u002FModbus interface for hardware unification, complemented by on-node edge processing that ensures standardized data from source. To lower the entry barrier for farmers, the system integrates the DeepSeek large language model (LLM) API, combining static agricultural knowledge with real-time sensor data, for context-aware, natural language-based decision support, functioning without local model deployment. Field deployment showed reference-verified sensor accuracy predominantly within manufacturer specifications, stable 4G connectivity during the 4-h benchmark but intermittent dropouts during the 20-day deployment, and remote-control success rates of 96% (normal) and 80% (weak signal at 300 m). Multi-node concurrent transmission resulted in average latency of 238.0 ± 27.1 ms and packet loss below 0.5%. The system achieves a 3–5× cost reduction compared to commercial platforms. These results demonstrate that addressing data heterogeneity from source through hardware unification and software edge processing, while leveraging AI to lower the barrier to system adoption, provides a more reliable foundation for data synchronization and accessible AI-driven analytics.","本文提出Agri-Synergy，一种采用软硬件协同设计的混合物联网（IoT）系统，旨在解决智慧农业中传感器数据异构性问题以及有效AI驱动决策支持的障碍。与基于软件的数据协调方式不同，我们提出了一种带有RS485\u002FModbus接口的通用传感器印刷电路板（PCB）以实现硬件统一，并辅以节点端边缘处理，确保从源头获得标准化数据。为降低农民的使用门槛，该系统集成了DeepSeek大语言模型（LLM）API，将静态农业知识与实时传感器数据相结合，提供上下文感知的、基于自然语言的决策支持，且无需本地部署模型即可运行。现场部署表明，经参考验证的传感器精度基本处于制造商规格范围内；在4小时基准测试期间4G连接稳定，但在20天部署期间出现间歇性断连；远程控制成功率分别为96%（正常信号）和80%（300 m处弱信号）。多节点并发传输的平均延迟为238.0 ± 27.1 ms，丢包率低于0.5%。与商业平台相比，该系统实现了3–5倍的成本降低。这些结果表明，通过硬件统一和软件边缘处理从源头解决数据异构性问题，同时利用AI降低系统采用门槛，可为数据同步和可及的AI驱动分析提供更可靠的基础。","AgriEngineering","2026-10-19T00:00:00Z",76,{"impact":17,"substance":113,"depth":19,"authority":199,"freshness":21,"relevant":22,"comment":200},12,"硬件统一接口与边缘处理结合大模型决策支持的农业物联网系统论文，方法新颖、有实地部署数据，但属细分技术进展，公共影响有限。",[202],{"name":195,"url":192},[27,28,29,204,205],"大语言模型","传感器标准化",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[208],{"code":62,"slug":209,"name":210,"path":211,"parent_code":37},"agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai",{"status":40,"locked":14,"version":41,"method":42,"evidence":213},{"candidates":214,"agricultural_evidence":47},[215,219,225],{"code":33,"score":19,"supported":47,"title":216,"abstract":217,"tags":218},[49],[142,143,49],[29],{"code":62,"score":17,"supported":47,"title":220,"abstract":222,"tags":224},[221],"decision support",[223,221],"large language model",[28,204],{"code":68,"score":226,"supported":47,"title":227,"abstract":228,"tags":230},7,[],[27,229],"smart agriculture",[27],[232,233],"Agri-Synergy 物联网 传感器","DeepSeek 决策支持","Agri-Synergy物联网传感器-5376","10.3390\u002Fagriengineering8100446",{"doi":235,"openalex_id":237,"authors":238,"venue":195,"cited_by_count":89,"oa_url":192,"card":251,"direction":96,"ingested_from":97},"W7222920424",[239,241,243,245,247,249],{"name":240,"orcid":9},"Leqi Lai",{"name":242,"orcid":9},"Langtao Duan",{"name":244,"orcid":9},"Yuhao Chen",{"name":246,"orcid":9},"Yuhang Zheng",{"name":248,"orcid":9},"Jinyu Yang",{"name":250,"orcid":9},"Mingyao Wen",{"tldr":252,"method":253,"finding":254,"direction":96,"opportunity":255},"提出软硬件协同的混合物联网系统，统一传感器接口并用大模型提供交互式决策支持。","通用RS485\u002FModbus传感器PCB、节点边缘处理、DeepSeek大模型A","传感器精度达标，多节点延迟238ms、丢包低于0.5%，成本降低3-5倍。","可研究弱信号下通信可靠性提升与边缘-云端LLM协同的长期田间鲁棒性。","2026-10-20T23:30:08.319698Z",{"id":258,"title":259,"url":260,"summary":261,"summary_zh":262,"content":9,"source_name":263,"source_url":260,"published_at":264,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":265,"score_detail":266,"sources":269,"tags":271,"primary_topic":274,"secondary_topics":280,"classification":287,"search_phrases":333,"slug":336,"view_count":89,"doi":9,"paper":337,"created_at":355},5313,"Real-Time IoT-Integrated Ensemble LSTM–Bi-LSTM Framework for Greenhouse Microclimate Forecasting","https:\u002F\u002Fwasrpublication.com\u002Findex.php\u002Findex\u002Farticle\u002Fview\u002F512","Accurate prediction of greenhouse microclimates is essential for optimizing crop growth and resource utilization in precision agriculture. The research presents an IoT-integrated ensemble deep learning framework for real-time greenhouse microclimate forecasting using Long Short-Term Memory (LSTM) and Bidirectional Long Short-Term Memory (Bi-LSTM) networks. Environmental variables, including temperature, light intensity, soil moisture, and humidity, were acquired through IoT sensor nodes deployed in a controlled greenhouse environment. A benchmark greenhouse dataset was first used for model development and hyperparameter optimization, while an original IoT-based dataset collected during cucumber, tomato, and pepper cultivation at the Federal University Wukari, Nigeria, was used for validation and real-world deployment. The data were preprocessed through noise filtering, missing value imputation, Z-score-based outlier removal, and Min–Max normalization. A weighted ensemble strategy was then applied to combine LSTM and Bi-LSTM predictions for improved robustness. Experimental results show that the Bi-LSTM model outperformed the LSTM model, achieving R² = 0.9981, MAE = 0.0088, RMSE = 0.0110, and MSE = 0.0001, compared to LSTM results of R² = 0.9956, MAE = 0.0133, RMSE = 0.0167, and MSE = 0.0003. Statistical validation using paired t-test, Wilcoxon signed-rank test, confidence intervals, and multi-run analysis confirmed that the performance improvement is statistically significant and robust. The proposed framework enables accurate real-time microclimate forecasting and intelligent environmental monitoring, providing a scalable solution for precision greenhouse management, improved crop productivity, optimized energy utilization, and efficient irrigation scheduling.","准确预测温室微气候对于精准农业中优化作物生长和资源利用至关重要。本研究提出了一种集成物联网的集成深度学习框架，利用长短期记忆网络（LSTM）和双向长短期记忆网络（Bi-LSTM）进行实时温室微气候预测。温度、光照强度、土壤湿度和湿度等环境变量通过部署在受控温室环境中的物联网传感器节点采集。研究首先使用基准温室数据集进行模型开发和超参数优化，随后使用在尼日利亚武卡里联邦大学黄瓜、番茄和辣椒种植期间采集的原始物联网数据集进行验证和实际部署。数据经过噪声滤波、缺失值插补、基于Z分数的异常值去除和最小—最大归一化预处理。然后采用加权集成策略融合LSTM和Bi-LSTM的预测结果以提高鲁棒性。实验结果表明，Bi-LSTM模型优于LSTM模型，其R² = 0.9981、MAE = 0.0088、RMSE = 0.0110、MSE = 0.0001，而LSTM的结果为R² = 0.9956、MAE = 0.0133、RMSE = 0.0167、MSE = 0.0003。采用配对t检验、Wilcoxon符号秩检验、置信区间和多轮运行分析进行的统计验证证实，性能提升具有统计显著性和鲁棒性。所提出的框架能够实现准确的实时微气候预测和智能环境监测，为精准温室管理、提高作物生产力、优化能源利用和高效灌溉调度提供了可扩展的解决方案。","World Journal of Multidisciplinary Studies","2026-10-16T00:00:00Z",67,{"impact":199,"substance":113,"depth":114,"authority":21,"freshness":267,"relevant":22,"comment":268},8,"方法扎实、验证充分，但属单点应用研究，公共影响有限，适合作为技术前沿条目而非头条精选。",[270],{"name":263,"url":260},[27,28,29,272,273],"温室环境","小气候预测",{"code":275,"slug":276,"name":277,"path":278,"parent_code":279},"B7","protected-horticulture","设施园艺与植物工厂","\u002Ftopic\u002Fhorticulture\u002Fprotected-horticulture","B",[281,282],{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},{"code":283,"slug":284,"name":285,"path":286,"parent_code":279},"B1","vegetables","蔬菜","\u002Ftopic\u002Fhorticulture\u002Fvegetables",{"status":40,"locked":14,"version":41,"method":42,"evidence":288},{"candidates":289,"agricultural_evidence":47},[290,294,301,310,316,322,329],{"code":33,"score":46,"supported":47,"title":291,"abstract":292,"tags":293},[49],[142,49],[29],{"code":275,"score":295,"supported":47,"title":296,"abstract":298,"tags":300},14,[297],"greenhouse",[299,297],"温室",[],{"code":283,"score":21,"supported":47,"title":302,"abstract":303,"tags":309},[],[304,305,306,307,308],"番茄","辣椒","黄瓜","tomato","cucumber",[],{"code":68,"score":226,"supported":47,"title":311,"abstract":312,"tags":315},[],[313,314],"精准农业","precision agriculture",[27],{"code":62,"score":226,"supported":47,"title":317,"abstract":318,"tags":321},[],[319,320],"深度学习","deep learning",[28],{"code":323,"score":100,"supported":47,"title":324,"abstract":325,"tags":328},"H2",[],[326,327],"土壤","soil",[],{"code":124,"score":100,"supported":47,"title":330,"abstract":331,"tags":332},[],[137,125],[],[334,335],"温室 小气候 LSTM 预测","尼日利亚 温室 IoT 黄瓜 番茄","温室小气候LSTM预测-5313",{"doi":9,"openalex_id":338,"authors":339,"venue":263,"cited_by_count":89,"oa_url":9,"card":350,"direction":96,"ingested_from":97},"W7222909837",[340,342,344,346,348],{"name":341,"orcid":9},"Siman Emmanuel*",{"name":343,"orcid":9},"Edward Onyebueke Agu",{"name":345,"orcid":9},"Musa Gani",{"name":347,"orcid":9},"Oladunjoye John Abiodun",{"name":349,"orcid":9},"Anagu Emmanuel John",{"tldr":351,"method":352,"finding":353,"direction":96,"opportunity":354},"提出物联网集成LSTM与Bi-LSTM加权集成框架，实时预测温室微气候。","IoT传感器采集温光湿数据，经预处理后用LSTM-Bi-LSTM加权集成建模。","Bi-LSTM优于LSTM，R²达0.9981，统计检验证实性能提升显著稳健。","可探索边缘端轻量化部署与多作物跨温室迁移学习，提升实时预测泛化性。","2026-10-19T23:30:13.295236Z",{"id":357,"title":358,"url":359,"summary":360,"summary_zh":361,"content":9,"source_name":362,"source_url":359,"published_at":363,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":364,"score_detail":365,"sources":367,"tags":369,"primary_topic":372,"secondary_topics":373,"classification":375,"search_phrases":397,"slug":400,"view_count":152,"doi":401,"paper":402,"created_at":414},4809,"Enhanced crop yield estimation in smart agriculture via AI and IoT using the GODL geo-optimized algorithm","https:\u002F\u002Fdoi.org\u002F10.1007\u002Fs12083-026-02299-3","Abstract Precision agriculture is one of the major breakthroughs which has been adopted to counteract some of the major challenges facing contemporary agriculture, such as maximizing crop yield, maximizing resource efficiency, and sustainability. The application of Artificial Intelligence (AI) and the Internet of Things (IoT) technologies can go a long way in enhancing crop yield prediction through data obtained from multiple sources, including satellite images, ground sensors, and weather stations. This research paper proposes a next-generation crop yield prediction system called Geo-Optimized Deep Learning (GODL) algorithm. The GODL algorithm integrates geospatial information with deep learning algorithms for crop yield analysis and prediction. This combination of deep learning and geospatial data allows accurate and scalable crop yield predictions using geospatial and sensor information. In order to ensure model optimization in terms of accuracy and efficiency, the GODL Algorithm makes use of advanced hyperparameter optimization methods that include Adam and Stochastic Gradient Descent (SGD). The research paper highlights the efficiency of the GODL Algorithm by conducting extensive experiments that compare the algorithm’s performance with other crop yield prediction models like Recurrent Neural Network (RNN) and other machine learning techniques. The evaluation takes into consideration important metrics such as accuracy, Jaccard Index, Fowlkes-Mallows Index (FMI), and G-mean, as well as IoT data delay and its effects on the prediction accuracy. In future research, the use of blockchain technology to facilitate data sharing and edge computing to process the data will be considered.","摘要 精准农业是为应对当代农业面临的若干重大挑战而采用的主要突破之一，这些挑战包括最大化作物产量、最大化资源效率以及可持续性。人工智能（AI）与物联网（IoT）技术的应用，能够通过从卫星图像、地面传感器和气象站等多源获取的数据，大幅提升作物产量预测能力。本文提出了一种下一代作物产量预测系统，称为地理优化深度学习（GODL）算法。GODL算法将地理空间信息与深度学习算法相结合，用于作物产量分析与预测。深度学习与地理空间数据的这种结合，使得利用地理空间和传感器信息进行准确且可扩展的作物产量预测成为可能。为确保模型在准确性和效率方面实现优化，GODL算法采用了先进的超参数优化方法，包括Adam和随机梯度下降（SGD）。本文通过开展大量实验，将该算法的性能与其他作物产量预测模型（如循环神经网络（RNN）及其他机器学习技术）进行比较，从而突出GODL算法的效率。评估考虑了准确率、Jaccard指数、Fowlkes-Mallows指数（FMI）和G均值等重要指标，以及物联网数据延迟及其对预测精度的影响。在未来的研究中，将考虑使用区块链技术促进数据共享，并利用边缘计算处理数据。","Peer-to-Peer Networking and Applications","2026-10-12T00:00:00Z",68,{"impact":199,"substance":19,"depth":17,"authority":20,"freshness":21,"relevant":22,"comment":366},"提出GODL地理优化深度学习算法融合AI与IoT进行作物估产，方法有创新且对比实验充分，但属学术论文，产业落地影响有限。",[368],{"name":362,"url":359},[27,28,29,370,371],"遥感监测","作物估产",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[374],{"code":62,"slug":209,"name":210,"path":211,"parent_code":37},{"status":40,"locked":14,"version":41,"method":42,"evidence":376},{"candidates":377,"agricultural_evidence":47},[378,382,386,392],{"code":33,"score":19,"supported":47,"title":379,"abstract":380,"tags":381},[49],[142,51,143,49,52],[29],{"code":68,"score":46,"supported":47,"title":383,"abstract":384,"tags":385},[229],[313,314],[27],{"code":62,"score":13,"supported":47,"title":387,"abstract":388,"tags":391},[],[319,389,390,320,148],"机器学习","machine learning",[28],{"code":393,"score":63,"supported":14,"title":394,"abstract":395,"tags":396},"I3",[],[],[371,370],[398,399],"GODL 作物估产 算法","AI IoT 智慧农业 产量预测","GODL作物估产算法-4809","10.1007\u002Fs12083-026-02299-3",{"doi":401,"openalex_id":403,"authors":404,"venue":362,"cited_by_count":89,"oa_url":359,"card":409,"direction":96,"ingested_from":97},"W7221560555",[405,407],{"name":406,"orcid":9},"Keerthana T",{"name":408,"orcid":9},"Janarthanan R",{"tldr":410,"method":411,"finding":412,"direction":96,"opportunity":413},"提出GODL算法，融合地理空间信息与深度学习，结合AI和IoT提升作物产量预测精度。","GODL算法，融合地理空间数据与深度学习，采用Adam和SGD超参数优化。","GODL在准确率、Jaccard指数、FMI和G-mean上优于RNN等模型，且IoT数据延迟影响预","可探索区块链保障数据共享安全、边缘计算降低IoT延迟，以及多源异构数据融合的轻量化模型。","2026-10-13T23:30:10.304344Z",{"id":416,"title":417,"url":418,"summary":419,"summary_zh":420,"content":9,"source_name":421,"source_url":418,"published_at":11,"category":12,"cover_url":9,"hotness":110,"is_selected":14,"score":422,"score_detail":423,"sources":425,"tags":429,"primary_topic":431,"secondary_topics":432,"classification":434,"search_phrases":449,"slug":452,"view_count":22,"doi":9,"paper":453,"created_at":469},4672,"AI-BASED ENERGY EFFICIENT COMMUNICATION IN IOT SYSTEM","https:\u002F\u002Fijmert.com\u002Findex.php\u002Fijmert\u002Farticle\u002Fview\u002F309","The rapid growth of the Internet of Things (IoT) has led to a massive increase in the number of connected devices across various domains such as smart homes, healthcare, agriculture, and industrial automation. As these devices are often battery-powered and deployed in remote or large scale environments, energy efficiency has become a critical concern in IoT communication systems. Traditional IoT networks typically operate using fixed transmission power levels, which do not adapt to changing environmental or network conditions, resulting in unnecessary energy consumption and reduced battery life.To address these limitations, there is a growing need for intelligent and adaptive communication mechanisms that can dynamically optimize power usage without compromising reliability. In this context, Artificial Intelligence (AI)-based adaptive transmission power control has emerged as a promising solution. This review paper presents an analysis of AI-driven techniques that enable IoT devices to adjust their transmission power based on real-time conditions and network requirements.The study explores approaches such as machine learning-based prediction models, which utilize parameters like Received Signal Strength Indicator (RSSI), distance, interference levels, and packet loss rates to determine optimal transmission power. Additionally, the role of edge computing is examined, enabling faster decision-making with reduced latency. The integration of Low Power Wide Area Network (LPWAN) technologies is also discussed for supporting long-range and energy-efficient communication.Furthermore, the paper highlights how adaptive power control improves network performance by reducing energy wastage, minimizing interference, and extending device lifetime. These benefits are especially useful in applications like smart cities, precision agriculture, and industrial automation.Despite these advantages, challenges such as implementation complexity, resource constraints, and data privacy concerns remain. The paper also outlines future research directions, including the development of lightweight AI models and improved integration with emerging technologies.In conclusion, AIbased adaptive transmission power control significantly enhances energy efficiency and communication reliability, making it essential for the sustainable development of next- generation IoT systems. It also supports scalable and cost-effective deployment of IoT networks in diverse real- world environments.","物联网（IoT）的快速发展使得智能家居、医疗健康、农业和工业自动化等各个领域的联网设备数量大幅增加。由于这些设备通常由电池供电，且部署在偏远或大规模环境中，能效已成为物联网通信系统中的关键问题。传统物联网网络通常采用固定传输功率运行，无法适应不断变化的环境或网络条件，导致不必要的能耗和电池寿命缩短。为解决这些局限，人们日益需要能够在不影响可靠性的前提下动态优化功率使用的智能自适应通信机制。在此背景下，基于人工智能（AI）的自适应传输功率控制已成为一种有前景的解决方案。本综述论文分析了使物联网设备能够根据实时条件和网络需求调整传输功率的AI驱动技术。研究探讨了基于机器学习的预测模型等方法，这些方法利用接收信号强度指示（RSSI）、距离、干扰水平和丢包率等参数来确定最优传输功率。此外，还考察了边缘计算的作用，其能够以更低延迟实现更快决策。论文还讨论了低功耗广域网（LPWAN）技术的集成，以支持长距离和节能通信。进一步地，本文强调了自适应功率控制如何通过减少能量浪费、最小化干扰和延长设备寿命来改善网络性能。这些优势在智慧城市、精准农业和工业自动化等应用中尤为有用。尽管具有这些优势，实施复杂性、资源约束和数据隐私问题等挑战仍然存在。论文还概述了未来研究方向，包括开发轻量级AI模型以及与新兴技术的改进集成。总之，基于AI的自适应传输功率控制显著提升了能效和通信可靠性，使其成为下一代物联网系统可持续发展的关键。它还支持物联网网络在多样化现实环境中的可扩展且具有成本效益的部署。","International Journal of Mechanical Engineering Research and Technology",47,{"impact":267,"substance":295,"depth":20,"authority":100,"freshness":100,"relevant":22,"comment":424},"综述性论文，方法层面有参考价值，但非农业专属突破，且期刊权威性一般，仅作主题聚合素材。",[426,427],{"name":421,"url":418},{"name":421,"url":428},"https:\u002F\u002Fijmert.com\u002Findex.php\u002Fijmert\u002Farticle\u002Fview\u002F310",[27,28,29,313,430],"低功耗通信",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[433],{"code":62,"slug":209,"name":210,"path":211,"parent_code":37},{"status":40,"locked":14,"version":41,"method":42,"evidence":435},{"candidates":436,"agricultural_evidence":47},[437,441,445],{"code":33,"score":19,"supported":47,"title":438,"abstract":439,"tags":440},[49],[142,51,143,49,52],[29],{"code":62,"score":13,"supported":47,"title":442,"abstract":443,"tags":444},[],[389,390,148],[28],{"code":68,"score":267,"supported":47,"title":446,"abstract":447,"tags":448},[],[313,314],[27,313],[450,451],"AI 物联网 传输功率控制","LPWAN 精准农业","AI物联网传输功率控制-4672",{"doi":9,"openalex_id":454,"authors":455,"venue":421,"cited_by_count":89,"oa_url":418,"card":464,"direction":96,"ingested_from":97},"W7221414266",[456,458,460,462],{"name":457,"orcid":9},"Shital Chavan",{"name":459,"orcid":9},"Anam Bedrekar",{"name":461,"orcid":9},"Sanika Babar",{"name":463,"orcid":9},"Prof. M. V. Naiknavare",{"tldr":465,"method":466,"finding":467,"direction":96,"opportunity":468},"综述AI自适应发射功率控制提升IoT通信能效，面向智慧农业等场景。","综述机器学习预测模型、边缘计算与LPWAN技术，基于RSSI等参数。","AI自适应功率控制可降低能耗、减少干扰并延长设备寿命，但存在实现复杂与隐私挑战。","可研究面向农田的轻量级AI功率控制模型，结合土壤墒情与作物需水优化能耗。","2026-10-11T23:30:16.397155Z",{"id":471,"title":472,"url":473,"summary":474,"summary_zh":475,"content":9,"source_name":421,"source_url":473,"published_at":11,"category":12,"cover_url":9,"hotness":110,"is_selected":14,"score":476,"score_detail":477,"sources":479,"tags":483,"primary_topic":484,"secondary_topics":485,"classification":487,"search_phrases":512,"slug":515,"view_count":22,"doi":9,"paper":516,"created_at":533},4671,"A COMPREHENSIVE REVIEW ON IOT AND EDGE AI-BASED PRECISION AGRICULTURE SYSTEM FOR WATER OPTIMIZATION","https:\u002F\u002Fijmert.com\u002Findex.php\u002Fijmert\u002Farticle\u002Fview\u002F316","Water scarcity and inefficient irrigation practices remain major challenges in modern agriculture, leading to reduced crop yield, resource wastage, and increased production costs. The proposed project, “IoT and Edge AI-Based Precision Agriculture System for Water Optimisation,” aims to develop an intelligent, real-time irrigation management system that optimises water usage through data-driven decision-making at the edge. The system integrates Internet of Things (IoT) sensors such as soil moisture, temperature, humidity, and light intensity sensors to continuously monitor field conditions. The collected data is processed locally using an Edge AI module, enabling real-time analysis and predictive irrigation control without relying entirely on cloud connectivity. By deploying machine learning algorithms at the edge, the system can determine optimal watering schedules based on soil conditions, crop requirements, and environmental factors, thereby minimising water wastage and ensuring efficient irrigation. A cloud platform is incorporated for long-term data storage, visualisation, and remote monitoring via a web or mobile dashboard. This hybrid architecture enhances scalability, reduces latency, and improves reliability in rural areas with limited internet connectivity. Automated irrigation control using smart valves or pumps ensures precise water delivery tailored to specific crop zones.","水资源短缺和灌溉效率低下仍是现代农业面临的主要挑战，导致作物减产、资源浪费和生产成本增加。本项目“基于物联网与边缘人工智能的水资源优化精准农业系统”旨在开发一种智能实时灌溉管理系统，通过边缘端的数据驱动决策来优化水资源利用。该系统集成了物联网（IoT）传感器，如土壤湿度、温度、湿度和光照强度传感器，以持续监测田间状况。采集的数据通过边缘人工智能模块在本地进行处理，实现实时分析和预测性灌溉控制，而无需完全依赖云连接。通过在边缘端部署机器学习算法，系统能够根据土壤条件、作物需求和环境因素确定最佳灌溉计划，从而最大限度地减少水资源浪费并确保高效灌溉。系统还包含一个云平台，用于长期数据存储、可视化以及通过网页或移动端仪表盘进行远程监控。这种混合架构增强了可扩展性，降低了延迟，并提高了互联网连接有限的农村地区的可靠性。利用智能阀门或水泵的自动灌溉控制可确保针对特定作物区域进行精准供水。",53,{"impact":199,"substance":295,"depth":20,"authority":100,"freshness":267,"relevant":22,"comment":478},"综述性论文，系统梳理IoT与边缘AI在水肥优化中的应用，方法框架清晰但缺乏原创数据与实证，属细分领域参考性内容。",[480,481],{"name":421,"url":473},{"name":421,"url":482},"https:\u002F\u002Fijmert.com\u002Findex.php\u002Fijmert\u002Farticle\u002Fview\u002F315",[27,28,29,51,122],{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[486],{"code":124,"slug":125,"name":126,"path":127,"parent_code":128},{"status":40,"locked":14,"version":41,"method":42,"evidence":488},{"candidates":489,"agricultural_evidence":47},[490,495,499,504,508],{"code":33,"score":491,"supported":47,"title":492,"abstract":493,"tags":494},19,[49],[142,143,49],[29,51],{"code":68,"score":46,"supported":47,"title":496,"abstract":497,"tags":498},[314],[313,314],[27],{"code":124,"score":13,"supported":47,"title":500,"abstract":501,"tags":503},[],[137,502,125],"水资源",[122],{"code":62,"score":226,"supported":47,"title":505,"abstract":506,"tags":507},[],[389,390],[28],{"code":323,"score":100,"supported":47,"title":509,"abstract":510,"tags":511},[],[326,327],[],[513,514],"IoT 边缘AI 精准灌溉","智慧灌溉 土壤墒情 传感器","IoT边缘AI精准灌溉-4671",{"doi":9,"openalex_id":517,"authors":518,"venue":421,"cited_by_count":89,"oa_url":473,"card":528,"direction":96,"ingested_from":97},"W7221411136",[519,521,523,525],{"name":520,"orcid":9},"Om Gaikwad",{"name":522,"orcid":9},"Shriraj Chavan",{"name":524,"orcid":9},"Krushnadev Gaikwad",{"name":526,"orcid":527},"Swapnil R. Takale","https:\u002F\u002Forcid.org\u002F0009-0001-3636-4450",{"tldr":529,"method":530,"finding":531,"direction":96,"opportunity":532},"综述基于物联网与边缘AI的精准农业系统，实现实时灌溉优化与节水。","集成土壤湿度等IoT传感器与边缘AI，本地处理数据并云端存储可视化。","边缘AI可实时预测灌溉需求，减少水浪费并适应网络受限农村环境。","可探索边缘AI模型轻量化与多源数据融合，提升复杂农田环境下的灌溉决策精度。","2026-10-11T23:30:16.298334Z"]