[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-2418":3,"related-2418":90},{"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":18,"tags":20,"primary_topic":9,"secondary_topics":26,"classification":27,"search_phrases":63,"slug":66,"view_count":58,"doi":67,"paper":68,"created_at":89},2418,"Artificial Intelligence-Driven Anti-Poaching Surveillance Systems: A Review of Advances, Challenges, and Future Directions","https:\u002F\u002Fdoi.org\u002F10.14445\u002F23488549\u002Fijece-v13i8p103","The UN Environment Programme reports an average 69% decline in wildlife population since 1970, with around 4000 species illegally poached, trafficked, and traded globally. The World Wildlife Crime Report shows a significant population revival for certain species, but overall protection remains inadequate. The 5-step approach to understanding wildlife decline can improve methods. Emerging smart technologies, such as computer vision, IoT systems, and AI, can help manage wildlife populations and prevent poaching. Drones, bio-loggers, camera traps, and GPS collars provide real-time data for wildlife monitoring and tracking. Unmanned Aerial Vehicles (UAVs) have also shown promise in wildlife welfare. This article aims to analyze the decline of wildlife populations, study smart technologies for anti-poaching efforts, and evaluate emerging technology-driven solutions to traditional wildlife welfare challenges. Human-induced factors like hunting, poaching, pollution, and habitat loss have compromised biodiversity. Poaching has led to population reduction and ecological imbalance. Traditional conservation practices have failed to revive wildlife populations. Integrating smart technologies, IoT, AI, and predictive analysis can help prevent poaching and promote ecological balance.","联合国环境规划署报告称，自1970年以来，野生动物种群平均减少了69%，全球约有4000个物种遭到非法偷猎、贩运和交易。《世界野生动物犯罪报告》显示，某些物种种群显著恢复，但总体保护仍然不足。理解野生动物减少的5步方法可以改进相关措施。新兴智能技术，如计算机视觉、物联网系统和人工智能，有助于管理野生动物种群并防止偷猎。无人机、生物记录器、相机陷阱和GPS项圈为野生动物监测和追踪提供实时数据。无人驾驶飞行器（UAVs）在野生动物福利方面也展现出前景。本文旨在分析野生动物种群的减少，研究反偷猎的智能技术，并评估新兴技术驱动的解决方案以应对传统野生动物福利挑战。狩猎、偷猎、污染和栖息地丧失等人为因素破坏了生物多样性。偷猎导致种群减少和生态失衡。传统保护实践未能恢复野生动物种群。整合智能技术、物联网、人工智能和预测分析有助于防止偷猎并促进生态平衡。",null,"International Journal of Electronics and Communication Engineering","2026-09-11T00:00:00Z","论文",10,false,0,{"impact":15,"substance":15,"depth":15,"authority":15,"freshness":15,"relevant":15,"comment":17},"主题为野生动物反盗猎智能监测，属生态保护领域，与三农、农业信息化、智慧农业无直接关联，不建议进入每日精选。",[19],{"name":10,"url":6},[21,22,23,24,25],"计算机视觉","物联网","无人机遥感","野生动物保护","反盗猎监测",[],{"status":28,"locked":14,"version":29,"method":30,"evidence":31},"needs_review","2026-10-22.2","rules",{"candidates":32,"agricultural_evidence":14,"reason":62},[33,42,50,56],{"code":34,"score":35,"supported":36,"title":37,"abstract":39,"tags":41},"I2",15,true,[38],"artificial intelligence",[21,40],"computer vision",[21],{"code":43,"score":44,"supported":36,"title":45,"abstract":46,"tags":49},"H6",7,[],[47,48],"生物多样性","biodiversity",[24],{"code":51,"score":44,"supported":36,"title":52,"abstract":53,"tags":55},"I5",[],[22,54],"IoT",[22],{"code":57,"score":58,"supported":14,"title":59,"abstract":60,"tags":61},"I3",1,[],[],[23],"农业相关性需要复核",[64,65],"野生动物保护 反盗猎监测 无人机遥感 计算机视觉","野生动物保护 反盗猎监测","野生动物保护反盗猎监测无人机遥感计算机视觉-2418","10.14445\u002F23488549\u002Fijece-v13i8p103",{"doi":67,"openalex_id":69,"authors":70,"venue":10,"cited_by_count":15,"oa_url":81,"card":82,"direction":86,"ingested_from":88},"W7212319763",[71,73,75,77,79],{"name":72,"orcid":9},"Rohit Samkaria",{"name":74,"orcid":9},"Rajesh Singh",{"name":76,"orcid":9},"Anita Gehlot",{"name":78,"orcid":9},"Charvi Joshi",{"name":80,"orcid":9},"Rahul Mahala","https:\u002F\u002Fwww.internationaljournalssrg.org\u002F..\u002FIJECE\u002F2026\u002FVolume13-Issue8\u002FIJECE-V13I8P103.pdf",{"tldr":83,"method":84,"finding":85,"direction":86,"opportunity":87},"综述AI、物联网与无人机等智能技术在反盗猎监测中的进展、挑战与未来方向。","综述计算机视觉、IoT、无人机、相机陷阱、GPS项圈等实时监测技术。","传统保护措施失效，AI与IoT融合可有效预防盗猎并促进生态平衡。","智慧农业 \u002F 农业物联网","可探索AI预测模型与多源传感数据融合在反盗猎实时预警中的落地验证。","openalex","2026-09-14T23:30:12.295479Z",{"total":91,"page":58,"page_size":91,"items":92},6,[93,160,214,308,409,470],{"id":94,"title":95,"url":96,"summary":97,"summary_zh":98,"content":9,"source_name":99,"source_url":96,"published_at":100,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":101,"score_detail":102,"sources":108,"tags":110,"primary_topic":115,"secondary_topics":120,"classification":121,"search_phrases":139,"slug":142,"view_count":15,"doi":143,"paper":144,"created_at":159},5587,"Calibration-Shift Inference for Computer Vision Agriculture Yield Models Under Retrofit Sensor Replacement: An Artificial Intelligence Post-Deployment Recalibration Guarantee","https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23544195","Computer vision yield models deployed in agricultural fields are routinely paired with retrofit sensor packages that replace or supplement the original imaging hardware, and the resulting calibration shift can silently invalidate the model's predictive guarantees. The analysis characterises the operating characteristics of the design under a declared variance structure with between-cluster standard deviation 6.0, residual standard deviation 4.5 and intraclass correlation 0.64, using 2000 replications per design cell across cluster counts of 8, 16, 24 and 32 and effect sizes of 3, 5 and 8 percentage points. The design attains 80% power at a 5 percentage point effect with 54 clusters. The paper states a formal guarantee: under stated assumptions on the stability of the calibration map, the before-after-control-impact estimator yields a consistent and asymptotically normal estimate of the calibration shift, and the associated confidence interval covers the true shift at the nominal rate. The contribution is a design-theoretic account of how retrofit-induced calibration shift can be detected and bounded without observational data, together with the sample-size requirements that follow.","部署在农田中的计算机视觉产量模型通常与改装传感器组件配合使用，这些组件用于替换或补充原始成像硬件，由此产生的校准偏移可能悄然使模型的预测保证失效。该分析在声明的方差结构下刻画了设计的工作特性，其中组间标准差为6.0、残差标准差为4.5、组内相关系数为0.64，每个设计单元采用2000次重复，聚类数分别为8、16、24和32，效应量为3、5和8个百分点。该设计在54个聚类、5个百分点效应下达到80%的检验力。论文给出了一项形式化保证：在关于校准映射稳定性的既定假设下，前后对照影响（before-after-control-impact）估计量给出校准偏移的一致且渐近正态的估计，且相关置信区间以名义覆盖率覆盖真实偏移。其贡献在于从设计理论角度说明了如何在没有观测数据的情况下检测并界定改装引起的校准偏移，以及随之而来的样本量要求。","Zenodo (CERN European Organization for Nuclear Research)","2026-10-21T00:00:00Z",65,{"impact":103,"substance":104,"depth":105,"authority":106,"freshness":13,"relevant":58,"comment":107},8,18,17,12,"方法学扎实、统计设计严谨的预印本论文，但偏理论设计层面，缺乏田间实证数据，产业影响有限，适合作为技术参考而非每日精选头条。",[109],{"name":99,"url":96},[111,112,21,113,114],"智慧农业","农业人工智能","产量预测","传感器校准",{"code":34,"slug":116,"name":117,"path":118,"parent_code":119},"agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai","I",[],{"status":122,"locked":14,"version":29,"method":30,"evidence":123},"classified",{"candidates":124,"agricultural_evidence":36},[125,130,135],{"code":34,"score":126,"supported":36,"title":127,"abstract":128,"tags":129},24,[38,40],[21,40],[112,21],{"code":131,"score":58,"supported":14,"title":132,"abstract":133,"tags":134},"I1",[],[],[111],{"code":57,"score":58,"supported":14,"title":136,"abstract":137,"tags":138},[],[],[113],[140,141],"计算机视觉 产量预测 传感器校准","农业人工智能 传感器校准 计算机视觉 产量预测","计算机视觉产量预测传感器校准-5587","10.5281\u002Fzenodo.23544195",{"doi":143,"openalex_id":145,"authors":146,"venue":99,"cited_by_count":15,"oa_url":96,"card":153,"direction":157,"ingested_from":88},"W7223104380",[147,149,151],{"name":148,"orcid":9},"Tigist Alemu Gebremariam",{"name":150,"orcid":9},"Hawi Tesfaye Desta",{"name":152,"orcid":9},"Mulugeta Bekele Wolde",{"tldr":154,"method":155,"finding":156,"direction":157,"opportunity":158},"提出改造传感器替换后视觉产量模型校准漂移的检测与置信区间保证方法。","用前后对照影响估计量，基于方差结构做2000次重复的模拟设计分析。","5个百分点效应下54个集群可达80%检验力，估计量一致且渐近正态。","农业人工智能与决策模型","可将该设计理论扩展到多传感器异构替换与在线校准监测，填补真实田间验证空白。","2026-10-22T11:21:01.554995Z",{"id":161,"title":162,"url":163,"summary":164,"summary_zh":165,"content":9,"source_name":99,"source_url":163,"published_at":100,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":166,"score_detail":167,"sources":171,"tags":173,"primary_topic":176,"secondary_topics":177,"classification":178,"search_phrases":194,"slug":197,"view_count":15,"doi":198,"paper":199,"created_at":213},5584,"Distributionally-Robust Evaluation of Computer Vision Agriculture Disease Classifiers Under Climate-Driven Pathogen Range Shifts: An Artificial Intelligence Worst-Case Generalization Bound","https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23544150","Computer vision classifiers for crop disease are evaluated almost exclusively on held-out samples drawn from the same distribution as their training data, yet climate-driven pathogen range shifts mean that deployment environments are systematically out of distribution. The framework treats the deployment distribution as an adversarial perturbation of the training distribution within a Wasserstein ball whose radius is calibrated to observed climate displacement. Using NASA POWER monthly reanalysis for Harare, Zimbabwe (2000-2023, 312 monthly observations), the analysis estimates a mean air temperature of 18.914 C, a standard deviation of 3.107 C, and a warming trend of 0.2001 C per decade. These estimates anchor the radius of the ambiguity set and the resulting bound. The bound decomposes worst-case risk into an empirical risk term, a Wasserstein radius term scaled by the Lipschitz constant of the classifier, and a concentration term. Under plausible Lipschitz constants for convolutional backbones, the radius term dominates the empirical risk term once the climate displacement exceeds a threshold that the Harare record already approaches. The contribution is a formal evaluation criterion that replaces the i. i. d. assumption with a climate-calibrated ambiguity set, together with the analytical machinery needed to compute it.","作物病害的计算机视觉分类器几乎完全在与训练数据同分布的留出样本上进行评估，然而气候驱动的病原体分布范围变化意味着部署环境系统性地处于分布之外。该框架将部署分布视为训练分布在Wasserstein球内的对抗性扰动，球的半径根据观测到的气候位移进行校准。利用NASA POWER对津巴布韦哈拉雷的月度再分析数据（2000—2023年，312个月度观测值），分析估计平均气温为18.914°C，标准差为3.107°C，每十年升温趋势为0.2001°C。这些估计值锚定了模糊集的半径及由此得到的界。该界将最坏情况风险分解为经验风险项、由分类器Lipschitz常数缩放后的Wasserstein半径项以及集中项。在卷积骨干网络合理的Lipschitz常数下，一旦气候位移超过哈拉雷记录已接近的阈值，半径项便主导经验风险项。本文的贡献在于提出了一种形式化评估准则，以气候校准的模糊集替代独立同分布（i.i.d.）假设，并提供了计算该准则所需的分析工具。",75,{"impact":106,"substance":168,"depth":104,"authority":169,"freshness":13,"relevant":58,"comment":170},22,13,"方法新颖、数据具体，但属方法论预印论文，产业影响有限，适合作为专业精选而非头条。",[172],{"name":99,"url":163},[111,112,21,174,175],"作物病害","气候变化",{"code":34,"slug":116,"name":117,"path":118,"parent_code":119},[],{"status":122,"locked":14,"version":29,"method":30,"evidence":179},{"candidates":180,"agricultural_evidence":36},[181,185,190],{"code":34,"score":126,"supported":36,"title":182,"abstract":183,"tags":184},[38,40],[21,40],[112,21],{"code":186,"score":58,"supported":14,"title":187,"abstract":188,"tags":189},"H7",[],[],[175],{"code":131,"score":58,"supported":14,"title":191,"abstract":192,"tags":193},[],[],[111],[195,196],"NASA POWER 哈拉雷 气温","Wasserstein 分布鲁棒 病害分类","NASAPOWER哈拉雷气温-5584","10.5281\u002Fzenodo.23544150",{"doi":198,"openalex_id":200,"authors":201,"venue":99,"cited_by_count":15,"oa_url":163,"card":208,"direction":157,"ingested_from":88},"W7223087892",[202,204,206],{"name":203,"orcid":9},"Kossi Afeli",{"name":205,"orcid":9},"Akouvi Amegashie",{"name":207,"orcid":9},"Komlan Sélom Dossou",{"tldr":209,"method":210,"finding":211,"direction":157,"opportunity":212},"提出气候校准的分布鲁棒评估准则，替代i.i.d.假设评价作物病害视觉分类器。","用Wasserstein球建模部署分布偏移，以NASA POWER哈拉雷气温数据","气候位移超阈值后，半径项主导经验风险，现有评估会高估模型泛化能力。","可将气候校准的鲁棒评估扩展到多作物多区域，并研究适应分布偏移的鲁棒训练方法。","2026-10-22T11:21:00.036937Z",{"id":215,"title":216,"url":217,"summary":218,"summary_zh":219,"content":9,"source_name":220,"source_url":217,"published_at":100,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":166,"score_detail":221,"sources":223,"tags":225,"primary_topic":228,"secondary_topics":232,"classification":233,"search_phrases":294,"slug":297,"view_count":15,"doi":298,"paper":299,"created_at":307},5556,"AI Powered IoT-Based System for Weed and Pest Control in Agriculture Using Deep Learning","https:\u002F\u002Fdoi.org\u002F10.25258\u002Fijddt.16.76s.32","Weeds, pest attacks, labour shortage and over-dependence on chemical pesticides are the serious challenges faced by the agricultural sector in the wake of ever growing weed infestation.Manual weed and pest control has potential negative environmental impacts, increased production costs and reduced yields.The application of Artificial Intelligence (AI), Internet of Things (IoT), Deep Learning, robotics, and laser technologies are a promising way of developing sustainable and autonomous agricultural systems.This study introduces an intelligent agricultural system based on Artificial Intelligence (AI) and Internet of Things (IoT) for weed identification, harmful insect identification, crop condition monitoring and automatic weed removal using a laser beam.The system will be based on Internet of Things (IoT) sensors, Computer vision modules, Convolutional neural networks (CNN), Unmanned robotic platforms, Environmental sensing systems, and Electronic Swarm Suppression (ESS) mechanisms to enhance agricultural productivity and reduce crop losses.The system uses deep learning algorithms for real time weed, crop, beneficial and harmful insect classification.High-resolution cameras and sensing devices collect environmental and crop data continuously and feed the data into an IoT system for processing in a cloud-edge computing system.High precision weed elimination with a robotic platform containing laser emitters which do not harm adjacent crops.Further, computer vision and environmental monitoring are employed to identify aerial insect swarms and to enable automatic attacks of insect swarms by means of a laser beam and ESS technology.Multiple agricultural blocks were used for experimental evaluation of the maize, wheat, soybean and vegetable crops.The results show substantial increases in the accuracy of weed detection, pest identification, crop protection and cost savings for operations over current practices.The proposed system achieved a 97.6% accuracy in the weed classification, 96.8% accuracy in the detection of the insect and a 42.3% reduction in pesticide use.In addition, the crop yield rose by 18.7% and the operational costs dropped by 28.4%.Overall, the results suggest that AI-powered IoT can significantly enhance precision agriculture and promote sustainable food production.","在杂草侵扰日益加剧的背景下，杂草、病虫害侵袭、劳动力短缺以及过度依赖化学农药是农业部门面临的严峻挑战。人工除草和病虫害防治具有潜在的环境负面影响，会增加生产成本并降低产量。人工智能（AI）、物联网（IoT）、深度学习、机器人技术及激光技术的应用，是发展可持续自主农业系统的有前景的途径。本研究提出一种基于人工智能（AI）和物联网（IoT）的智能农业系统，用于杂草识别、有害昆虫识别、作物状况监测以及利用激光束自动除草。该系统将基于物联网（IoT）传感器、计算机视觉模块、卷积神经网络（CNN）、无人机器人平台、环境感知系统及电子蜂群抑制（ESS）机制，以提高农业生产力并减少作物损失。该系统采用深度学习算法对杂草、作物、有益昆虫和有害昆虫进行实时分类。高分辨率摄像头和传感设备持续采集环境与作物数据，并将数据输入物联网系统，在云边计算系统中进行处理。通过配备激光发射器的机器人平台实现高精度除草，且不伤害邻近作物。此外，利用计算机视觉和环境监测来识别空中昆虫蜂群，并通过激光束和ESS技术实现对昆虫蜂群的自动攻击。研究采用多个农业区块对玉米、小麦、大豆和蔬菜作物进行了实验评估。结果表明，与当前做法相比，杂草检测、病虫害识别、作物保护的准确性显著提高，作业成本得以节省。所提出的系统在杂草分类中达到97.6%的准确率，昆虫检测准确率为96.8%，农药使用量减少42.3%。此外，作物产量提高18.7%，作业成本下降28.4%。总体而言，结果表明人工智能驱动的物联网能够显著增强精准农业并促进可持续粮食生产。","International Journal of Drug Delivery Technology",{"impact":104,"substance":168,"depth":105,"authority":103,"freshness":13,"relevant":58,"comment":222},"论文给出97.6%杂草识别率、42.3%农药减量等量化结论，方法新颖且数据完整，但期刊权威性一般，属细分领域技术进展。",[224],{"name":220,"url":217},[111,112,22,226,227],"杂草识别","病虫害监测",{"code":119,"slug":229,"name":230,"path":231,"parent_code":9},"digital-agriculture","数字农业与前沿技术","\u002Ftopic\u002Fdigital-agriculture",[],{"status":234,"locked":14,"version":29,"method":30,"evidence":235},"partial",{"candidates":236,"agricultural_evidence":36,"reason":293},[237,244,250,259,265,272,279,286],{"code":34,"score":104,"supported":36,"title":238,"abstract":240,"tags":243},[239],"deep learning",[241,21,242,239,38,40],"深度学习","病虫害识别",[112],{"code":51,"score":104,"supported":36,"title":245,"abstract":246,"tags":249},[54],[22,247,54,248],"internet of things","edge computing",[22],{"code":251,"score":13,"supported":36,"title":252,"abstract":253,"tags":258},"F4",[],[254,255,256,257],"农药","除草","pesticide","crop protection",[226],{"code":131,"score":44,"supported":36,"title":260,"abstract":261,"tags":264},[],[262,263],"精准农业","precision agriculture",[111],{"code":266,"score":91,"supported":36,"title":267,"abstract":268,"tags":271},"A2",[],[269,270],"小麦","wheat",[],{"code":273,"score":91,"supported":36,"title":274,"abstract":275,"tags":278},"A3",[],[276,277],"玉米","maize",[],{"code":280,"score":91,"supported":36,"title":281,"abstract":282,"tags":285},"A4",[],[283,284],"大豆","soybean",[],{"code":287,"score":91,"supported":36,"title":288,"abstract":289,"tags":292},"B1",[],[290,291],"蔬菜","vegetable",[],"一级领域明确，二级方向需要细分",[295,296],"AI IoT 激光除草 病虫害识别","深度学习 精准农业 农药减量","AIIoT激光除草病虫害识别-5556","10.25258\u002Fijddt.16.76s.32",{"doi":298,"openalex_id":300,"authors":301,"venue":220,"cited_by_count":15,"oa_url":217,"card":302,"direction":86,"ingested_from":88},"W7223145687",[],{"tldr":303,"method":304,"finding":305,"direction":86,"opportunity":306},"提出AI+IoT系统，用深度学习和激光机器人实现杂草与害虫识别及自动清除。","IoT传感器、CNN视觉分类、无人机\u002F机器人平台与激光除草，云边协同处理。","杂草分类准确率97.6%、害虫识别96.8%，农药减42.3%，增产18.7%。","可探索激光除草对非靶标生物与土壤的长期生态影响，及多作物跨域模型泛化。","2026-10-22T11:20:18.540359Z",{"id":309,"title":310,"url":311,"summary":312,"summary_zh":313,"content":9,"source_name":314,"source_url":311,"published_at":100,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":315,"score_detail":316,"sources":319,"tags":321,"primary_topic":323,"secondary_topics":324,"classification":335,"search_phrases":385,"slug":388,"view_count":15,"doi":389,"paper":390,"created_at":408},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)",80,{"impact":104,"substance":168,"depth":104,"authority":169,"freshness":317,"relevant":58,"comment":318},9,"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。",[320],{"name":314,"url":311},[111,112,22,262,322],"气候韧性",{"code":34,"slug":116,"name":117,"path":118,"parent_code":119},[325,329],{"code":51,"slug":326,"name":327,"path":328,"parent_code":119},"agricultural-iot","物联网传感与边缘计算","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-iot",{"code":330,"slug":331,"name":332,"path":333,"parent_code":334},"H3","irrigation","水资源与灌溉","\u002Ftopic\u002Fenvironment\u002Firrigation","H",{"status":122,"locked":14,"version":29,"method":30,"evidence":336},{"candidates":337,"agricultural_evidence":36},[338,344,349,354,360,365,372,379],{"code":34,"score":104,"supported":36,"title":339,"abstract":340,"tags":343},[38],[241,341,21,342,239,40],"机器学习","machine learning",[112],{"code":131,"score":345,"supported":36,"title":346,"abstract":347,"tags":348},16,[263],[262,263],[111,262],{"code":51,"score":13,"supported":36,"title":350,"abstract":351,"tags":353},[],[22,352,54,248],"边缘计算",[22],{"code":330,"score":317,"supported":36,"title":355,"abstract":356,"tags":359},[],[357,358,331],"灌溉","水资源",[],{"code":186,"score":44,"supported":36,"title":361,"abstract":362,"tags":364},[],[175,363],"climate change",[322],{"code":366,"score":91,"supported":36,"title":367,"abstract":368,"tags":371},"H2",[],[369,370],"土壤","soil",[],{"code":373,"score":374,"supported":36,"title":375,"abstract":376,"tags":378},"A7",3,[],[377],"间作",[],{"code":380,"score":374,"supported":36,"title":381,"abstract":382,"tags":384},"F3",[],[383],"养分管理",[],[386,387],"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":389,"openalex_id":391,"authors":392,"venue":314,"cited_by_count":15,"oa_url":311,"card":403,"direction":86,"ingested_from":88},"W7223121456",[393,396,398,400],{"name":394,"orcid":395},"Olumayowa J. Iyanda","https:\u002F\u002Forcid.org\u002F0000-0002-5340-9884",{"name":397,"orcid":9},"Oluwadare OYEBODE",{"name":399,"orcid":9},"Iretiayo Adelaiye",{"name":401,"orcid":402},"Ayomide Afolabi","https:\u002F\u002Forcid.org\u002F0000-0002-5171-886X",{"tldr":404,"method":405,"finding":406,"direction":157,"opportunity":407},"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",{"id":410,"title":411,"url":412,"summary":413,"summary_zh":9,"content":9,"source_name":414,"source_url":9,"published_at":415,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":416,"score_detail":417,"sources":420,"tags":422,"primary_topic":425,"secondary_topics":431,"classification":433,"search_phrases":456,"slug":459,"view_count":15,"doi":9,"paper":460,"created_at":469},5532,"精准棉花病害管理：融合无人机影像与定制YOLO深度学习框架","https:\u002F\u002Farxiv.org\u002Fabs\u002F2610.24182","研究提出单阶段全卷积检测器Cotton-Vision，融合幽灵瓶颈通道注意力与跨阶段部分融合三项轻量化设计，参数减少20%。在巴基斯坦旁遮普省7个棉田采集的2700余幅高分辨率无人机影像上测试，精确率97.7%、召回率77.7%、mAP@50为85.6%，较主流基线提升10至20个百分点。预印本，仅在巴基斯坦单一区域验证。","arXiv 预印本 2026-10-16","2026-10-16T00:00:00Z",69,{"impact":345,"substance":418,"depth":105,"authority":317,"freshness":91,"relevant":58,"comment":419},21,"方法轻量化与实测数据有增量，但仅单区域预印本验证，属细分技术进展，可入精选备选。",[421],{"name":414,"url":412},[111,112,423,23,424],"目标检测","棉花病害",{"code":426,"slug":427,"name":428,"path":429,"parent_code":430},"A6","cash-crops","棉麻糖与特色经作","\u002Ftopic\u002Ffield-crops\u002Fcash-crops","A",[432],{"code":34,"slug":116,"name":117,"path":118,"parent_code":119},{"status":122,"locked":14,"version":29,"method":30,"evidence":434},{"candidates":435,"agricultural_evidence":36},[436,441,448,452],{"code":34,"score":104,"supported":36,"title":437,"abstract":439,"tags":440},[241,438],"YOLO",[],[112,423],{"code":426,"score":442,"supported":36,"title":443,"abstract":445,"tags":447},11,[444],"棉花",[446],"cotton",[],{"code":131,"score":58,"supported":14,"title":449,"abstract":450,"tags":451},[],[],[111],{"code":57,"score":58,"supported":14,"title":453,"abstract":454,"tags":455},[],[],[23],[457,458],"Cotton-Vision 棉花病害","无人机影像 棉花 检测","Cotton-Vision棉花病害-5532",{"doi":9,"openalex_id":9,"authors":461,"venue":9,"cited_by_count":15,"oa_url":9,"card":462,"direction":466,"ingested_from":468},[],{"tldr":463,"method":464,"finding":465,"direction":466,"opportunity":467},"提出轻量化检测器Cotton-Vision，用无人机影像精准识别棉花病害。","基于YOLO改进，融合幽灵瓶颈通道注意力与跨阶段部分融合，用2700余幅无人机影","精确率97.7%、mAP@50达85.6%，较主流基线提升10至20个百分点。","农业遥感与作物表型","仅在巴基斯坦单一区域验证，可扩展至多区域多品种，并探索边缘设备实时部署。","agent","2026-10-22T00:07:09.816790Z",{"id":471,"title":472,"url":473,"summary":474,"summary_zh":475,"content":9,"source_name":99,"source_url":473,"published_at":476,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":477,"score_detail":478,"sources":481,"tags":483,"primary_topic":487,"secondary_topics":488,"classification":489,"search_phrases":513,"slug":516,"view_count":15,"doi":517,"paper":518,"created_at":532},5495,"Artificial Neural Network-Based Prediction and Mapping of Elephant Corridors in West Bengal Using Geospatial and Environmental Variables","https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23496876","The fragmentation of forest habitats and increasing anthropogenic pressures have significantly altered elephant movementpatternsinWestBengal,India,resultinginanescalatinghuman-elephantconflict.Accurateidentification and mapping of elephant corridors are essential for landscape-level conservation planning and mitigation.Traditional corridor mapping methods, including expert-based delineation and least-cost path analyses, oftenstruggle to capture the complex and nonlinear relationships between environmental variables influencing elephant movement. This study proposes an Artificial Neural Network (ANN)-based framework to predict potential elephant corridors across West Bengal using remotely sensed and geographic information system (GIS) datasets. The predictor variables included forest cover, normalized difference vegetation index (NDVI), elevation, slope,proximity to water sources, distance from roads, railways, settlements, agricultural land, and historical elephant occurrence records. The ANN model was trained using known elephant movement locations and validated using independent datasets. The predicted suitable surface was integrated into a GIS environment to identify probable movement corridors and prioritize areas for conservation. The proposed approach demonstrates the ability of machine learning to improve corridor prediction compared with conventional GIS-based techniques and provides a decision support tool for forest managers and policymakers.","森林栖息地的破碎化和日益增加的人为压力显著改变了印度西孟加拉邦大象的活动模式，导致人象冲突不断加剧。准确识别和绘制大象廊道对于景观层面的保护规划和冲突缓解至关重要。传统的廊道制图方法，包括基于专家的 delineation（边界划定）和最小成本路径分析，往往难以捕捉影响大象活动的环境变量之间复杂的非线性关系。本研究提出了一种基于人工神经网络（Artificial Neural Network, ANN）的框架，利用遥感数据和地理信息系统（Geographic Information System, GIS）数据集预测西孟加拉邦潜在的大象廊道。预测变量包括森林覆盖、归一化植被指数（Normalized Difference Vegetation Index, NDVI）、海拔、坡度、距水源的邻近度、距道路、铁路、居民点、农田的距离以及历史大象出现记录。ANN模型利用已知的大象活动位置进行训练，并使用独立数据集进行验证。将预测得到的适宜性表面整合到GIS环境中，以识别可能的迁移廊道并确定保护优先区域。所提出的方法展示了机器学习相较于传统基于GIS的技术在改进廊道预测方面的能力，并为森林管理者和政策制定者提供了决策支持工具。","2026-10-20T00:00:00Z",70,{"impact":106,"substance":479,"depth":105,"authority":106,"freshness":317,"relevant":58,"comment":480},20,"该研究用人工神经网络结合遥感与GIS预测大象迁徙走廊，方法新颖、数据扎实，对生态保护与乡村人兽冲突治理有参考价值，但属区域性研究，公共影响有限。",[482],{"name":99,"url":473},[112,484,485,24,486],"遥感","GIS","人兽冲突",{"code":119,"slug":229,"name":230,"path":231,"parent_code":9},[],{"status":234,"locked":14,"version":29,"method":30,"evidence":490},{"candidates":491,"agricultural_evidence":36,"reason":293},[492,497,504,509],{"code":34,"score":13,"supported":36,"title":493,"abstract":494,"tags":496},[],[341,342,495],"decision support",[112],{"code":498,"score":13,"supported":36,"title":499,"abstract":500,"tags":503},"I4",[],[501,485,502],"地理信息","geographic information",[485],{"code":57,"score":505,"supported":36,"title":506,"abstract":507,"tags":508},4,[],[484],[484],{"code":43,"score":58,"supported":14,"title":510,"abstract":511,"tags":512},[],[],[24],[514,515],"西孟加拉邦 大象走廊","人工神经网络 大象走廊","西孟加拉邦大象走廊-5495","10.5281\u002Fzenodo.23496876",{"doi":517,"openalex_id":519,"authors":520,"venue":99,"cited_by_count":15,"oa_url":473,"card":527,"direction":157,"ingested_from":88},"W7222999966",[521,523,525],{"name":522,"orcid":9},"Bhutia Dechen Doma",{"name":524,"orcid":9},"Swarup Das",{"name":526,"orcid":9},"Mandal Rakesh Kumar",{"tldr":528,"method":529,"finding":530,"direction":466,"opportunity":531},"用人工神经网络结合遥感与GIS变量预测并制图西孟加拉邦大象迁徙廊道。","ANN模型，输入遥感与GIS环境变量及历史象群出现记录，GIS集成制图。","ANN能刻画环境变量非线性关系，廊道预测优于传统GIS方法，可作决策支持工具。","可将ANN廊道预测与人类活动、农田冲突热点耦合，探索农业景观中野生动物通道与作物保护的协同规划。","2026-10-21T23:30:43.936453Z"]