[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-1891":3,"related-1891":74},{"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":22,"classification":23,"search_phrases":54,"slug":55,"view_count":15,"doi":56,"paper":57,"created_at":73},1891,"Computer Vision and Artificial Intelligence-Driven Structural Health Monitoring in Water Conservancy: Intelligent Early Warning and Economic Benefit Optimization","https:\u002F\u002Fdoi.org\u002F10.62177\u002Fapemr.v3i7.1710","Water conservancy structures serve as the core infrastructure for the operation of water conservancy projects; their structural health directly impacts regional flood control safety, water resource supply security, and ecological environment safety. Traditional health monitoring of water conservancy structures relies on manual inspections and wired sensing devices, but this approach suffers from limitations such as limited coverage, insufficient detection timeliness, and high maintenance and operation costs. This paper investigates the application of computer vision and artificial intelligence (AI) technologies in the field of water conservancy structure health monitoring. It outlines the technical pathways employed by computer vision techniques to identify structural anomalies—such as surface cracks, deformation, or damage—and analyzes the predictive logic of AI algorithms for forecasting the evolution trends of structural health. The paper proposes a framework for an intelligent safety early-warning system tailored for water conservancy structures and systematically evaluates the economic benefits derived from the application of these technologies from both direct and indirect perspectives. Case studies based on actual engineering applications demonstrate that this technical framework can significantly enhance the accuracy of anomaly detection and the timeliness of early warnings for water conservancy structures, reduce manual maintenance and operation costs, and extend the service life of these structures. Finally, this paper identifies the challenges currently faced during the implementation phase of these technologies—particularly in data processing and privacy protection—and proposes corresponding solutions. It also offers insights into future technological development directions in this field, providing a valuable reference for the intelligent operation and maintenance of water conservancy projects.","水利构筑物是水利工程运行的核心基础设施，其结构健康直接关系到区域防洪安全、水资源供给安全及生态环境安全。传统水利构筑物健康监测依赖人工巡检与有线传感设备，存在覆盖范围有限、检测时效性不足及运维成本高等局限。本文探讨了计算机视觉与人工智能（AI）技术在水利构筑物健康监测领域的应用，概述了计算机视觉技术识别结构异常（如表面裂缝、变形或损伤）的技术路径，并分析了AI算法预测结构健康演化趋势的逻辑。文章提出了一套面向水利构筑物的智能安全预警系统框架，并从直接与间接两个维度系统评估了这些技术应用所带来的经济效益。基于实际工程应用的案例研究表明，该技术框架能显著提升水利构筑物异常检测的准确性和预警的时效性，降低人工运维成本，并延长构筑物的使用寿命。最后，本文指出了当前技术实施阶段面临的挑战，特别是数据处理与隐私保护方面的问题，并提出了相应的解决对策，同时对该领域未来的技术发展方向进行了展望，为水利工程的智能化运维提供了有益参考。",null,"Asia Pacific Economic and Management Review","2026-09-06T00:00:00Z","论文",10,false,0,{"impact":15,"substance":15,"depth":15,"authority":15,"freshness":15,"relevant":15,"comment":17},"主题为水利工程健康监测，不属于三农、农业信息化或智慧农业范畴。",[19],{"name":10,"url":6},[21],"智慧农业",[],{"status":24,"locked":14,"version":25,"method":26,"evidence":27},"needs_review","2026-10-22.2","rules",{"candidates":28,"agricultural_evidence":14,"reason":53},[29,39,47],{"code":30,"score":31,"supported":32,"title":33,"abstract":36,"tags":38},"I2",25,true,[34,35],"artificial intelligence","computer vision",[37,34,35],"计算机视觉",[],{"code":40,"score":41,"supported":32,"title":42,"abstract":43,"tags":46},"H3",6,[],[44,45],"水资源","water resource",[],{"code":48,"score":49,"supported":14,"title":50,"abstract":51,"tags":52},"I1",1,[],[],[21],"农业相关性需要复核",[],"ComputerVisionandArtificialIntelligence--1891","10.62177\u002Fapemr.v3i7.1710",{"doi":56,"openalex_id":58,"authors":59,"venue":10,"cited_by_count":15,"oa_url":64,"card":65,"direction":71,"ingested_from":72},"W7210270447",[60,62],{"name":61,"orcid":9},"Yunshu Luo",{"name":63,"orcid":9},"Hanwen Xue","https:\u002F\u002Fojs.apspublisher.com\u002Findex.php\u002Fapemr\u002Farticle\u002Fdownload\u002F1710\u002F1332",{"tldr":66,"method":67,"finding":68,"direction":69,"opportunity":70},"研究计算机视觉与AI在水工结构健康监测中的应用，提出智能预警框架并评估经济效益。","综述技术路径，结合案例评估，提出智能预警系统框架。","该框架能提高异常检测精度和预警及时性，降低维护成本，延长结构寿命。","其他","可延伸研究水工结构监测中的数据处理与隐私保护，以及多模态数据融合的智能预警算法。","智慧农业 \u002F 农业物联网","openalex","2026-09-08T23:30:11.649850Z",{"total":41,"page":49,"page_size":41,"items":75},[76,132,179,237,280,336],{"id":77,"title":78,"url":79,"summary":80,"summary_zh":9,"content":9,"source_name":81,"source_url":9,"published_at":82,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":83,"score_detail":84,"sources":89,"tags":91,"primary_topic":9,"secondary_topics":96,"classification":97,"search_phrases":118,"slug":121,"view_count":15,"doi":9,"paper":122,"created_at":131},5658,"基于混合卷积Transformer与知识自蒸馏的香蕉品种跨等级分类泛化增强","https:\u002F\u002Farxiv.org\u002Fabs\u002F2610.29008","预印本，未同行评审。针对香蕉品种分类中训练等级与实际品质等级的域偏移，用31678张图像、三个品种三个等级数据并采用留一等级交叉验证，比较卷积网络、视觉Transformer与双流架构；提出的混合模型加自蒸馏平均准确率达81.25%，优于两类基线。","arXiv 2026-10-21","2026-10-20T16:00:00Z",56,{"impact":85,"substance":86,"depth":87,"authority":41,"freshness":85,"relevant":49,"comment":88},8,18,16,"预印本提出混合卷积Transformer与知识自蒸馏方法，在香蕉品种跨等级分类上取得81.25%准确率，方法有增量但属细分技术进展，未同行评审，公共影响有限。",[90],{"name":81,"url":79},[21,92,93,94,95],"农业人工智能","作物分类","香蕉","领域泛化",[],{"status":24,"locked":14,"version":25,"method":26,"evidence":98},{"candidates":99,"agricultural_evidence":14,"reason":53},[100,105,109,113],{"code":101,"score":49,"supported":14,"title":102,"abstract":103,"tags":104},"B2",[],[],[94],{"code":48,"score":49,"supported":14,"title":106,"abstract":107,"tags":108},[],[],[21],{"code":30,"score":49,"supported":14,"title":110,"abstract":111,"tags":112},[],[],[92],{"code":114,"score":49,"supported":14,"title":115,"abstract":116,"tags":117},"I3",[],[],[93],[119,120],"香蕉品种 跨等级分类","混合卷积Transformer 知识自蒸馏","香蕉品种跨等级分类-5658",{"doi":9,"openalex_id":9,"authors":123,"venue":9,"cited_by_count":15,"oa_url":9,"card":124,"direction":128,"ingested_from":130},[],{"tldr":125,"method":126,"finding":127,"direction":128,"opportunity":129},"提出混合卷积Transformer加知识自蒸馏，提升香蕉品种跨等级分类泛化能力。","31678张图像、三品种三等级，留一等级交叉验证，对比CNN、ViT与双流架构。","混合模型加自蒸馏平均准确率81.25%，优于卷积网络和视觉Transformer基线。","农业人工智能与决策模型","可探索知识自蒸馏在更多农产品跨域分级中的应用，并研究轻量化部署与真实分拣场景验证。","agent","2026-10-23T00:32:42.976104Z",{"id":133,"title":134,"url":135,"summary":136,"summary_zh":9,"content":9,"source_name":137,"source_url":9,"published_at":138,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":139,"score_detail":140,"sources":145,"tags":147,"primary_topic":9,"secondary_topics":152,"classification":153,"search_phrases":166,"slug":169,"view_count":15,"doi":9,"paper":170,"created_at":178},5657,"ADA 2025：甜菜水分亏缺胁迫检测多模态数据集","https:\u002F\u002Farxiv.org\u002Fabs\u002F2610.23609","预印本，未同行评审。发布超过500GB的甜菜缺水胁迫多模态公开数据集，融合热红外、高光谱、彩色影像与深度影像，附环境元数据及相对含水量、叶厚两类真值，填补作物水分胁迫领域公开数据稀缺的空白；仅展示数据有效性，未系统评测模型性能。","arXiv 2026-10-16","2026-10-15T16:00:00Z",75,{"impact":86,"substance":141,"depth":142,"authority":143,"freshness":41,"relevant":49,"comment":144},22,17,12,"500GB级公开多模态数据集填补作物水分胁迫数据空白，对智慧农业研究有实质价值，但仅预印本且未评测模型，时效偏旧。",[146],{"name":137,"url":135},[21,148,149,150,151],"甜菜","遥感监测","水分胁迫","多模态数据",[],{"status":154,"locked":14,"version":25,"method":26,"evidence":155},"unclassified",{"candidates":156,"agricultural_evidence":32,"reason":165},[157,161],{"code":48,"score":49,"supported":14,"title":158,"abstract":159,"tags":160},[],[],[21],{"code":114,"score":49,"supported":14,"title":162,"abstract":163,"tags":164},[],[],[149],"缺少足够的主题证据",[167,168],"甜菜 水分亏缺 多模态数据集","甜菜 高光谱 热红外 胁迫检测","甜菜水分亏缺多模态数据集-5657",{"doi":9,"openalex_id":9,"authors":171,"venue":9,"cited_by_count":15,"oa_url":9,"card":172,"direction":176,"ingested_from":130},[],{"tldr":173,"method":174,"finding":175,"direction":176,"opportunity":177},"发布超500GB甜菜水分亏缺多模态公开数据集，含热红外、高光谱、彩色与深度影像及两类真值。","融合热红外、高光谱、彩色、深度影像与环境元数据，附相对含水量和叶厚真值。","数据集有效可用，填补作物水分胁迫公开数据稀缺空白，但未系统评测模型性能。","农业遥感与作物表型","可基于该多模态数据集系统评测水分胁迫检测模型，探索多源影像融合与跨模态特征学习。","2026-10-23T00:32:42.559583Z",{"id":180,"title":181,"url":182,"summary":183,"summary_zh":9,"content":9,"source_name":184,"source_url":9,"published_at":185,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":186,"score_detail":187,"sources":190,"tags":192,"primary_topic":197,"secondary_topics":203,"classification":204,"search_phrases":225,"slug":228,"view_count":15,"doi":9,"paper":229,"created_at":236},5655,"面向节能温室运行的LED照明控制策略比较研究","https:\u002F\u002Farxiv.org\u002Fabs\u002F2610.28570","预印本，未同行评审。在同一温室并行对比常开、启发式与基于马尔可夫链太阳预报的预测补光控制，2026年2—4月番茄两重复试验显示预测策略达到目标日光积分且成本较常开降低65%，单位电费干物质18.976克、约为常开两倍多；仅单温室验证。","arXiv 2026-10-20","2026-10-19T16:00:00Z",66,{"impact":143,"substance":188,"depth":142,"authority":85,"freshness":85,"relevant":49,"comment":189},21,"预印本单温室验证，节能数据有参考价值但可靠性有限，属细分技术进展，可作候选而非重点精选。",[191],{"name":184,"url":182},[21,193,194,195,196],"LED补光","温室环控","节能控制","番茄栽培",{"code":198,"slug":199,"name":200,"path":201,"parent_code":202},"B7","protected-horticulture","设施园艺与植物工厂","\u002Ftopic\u002Fhorticulture\u002Fprotected-horticulture","B",[],{"status":205,"locked":14,"version":25,"method":26,"evidence":206},"classified",{"candidates":207,"agricultural_evidence":32},[208,214,221],{"code":198,"score":209,"supported":32,"title":210,"abstract":212,"tags":213},11,[211],"温室",[211],[],{"code":215,"score":216,"supported":32,"title":217,"abstract":218,"tags":220},"B1",3,[],[219],"番茄",[],{"code":48,"score":49,"supported":14,"title":222,"abstract":223,"tags":224},[],[],[21],[226,227],"温室 LED补光 节能控制","马尔可夫链 太阳预报 补光","温室LED补光节能控制-5655",{"doi":9,"openalex_id":9,"authors":230,"venue":9,"cited_by_count":15,"oa_url":9,"card":231,"direction":71,"ingested_from":130},[],{"tldr":232,"method":233,"finding":234,"direction":71,"opportunity":235},"在同一温室对比三种LED补光控制策略，预测策略节能且增产。","并行对比常开、启发式与马尔可夫链太阳预报预测控制，番茄两重复试验。","预测策略达标日光积分且成本降65%，单位电费干物质约为常开两倍多。","仅单温室验证，可扩展多温室多作物验证，并融合实时传感与电价优化控制。","2026-10-23T00:32:42.146061Z",{"id":238,"title":239,"url":240,"summary":241,"summary_zh":9,"content":9,"source_name":242,"source_url":9,"published_at":243,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":244,"score_detail":245,"sources":248,"tags":250,"primary_topic":9,"secondary_topics":255,"classification":256,"search_phrases":268,"slug":271,"view_count":15,"doi":9,"paper":272,"created_at":279},5654,"MultiAltitudeSR：基于物理地面采样距离条件化的多航高农业无人机图像超分辨率","https:\u002F\u002Farxiv.org\u002Fabs\u002F2610.24992","预印本，未同行评审。针对农业无人机航高与空间分辨率的矛盾，提出以源与目标地面采样距离为条件的超分框架，用10—70米真实航拍图联合训练，21种航高组合免微调即可支持1.18至6.03倍放大；下游仅做零样本分割初步验证。","arXiv 2026-10-18","2026-10-17T16:00:00Z",70,{"impact":87,"substance":188,"depth":86,"authority":246,"freshness":41,"relevant":49,"comment":247},9,"预印本提出按地面采样距离条件化的多航高超分框架，覆盖21种航高组合且免微调，方法新颖、数据规模可观，但下游仅零样本分割初步验证且未经同行评审，属细分领域值得关注的进展。",[249],{"name":242,"url":240},[21,251,252,253,254],"无人机","农业遥感","作物监测","图像超分辨率",[],{"status":154,"locked":14,"version":25,"method":26,"evidence":257},{"candidates":258,"agricultural_evidence":32,"reason":165},[259,264],{"code":114,"score":260,"supported":14,"title":261,"abstract":262,"tags":263},2,[],[],[253,252],{"code":48,"score":49,"supported":14,"title":265,"abstract":266,"tags":267},[],[],[21],[269,270],"农业无人机 超分辨率","地面采样距离 无人机遥感","农业无人机超分辨率-5654",{"doi":9,"openalex_id":9,"authors":273,"venue":9,"cited_by_count":15,"oa_url":9,"card":274,"direction":176,"ingested_from":130},[],{"tldr":275,"method":276,"finding":277,"direction":176,"opportunity":278},"提出按源与目标地面采样距离条件化的超分框架，实现多航高农业无人机图像免微调放大。","以源\u002F目标地面采样距离为条件，用10—70米真实航拍图联合训练超分模型。","21种航高组合免微调支持1.18至6.03倍放大，下游仅零样本分割初步验证。","下游任务仅零样本分割初验，可系统评估超分对作物表型与分割精度的影响及跨航高泛化。","2026-10-23T00:32:42.046879Z",{"id":281,"title":282,"url":283,"summary":284,"summary_zh":9,"content":9,"source_name":285,"source_url":9,"published_at":286,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":287,"score_detail":288,"sources":290,"tags":292,"primary_topic":296,"secondary_topics":300,"classification":301,"search_phrases":324,"slug":327,"view_count":15,"doi":9,"paper":328,"created_at":335},5653,"蔬菜田半自动驾驶精准喷药系统多速度田间性能评价","https:\u002F\u002Farxiv.org\u002Fabs\u002F2610.24893","预印本，未同行评审。用五个生长季10599张图、84311个标注实例训练目标检测模型，结合多目标跟踪与12喷头装置在萝卜田开展三档速度田间试验。最低速平均精度达94.6%，提速后杂草命中率由78.3%降至35.0%，未达预设商业门槛。","arXiv 2026-10-17","2026-10-16T16:00:00Z",64,{"impact":143,"substance":188,"depth":142,"authority":85,"freshness":41,"relevant":49,"comment":289},"数据规模扎实、结论明确的预印本田间评价研究，但提速后命中率大幅下降且未达商业门槛，属细分技术进展，可作候选而非必选。",[291],{"name":285,"url":283},[21,92,293,294,295],"杂草识别","田间试验","精准喷药",{"code":215,"slug":297,"name":298,"path":299,"parent_code":202},"vegetables","蔬菜","\u002Ftopic\u002Fhorticulture\u002Fvegetables",[],{"status":205,"locked":14,"version":25,"method":26,"evidence":302},{"candidates":303,"agricultural_evidence":32},[304,309,315,320],{"code":215,"score":209,"supported":32,"title":305,"abstract":306,"tags":308},[298],[307],"萝卜",[],{"code":30,"score":310,"supported":32,"title":311,"abstract":312,"tags":314},4,[],[313],"目标检测",[92],{"code":316,"score":49,"supported":14,"title":317,"abstract":318,"tags":319},"F4",[],[],[293],{"code":48,"score":49,"supported":14,"title":321,"abstract":322,"tags":323},[],[],[21],[325,326],"萝卜田 精准喷药 田间试验","半自动驾驶 喷药 杂草检测","萝卜田精准喷药田间试验-5653",{"doi":9,"openalex_id":9,"authors":329,"venue":9,"cited_by_count":15,"oa_url":9,"card":330,"direction":71,"ingested_from":130},[],{"tldr":331,"method":332,"finding":333,"direction":71,"opportunity":334},"在萝卜田测试半自动精准喷药系统，评估三档速度下的检测与喷施性能。","五个生长季10599张图训练目标检测，结合多目标跟踪与12喷头装置开展田间试验。","最低速精度94.6%，提速后杂草命中率由78.3%降至35.0%，未达商业门槛。","高速下跟踪与喷施协同失效是瓶颈，可研究速度自适应跟踪与喷头时序补偿算法。","2026-10-23T00:32:41.788532Z",{"id":337,"title":338,"url":339,"summary":340,"summary_zh":9,"content":9,"source_name":341,"source_url":9,"published_at":342,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":343,"score_detail":344,"sources":347,"tags":349,"primary_topic":353,"secondary_topics":354,"classification":355,"search_phrases":370,"slug":373,"view_count":15,"doi":9,"paper":374,"created_at":381},5652,"Bombadil：面向植物电生理信号的基础模型","https:\u002F\u002Farxiv.org\u002Fabs\u002F2610.26597","预印本，未同行评审。以100个植物物种、十类环境（实验室、温室、商业果园、天然林）超400万小时植物电信号无标签预训练基础模型，可从同一信号追踪七项胁迫生理指标变化，并覆盖训练中无标签的物种与胁迫类型；跨场景田间验证仍有限。","arXiv 2026-10-19","2026-10-18T16:00:00Z",74,{"impact":86,"substance":141,"depth":345,"authority":246,"freshness":41,"relevant":49,"comment":346},19,"预印本尚未经同行评审，但以400万小时、100个物种的植物电信号预训练基础模型，在方法新颖性与数据规模上具有实质增量，值得关注但田间验证有限。",[348],{"name":341,"url":339},[21,92,350,351,352],"基础模型","植物电生理","胁迫监测",{"code":198,"slug":199,"name":200,"path":201,"parent_code":202},[],{"status":205,"locked":14,"version":25,"method":26,"evidence":356},{"candidates":357,"agricultural_evidence":32},[358,362,366],{"code":198,"score":216,"supported":32,"title":359,"abstract":360,"tags":361},[],[211],[],{"code":48,"score":49,"supported":14,"title":363,"abstract":364,"tags":365},[],[],[21],{"code":30,"score":49,"supported":14,"title":367,"abstract":368,"tags":369},[],[],[92],[371,372],"Bombadil 植物电信号 基础模型","植物电生理 胁迫监测","Bombadil植物电信号基础模型-5652",{"doi":9,"openalex_id":9,"authors":375,"venue":9,"cited_by_count":15,"oa_url":9,"card":376,"direction":128,"ingested_from":130},[],{"tldr":377,"method":378,"finding":379,"direction":128,"opportunity":380},"构建植物电生理信号基础模型，用超400万小时无标签数据预训练。","100个物种、十类环境超400万小时植物电信号无标签预训练。","同一信号可追踪七项胁迫生理指标，并泛化到未见物种与胁迫类型。","跨场景田间验证不足，可研究多环境迁移与胁迫早期预警的落地评估。","2026-10-23T00:32:41.455307Z"]