[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-2176":3,"related-2176":93},{"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":9,"secondary_topics":32,"classification":33,"search_phrases":65,"slug":68,"view_count":69,"doi":70,"paper":71,"created_at":92},2176,"A Robust and Efficient Scheme for Real-Time SAR Speckle Noise Reduction based on Attentional Self-Supervised Contrastive Deep Learning","https:\u002F\u002Fdoi.org\u002F10.17725\u002Fj.rensit.2026.18.429","Synthetic aperture radar (SAR) is a critical and essential technology for remote sensing, achieving high resolution under all weather and day\u002Fnight conditions. But SAR images suffer from multiplicative speckle noise, making the images poor in quality and resulting in adverse effects on the image interpretation and automatic analysis. In order to solve this problem, a novel Residual Attention Swin Transformer Network (RAST-Net) is proposed to efficiently despeckle the SAR images.To improve local feature representation, the suggested architecture combines residual convolutional learning with the Convolutional Block Attention Module (CBAM); light Swin Transformer blocks are added to catch long-range contextual dependencies.Additionally, the self-supervised contrastive learning method is applied to enhance the capacity to differentiate features and achieve generalization without much dependence on clear reference data. The techniques of quantization-aware optimization are also incorporated into the proposed design to ensure the successful implementation of the method on real devices like NVIDIA Jetson Orin NX, and Xilinx ZynqUltraScale+ MPSoC FPGA systems. Comprehensive experimental results indicate that the proposed model demonstrates better performance than conventional filters, CNN networks, self-supervised models, and some other recently developed transformers and diffusion-based approaches operating with various objective indices such as PSNR, SSIM, FSIM ENL, and EPI.The proposed network not only achieves high restoration accuracy, but also has moderate computational complexity, low inference delay and excellent deployment efficiency on embedded hardware platform. Based on these results, the proposed RAST-Net is an effective and practical solution for real-time SAR image restoration in intelligent remote sensing and edge AI applications.","合成孔径雷达（SAR）是一种关键且必不可少的遥感技术，能够在全天候、全天时条件下实现高分辨率成像。但SAR图像受到乘性斑点噪声的影响，导致图像质量下降，并对图像解译和自动分析产生不利影响。为了解决这一问题，提出了一种新型残差注意力Swin Transformer网络（RAST-Net），用于高效地对SAR图像进行去斑。为提升局部特征表达能力，所提出的架构将残差卷积学习与卷积块注意力模块（CBAM）相结合；并加入轻量Swin Transformer块以捕获长距离上下文依赖关系。此外，采用自监督对比学习方法增强特征区分能力，并在不过度依赖清晰参考数据的情况下实现泛化。该设计还融入了量化感知优化技术，以确保该方法能够在NVIDIA Jetson Orin NX和Xilinx ZynqUltraScale+ MPSoC FPGA系统等真实设备上成功部署。综合实验结果表明，在PSNR、SSIM、FSIM、ENL和EPI等多种客观指标下，所提出的模型相比传统滤波器、CNN网络、自监督模型以及一些其他近期开发的Transformer和基于扩散的方法表现出更好的性能。所提出的网络不仅实现了较高的恢复精度，而且具有适中的计算复杂度、较低的推理延迟以及在嵌入式硬件平台上的出色部署效率。基于这些结果，所提出的RAST-Net是智能遥感和边缘AI应用中实时SAR图像恢复的一种有效且实用的解决方案。",null,"Radioelectronics Nanosystems Information Technologies","2026-09-08T00:00:00Z","论文",10,false,75,{"impact":17,"substance":18,"depth":19,"authority":20,"freshness":21,"relevant":22,"comment":23},16,21,18,12,8,1,"提出RAST-Net自监督对比学习去斑方案并在嵌入式平台验证，方法新颖、指标充分，对农业遥感影像智能解译有实用价值，但属细分技术进展，影响层级有限。",[25],{"name":10,"url":6},[27,28,29,30,31],"智慧农业","农业人工智能","边缘计算","遥感","SAR影像",[],{"status":34,"locked":14,"version":35,"method":36,"evidence":37},"needs_review","2026-10-22.2","rules",{"candidates":38,"agricultural_evidence":14,"reason":64},[39,47,54,59],{"code":40,"score":41,"supported":42,"title":43,"abstract":45,"tags":46},"I2",9,true,[44],"deep learning",[],[28],{"code":48,"score":49,"supported":42,"title":50,"abstract":51,"tags":53},"I3",7,[],[30,52],"remote sensing",[30],{"code":55,"score":22,"supported":14,"title":56,"abstract":57,"tags":58},"I1",[],[],[27],{"code":60,"score":22,"supported":14,"title":61,"abstract":62,"tags":63},"I5",[],[],[29],"农业相关性需要复核",[66,67],"农业人工智能 智慧农业 边缘计算 SAR影像","农业人工智能 智慧农业","农业人工智能智慧农业边缘计算SAR影像-2176",4,"10.17725\u002Fj.rensit.2026.18.429",{"doi":70,"openalex_id":72,"authors":73,"venue":10,"cited_by_count":83,"oa_url":84,"card":85,"direction":89,"ingested_from":91},"W7211960754",[74,76,79,81],{"name":75,"orcid":9},"Marwah M. Hassooni",{"name":77,"orcid":78},"Abdullah Al-Obaidi","https:\u002F\u002Forcid.org\u002F0000-0002-6541-8999",{"name":80,"orcid":9},"Mustafa Nadhim Ghazal",{"name":82,"orcid":9},"Anas Fouad Ahmed.",0,"http:\u002F\u002Fen.rensit.ru\u002Fvypuski\u002Farticle\u002F774\u002F18(4)429-442e.pdf",{"tldr":86,"method":87,"finding":88,"direction":89,"opportunity":90},"提出RAST-Net，用注意力自监督对比深度学习实现SAR图像实时去斑。","残差卷积+CBAM+轻量Swin Transformer，自监督对比学习与量化优","在多项指标上优于传统滤波、CNN、自监督及扩散方法，且嵌入式部署高效。","农业遥感与作物表型","可探索该轻量去斑模型在农业SAR作物监测边缘设备上的迁移与实时应用。","openalex","2026-09-11T23:30:34.257413Z",{"total":94,"page":22,"page_size":94,"items":95},6,[96,193,273,335,390,500],{"id":97,"title":98,"url":99,"summary":100,"summary_zh":101,"content":9,"source_name":102,"source_url":99,"published_at":103,"category":12,"cover_url":9,"hotness":13,"is_selected":42,"score":104,"score_detail":105,"sources":109,"tags":111,"primary_topic":113,"secondary_topics":118,"classification":124,"search_phrases":154,"slug":157,"view_count":158,"doi":159,"paper":160,"created_at":192},3842,"Raspberry Pi Lightweight Deep Learning Models Deployed for Corn Growth Stage Classification Using PhenoCam Images","https:\u002F\u002Fdoi.org\u002F10.3390\u002Frs18193337","Identifying crop growth stages is essential for comprehensive assessment that facilitates timely and precise agronomic management. However, conventional field-based manual scouting is labor-intensive, time-consuming, and lacks scalability. Digital field images with advanced models deployed on portable edge-computing devices offer an alternative solution for real-time crop monitoring and decision support. Therefore, in this study, the standard 8 corn growth stages were defined as 8 classes based on distinct visual canopy appearance patterns observed in PhenoCam (near-surface [proximal] remote sensing network camera) imagery collected from 10 PhenoCam sites. These classes do not correspond directly to standard agronomic vegetation and reproductive growth stages. For modeling, four lightweight deep learning (DL) models, ELiteCrop0, ELiteCrop1, ELiteCrop4, and MobNetCropV2, were developed and evaluated across five image vertical clipping levels (0–40%) using a supercomputer (NDSU CCAST). The optimized model was subsequently deployed on a single-board Raspberry Pi 5 computer for edge inference. Model training accounted for the majority of the total CPU time, exceeding 97%, while the testing times ranged from 0.01–0.12 min, enabling real-time applications. Among the models, ELiteCrop0 achieved the most balanced performance with a confusion-matrix diagonal ratio (CMDR) of 0.93, followed by ELiteCrop1 (CMDR = 0.92). Overall, model performance decreased with increasing vertical clipping; therefore, a moderate image clipping (0–10%) was recommended for improved computational efficiency. Analysis with a supercomputer produced an intrasite (same training sites) accuracy of 0.90–0.93 (Raspberry Pi: 0.78–0.81) and an intersite (unseen test sites) accuracy of 0.48–0.50 (Raspberry Pi: 0.41–0.43), indicating challenges with model generalization. Raspberry Pi processed ≈ 1000 images\u002Fmin at 68 °C, with a PhenoCam:Raspberry Pi processing ratio of 1:2500 at 50% capacity over an 8 h day. Overall, this study presents a scalable and cost-effective solution for real-time corn growth stage monitoring in precision agriculture, which could be extended to other crops.","识别作物生长阶段对于全面评估、促进及时精准的农艺管理至关重要。然而，传统的田间人工巡查劳动强度大、耗时且缺乏可扩展性。部署在便携式边缘计算设备上的数字田间图像与先进模型为实时作物监测和决策支持提供了替代方案。因此，本研究基于从10个PhenoCam（近地面[近端]遥感网络相机）站点采集的影像中观察到的不同冠层视觉外观模式，将标准的8个玉米生长阶段定义为8个类别。这些类别并不直接对应于标准的农艺营养生长和生殖生长阶段。在建模方面，开发了四种轻量级深度学习（DL）模型——ELiteCrop0、ELiteCrop1、ELiteCrop4和MobNetCropV2，并利用超级计算机（NDSU CCAST）在五种图像垂直裁剪水平（0–40%）下进行了评估。随后，优化后的模型被部署到单板Raspberry Pi 5计算机上进行边缘推理。模型训练占总CPU时间的大部分，超过97%，而测试时间为0.01–0.12分钟，支持实时应用。在各模型中，ELiteCrop0实现了最均衡的性能，混淆矩阵对角线比率（CMDR）为0.93，其次是ELiteCrop1（CMDR = 0.92）。总体而言，模型性能随垂直裁剪比例的增加而下降；因此，建议采用适度的图像裁剪（0–10%）以提高计算效率。使用超级计算机的分析得出站内（相同训练站点）准确率为0.90–0.93（Raspberry Pi：0.78–0.81），站间（未见测试站点）准确率为0.48–0.50（Raspberry Pi：0.41–0.43），表明模型泛化面临挑战。Raspberry Pi在68 °C下处理约1000张图像\u002F分钟，在8小时工作日50%容量下，PhenoCam与Raspberry Pi的处理比为1:2500。总体而言，本研究为精准农业中的实时玉米生长阶段监测提供了一种可扩展且经济高效的解决方案，并可扩展至其他作物。","Remote Sensing","2026-09-29T00:00:00Z",81,{"impact":19,"substance":106,"depth":19,"authority":107,"freshness":41,"relevant":22,"comment":108},22,14,"该研究将轻量深度学习模型部署于树莓派实现玉米生育期实时识别，方法新颖、数据扎实，对精准农业边缘监测有参考价值。",[110],{"name":102,"url":99},[27,28,29,112,30],"玉米",{"code":114,"slug":115,"name":112,"path":116,"parent_code":117},"A3","maize","\u002Ftopic\u002Ffield-crops\u002Fmaize","A",[119],{"code":40,"slug":120,"name":121,"path":122,"parent_code":123},"agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai","I",{"status":125,"locked":14,"version":35,"method":36,"evidence":126},"classified",{"candidates":127,"agricultural_evidence":42},[128,134,140,146,150],{"code":40,"score":19,"supported":42,"title":129,"abstract":130,"tags":133},[44],[131,44,132],"深度学习","decision support",[28],{"code":114,"score":135,"supported":42,"title":136,"abstract":138,"tags":139},15,[137],"corn",[112,137],[112],{"code":55,"score":49,"supported":42,"title":141,"abstract":142,"tags":145},[],[143,144],"精准农业","precision agriculture",[27],{"code":48,"score":49,"supported":42,"title":147,"abstract":148,"tags":149},[],[30,52],[30],{"code":60,"score":69,"supported":42,"title":151,"abstract":152,"tags":153},[],[29],[29],[155,156],"Raspberry Pi 玉米 生长阶段","PhenoCam 玉米 深度学习","RaspberryPi玉米生长阶段-3842",5,"10.3390\u002Frs18193337",{"doi":159,"openalex_id":161,"authors":162,"venue":102,"cited_by_count":83,"oa_url":99,"card":187,"direction":89,"ingested_from":91},"W7214787420",[163,166,169,172,175,178,181,184],{"name":164,"orcid":165},"Anne Joice","https:\u002F\u002Forcid.org\u002F0000-0001-7499-9222",{"name":167,"orcid":168},"Humeera Tazeen","https:\u002F\u002Forcid.org\u002F0000-0003-2411-0725",{"name":170,"orcid":171},"Talha Tufaique","https:\u002F\u002Forcid.org\u002F0000-0002-4490-5341",{"name":173,"orcid":174},"Mathala Juliet Gupta","https:\u002F\u002Forcid.org\u002F0000-0002-3886-2429",{"name":176,"orcid":177},"Cannayen Igathinathane","https:\u002F\u002Forcid.org\u002F0000-0001-8884-7959",{"name":179,"orcid":180},"Nitin Rai","https:\u002F\u002Forcid.org\u002F0000-0002-9350-9601",{"name":182,"orcid":183},"Craig W. Whippo","https:\u002F\u002Forcid.org\u002F0000-0002-7626-6919",{"name":185,"orcid":186},"David Walter Archer","https:\u002F\u002Forcid.org\u002F0000-0002-4816-7040",{"tldr":188,"method":189,"finding":190,"direction":89,"opportunity":191},"用PhenoCam图像训练轻量深度学习模型，在树莓派上实时分类玉米8个生长阶段。","基于10个PhenoCam站点图像，开发4种轻量DL模型，在树莓派5上边缘推理。","ELiteCrop0表现最佳（CMDR=0.93），站点内精度0.90-0.93，但站点间泛化差（0","站点间泛化差是主要瓶颈，可研究跨站点迁移学习或域适应以提升模型鲁棒性。","2026-09-30T23:30:29.168194Z",{"id":194,"title":195,"url":196,"summary":197,"summary_zh":198,"content":9,"source_name":199,"source_url":196,"published_at":200,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":201,"score_detail":202,"sources":204,"tags":206,"primary_topic":9,"secondary_topics":208,"classification":209,"search_phrases":230,"slug":233,"view_count":69,"doi":234,"paper":235,"created_at":272},2019,"AgroInsect: A hybrid AI-driven edge-cloud architecture for precision pest detection and geostatistical interpolation in agriculture","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.compag.2026.112386","AgroInsect: A hybrid AI-driven edge-cloud architecture for precision pest detection and geostatistical interpolation in agriculture。Computers and Electronics in Agriculture","AgroInsect：一种用于农业精准害虫检测与地统计插值的混合人工智能驱动边缘-云架构。","Computers and Electronics in Agriculture","2026-09-09T00:00:00Z",79,{"impact":19,"substance":18,"depth":19,"authority":107,"freshness":21,"relevant":22,"comment":203},"核心期刊论文，提出边缘-云协同的害虫精准检测与地统计插值架构，方法新颖且面向田间应用，具备智慧农业技术参考价值。",[205],{"name":199,"url":196},[27,28,29,30,207],"病虫害监测",[],{"status":210,"locked":14,"version":35,"method":36,"evidence":211},"unclassified",{"candidates":212,"agricultural_evidence":42,"reason":229},[213,217,221,225],{"code":55,"score":22,"supported":14,"title":214,"abstract":215,"tags":216},[],[],[27],{"code":40,"score":22,"supported":14,"title":218,"abstract":219,"tags":220},[],[],[28],{"code":48,"score":22,"supported":14,"title":222,"abstract":223,"tags":224},[],[],[30],{"code":60,"score":22,"supported":14,"title":226,"abstract":227,"tags":228},[],[],[29],"缺少足够的主题证据",[231,232],"农业人工智能 病虫害监测 智慧农业 边缘计算","农业人工智能 病虫害监测","农业人工智能病虫害监测智慧农业边缘计算-2019","10.1016\u002Fj.compag.2026.112386",{"doi":234,"openalex_id":236,"authors":237,"venue":199,"cited_by_count":83,"oa_url":196,"card":266,"direction":270,"ingested_from":91},"W7212076527",[238,240,243,246,249,252,255,258,261,264],{"name":239,"orcid":9},"Guilherme Pires Silva de Almeida",{"name":241,"orcid":242},"Leonardo Nazário Silva dos Santos","https:\u002F\u002Forcid.org\u002F0000-0002-3951-3888",{"name":244,"orcid":245},"Ruy de Oliveira","https:\u002F\u002Forcid.org\u002F0009-0004-4502-8557",{"name":247,"orcid":248},"Marconi Batista Teixeira","https:\u002F\u002Forcid.org\u002F0000-0002-0152-256X",{"name":250,"orcid":251},"Mario De Oliveira","https:\u002F\u002Forcid.org\u002F0000-0002-3619-3989",{"name":253,"orcid":254},"Pablo da Costa Gontijo","https:\u002F\u002Forcid.org\u002F0000-0001-8173-0539",{"name":256,"orcid":257},"Adriano Soares de Oliveira Bailão","https:\u002F\u002Forcid.org\u002F0009-0008-0122-9639",{"name":259,"orcid":260},"Heyde Francielle do Carmo França","https:\u002F\u002Forcid.org\u002F0000-0002-2696-1083",{"name":262,"orcid":263},"Sergio Souza Novak","https:\u002F\u002Forcid.org\u002F0009-0007-6864-2332",{"name":265,"orcid":9},"Jéssica  Lauanda Stirle",{"tldr":267,"method":268,"finding":269,"direction":270,"opportunity":271},"提出AgroInsect边缘-云混合AI架构，实现农田害虫精准检测与地统计插值。","边缘端AI检测害虫，云端地统计插值，构建边缘-云协同架构。","该架构可精准检测害虫并空间插值，提升农田虫害监测覆盖与精度。","智慧农业 \u002F 农业物联网","可探索边缘检测模型轻量化与多源数据融合的虫害时空预测，降低部署成本。","2026-09-10T23:30:01.641871Z",{"id":274,"title":275,"url":276,"summary":277,"summary_zh":278,"content":9,"source_name":279,"source_url":276,"published_at":280,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":281,"score_detail":282,"sources":284,"tags":286,"primary_topic":289,"secondary_topics":290,"classification":291,"search_phrases":318,"slug":321,"view_count":83,"doi":322,"paper":323,"created_at":334},5588,"Certified Temporal Fusion of Artificial Intelligence Climate Analytics for Crop Failure Early Warning Under Sparse Agricultural Sensors: Error Bounds and Famine Relief Targeting","https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23545489","Sparse agricultural sensor networks across sub-Saharan Africa leave crop-failure early-warning systems with too little ground observation to support reliable famine relief targeting. The framework treats sensor sparsity as a missing-data problem with a known observation operator and propagates climate-analytics uncertainty through a temporal Kalman-Buys filter whose gain is constrained by a certified risk envelope. Using NASA POWER monthly reanalysis for Accra, Ghana (2000-2023, 312 observations), the analysis establishes a mean temperature of 26.521 C and a warming trend of 0.1622 C per decade, which calibrates the climate-analytics prior. The derived bounds show that relief-targeting error scales with the square root of sensor density and with the Lipschitz constant of the crop-response function, yielding an explicit sensor-placement rule for a target misclassification rate. The contribution is a certified, assumption-explicit bridge between climate analytics and famine relief targeting that states the conditions under which early warning is trustworthy.","撒哈拉以南非洲稀疏的农业传感器网络使得作物歉收预警系统缺乏足够的地面观测数据，难以支撑可靠的饥荒救援目标定位。该框架将传感器稀疏性视为具有已知观测算子的缺失数据问题，并通过时间Kalman-Buys滤波器传播气候分析不确定性，该滤波器的增益受经认证的风险包络约束。利用NASA POWER对加纳阿克拉的月度再分析数据（2000—2023年，312个观测值），分析确定了平均气温为26.521°C、每十年0.1622°C的增暖趋势，据此校准了气候分析先验。推导出的界限表明，救援目标定位误差随传感器密度的平方根以及作物响应函数的Lipschitz常数缩放，从而为目标误分类率给出了明确的传感器布设规则。该研究的贡献在于构建了一个经认证的、假设明确的气候分析与饥荒救援目标定位之间的桥梁，阐明了早期预警可信赖的条件。","Zenodo (CERN European Organization for Nuclear Research)","2026-10-21T00:00:00Z",74,{"impact":20,"substance":106,"depth":19,"authority":20,"freshness":13,"relevant":22,"comment":283},"该研究提出一种带误差界认证的时序融合方法，用于稀疏传感器条件下的作物歉收预警与饥荒救援瞄准，方法新颖、结论明确，但案例局限于加纳阿克拉，产业影响有限，适合作为技术前沿条目收录。",[285],{"name":279,"url":276},[27,28,287,30,288],"粮食安全","气候分析",{"code":40,"slug":120,"name":121,"path":122,"parent_code":123},[],{"status":125,"locked":14,"version":35,"method":36,"evidence":292},{"candidates":293,"agricultural_evidence":42},[294,299,305,310,314],{"code":40,"score":41,"supported":42,"title":295,"abstract":297,"tags":298},[296],"artificial intelligence",[],[28],{"code":60,"score":300,"supported":42,"title":301,"abstract":302,"tags":304},3,[],[303],"农业传感",[],{"code":306,"score":22,"supported":14,"title":307,"abstract":308,"tags":309},"A8",[],[],[287],{"code":55,"score":22,"supported":14,"title":311,"abstract":312,"tags":313},[],[],[27],{"code":48,"score":22,"supported":14,"title":315,"abstract":316,"tags":317},[],[],[30],[319,320],"NASA POWER 加纳 气候分析","作物歉收 早期预警 传感器","NASAPOWER加纳气候分析-5588","10.5281\u002Fzenodo.23545489",{"doi":322,"openalex_id":324,"authors":325,"venue":279,"cited_by_count":83,"oa_url":276,"card":328,"direction":333,"ingested_from":91},"W7223111766",[326],{"name":327,"orcid":9},"Fatima-Zahra El Khattabi",{"tldr":329,"method":330,"finding":331,"direction":89,"opportunity":332},"提出带误差界的时间融合框架，在稀疏传感器下用气候分析预警作物歉收。","卡尔曼-布西滤波+NASA POWER再分析数据，风险包络约束增益。","救济瞄准误差随传感器密度平方根和作物响应Lipschitz常数缩放，给出布点规则。","可将该认证误差界框架扩展到多作物多区域，结合卫星遥感弥补地面传感器稀疏。","农业人工智能与决策模型","2026-10-22T11:21:02.682497Z",{"id":336,"title":337,"url":338,"summary":339,"summary_zh":340,"content":9,"source_name":279,"source_url":338,"published_at":280,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":341,"score_detail":342,"sources":346,"tags":348,"primary_topic":351,"secondary_topics":352,"classification":357,"search_phrases":374,"slug":377,"view_count":83,"doi":378,"paper":379,"created_at":389},5570,"Evaluation of Counterfactual Explanation Robustness for Computer Vision Agriculture Yield-Forecast Models Under Satellite Cloud-Mask Gaps: An Interpretability Reliability Bound for Cooperative Storage Decisions","https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23545079","Counterfactual explanations are increasingly proposed as the interface through which cooperative storage managers interrogate computer-vision yield-forecast models, yet their reliability under the satellite cloud-mask gaps that routinely afflict agricultural remote sensing has not been bounded. The bound is derived from a Lipschitz continuity assumption on the forecast model and a masking-operator perturbation analysis, and it is expressed as a function of the cloud-mask gap fraction, the model's local Lipschitz constant, and the counterfactual search tolerance. A replication-based evaluation of the before-after-control-impact estimator was conducted to characterise the operating characteristics of the design under a declared variance structure (between-cluster SD = 6.0, residual SD = 4.5, intraclass correlation = 0.64), with power estimated across 2000 design cells. The reliability bound degrades non-linearly as the gap fraction rises, and that the intraclass correlation of 0.64 implies that cooperative-level aggregation of counterfactual recourse is necessary to preserve explanation stability. The contribution is a decision-relevant reliability bound that storage cooperatives can use to set a maximum tolerable cloud-mask gap before counterfactual recourse is suspended.","反事实解释日益被提出作为合作式仓储管理者质询计算机视觉产量预测模型的接口，然而在农业遥感中经常出现的卫星云掩膜缺口条件下，其可靠性尚未得到界定。该界基于预测模型上的Lipschitz连续性假设和掩膜算子扰动分析推导得出，并表示为云掩膜缺口比例、模型局部Lipschitz常数和反事实搜索容差的函数。对前后对照影响估计量进行了基于重复的评估，以表征该设计在给定方差结构（组间标准差=6.0，残差标准差=4.5，组内相关系数=0.64）下的运行特征，并在2000个设计单元中估计了检验效能。可靠性界随缺口比例上升呈非线性退化，且组内相关系数0.64意味着需要对反事实救济进行合作社层面的聚合，以保持解释的稳定性。本文的贡献在于提供了一个与决策相关的可靠性界，仓储合作社可据此设定在暂停反事实救济之前可容忍的最大云掩膜缺口。",67,{"impact":21,"substance":343,"depth":344,"authority":20,"freshness":13,"relevant":22,"comment":345},20,17,"该论文提出反事实解释在卫星云掩膜缺口下的可靠性边界，方法新颖且结论可用于合作社决策，但属细分方法学研究，产业影响有限。",[347],{"name":279,"url":338},[27,28,349,30,350],"产量预测","可解释性",{"code":40,"slug":120,"name":121,"path":122,"parent_code":123},[353],{"code":48,"slug":354,"name":355,"path":356,"parent_code":123},"remote-sensing","遥感与农情监测","\u002Ftopic\u002Fdigital-agriculture\u002Fremote-sensing",{"status":125,"locked":14,"version":35,"method":36,"evidence":358},{"candidates":359,"agricultural_evidence":42},[360,366,370],{"code":40,"score":20,"supported":42,"title":361,"abstract":363,"tags":365},[362],"computer vision",[364],"计算机视觉",[28],{"code":48,"score":21,"supported":42,"title":367,"abstract":368,"tags":369},[],[30,52],[349,30],{"code":55,"score":22,"supported":14,"title":371,"abstract":372,"tags":373},[],[],[27],[375,376],"农业人工智能 产量预测 可解释性 智慧农业","农业人工智能 产量预测","农业人工智能产量预测可解释性智慧农业-5570","10.5281\u002Fzenodo.23545079",{"doi":378,"openalex_id":380,"authors":381,"venue":279,"cited_by_count":83,"oa_url":338,"card":384,"direction":89,"ingested_from":91},"W7223114002",[382],{"name":383,"orcid":9},"Kadiatou Barry",{"tldr":385,"method":386,"finding":387,"direction":89,"opportunity":388},"推导卫星云掩膜缺口下反事实解释可靠性的理论界，并评估其对合作社储粮决策的影响。","基于Lipschitz连续性假设与掩膜扰动分析推导可靠性界，并用2000个设计单","缺口比例上升使可靠性非线性下降；ICC=0.64表明需合作社层级聚合以保持解释稳定。","可研究不同作物区云掩膜统计特性下的自适应解释暂停阈值与聚合策略。","2026-10-22T11:20:39.240791Z",{"id":391,"title":392,"url":393,"summary":394,"summary_zh":395,"content":9,"source_name":396,"source_url":393,"published_at":280,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":201,"score_detail":397,"sources":399,"tags":401,"primary_topic":404,"secondary_topics":410,"classification":416,"search_phrases":471,"slug":474,"view_count":83,"doi":475,"paper":476,"created_at":499},5564,"Dynamic thermal imaging for crop water stress detection across phenological stages: Enhancing irrigation scheduling precision","https:\u002F\u002Fdoi.org\u002F10.14719\u002Fpst.17728","Thermal remote sensing and derived indices such as the crop water stress index (CWSI) can assess crop water status and support precision irrigation. However, most operational and artificial intelligence (AI)-based approaches rely on fixed, season-long thresholds to interpret canopy–air temperature differences and often overlook the influence of crop phenology. This review synthesises evidence from grapevine, wheat, maize, potato and several other crops on how developmental stage affects the non-water-stressed baseline (NWSB), CWSI thresholds and irrigation-scheduling algorithms. It covers CWSI theory and estimation practices; crop-specific, phenology-dependent behaviour of baselines and thresholds; a comparison of threshold-based, machine learning, Internet of Things (IoT), digital twin and Food and Agriculture Organization (FAO)-56 approaches with respect to crop phenology; and a conceptual framework for phenology-aware scheduling. Across the reviewed crops, baseline behaviour generally varied systematically across the growing season, the relationship between CWSI and crop water status generally strengthened after canopy closure and CWSI measured during reproductive stages was generally most closely related to yield. A quantitative audit of 56 AI-, thermal-imaging-, or CWSI-based studies found that 78.6 % either did not explicitly incorporate crop phenology or relied on day of year as a temporal proxy, whereas only 5.4 % explicitly encoded phenological stage. The proposed framework considers phenology explicitly in baseline estimation, threshold setting and data fusion or machine learning for irrigation scheduling; the framework will require multi-year, multi-site field validation before operational use. Future research should prioritise standardised phenology definitions and long-term, multi-location studies to establish and validate stage-specific parameters.","热红外遥感及其衍生的作物水分胁迫指数（crop water stress index, CWSI）等指标可评估作物水分状况并支持精准灌溉。然而，大多数业务化方法和基于人工智能（artificial intelligence, AI）的方法依赖固定的全生育期阈值来解释冠层—空气温差，且往往忽视作物物候的影响。本文综述了葡萄、小麦、玉米、马铃薯及其他若干作物在发育阶段如何影响非水分胁迫基线（non-water-stressed baseline, NWSB）、CWSI阈值和灌溉调度算法的相关证据。综述内容涵盖CWSI理论与估算实践；基线和阈值的作物特异性、物候依赖性行为；基于阈值、机器学习、物联网（Internet of Things, IoT）、数字孪生及联合国粮农组织（Food and Agriculture Organization, FAO）-56等方法在作物物候方面的比较；以及物候感知调度的概念框架。在所综述的作物中，基线行为在整个生长季通常呈系统性变化，CWSI与作物水分状况之间的关系在冠层封垄后总体增强，且在生殖阶段测得的CWSI与产量的关系总体最为密切。对56项基于AI、热成像或CWSI的研究进行的定量审查发现，78.6%的研究要么未明确纳入作物物候，要么依赖年积日作为时间代理变量，而仅5.4%的研究明确编码了物候阶段。所提出的框架在基线估算、阈值设定以及用于灌溉调度的数据融合或机器学习中明确考虑物候；该框架在投入业务应用前尚需多年、多站点的田间验证。未来研究应优先建立标准化的物候定义，并开展长期、多地点研究以建立和验证阶段特异性参数。","Plant Science Today",{"impact":19,"substance":106,"depth":19,"authority":20,"freshness":41,"relevant":22,"comment":398},"系统综述热成像与CWSI在物候期感知灌溉中的应用，含56项研究量化审计，方法学价值突出，但属综述类论文，影响偏学术层面。",[400],{"name":396,"url":393},[27,28,30,402,403],"作物水分胁迫","精准灌溉",{"code":405,"slug":406,"name":407,"path":408,"parent_code":409},"H3","irrigation","水资源与灌溉","\u002Ftopic\u002Fenvironment\u002Firrigation","H",[411,412],{"code":40,"slug":120,"name":121,"path":122,"parent_code":123},{"code":60,"slug":413,"name":414,"path":415,"parent_code":123},"agricultural-iot","物联网传感与边缘计算","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-iot",{"status":125,"locked":14,"version":35,"method":36,"evidence":417},{"candidates":418,"agricultural_evidence":42},[419,424,430,437,443,447,454,458,465],{"code":405,"score":135,"supported":42,"title":420,"abstract":421,"tags":423},[406],[422,406],"灌溉",[403],{"code":40,"score":13,"supported":42,"title":425,"abstract":426,"tags":429},[],[427,428,296],"机器学习","machine learning",[28],{"code":60,"score":41,"supported":42,"title":431,"abstract":432,"tags":436},[],[433,434,435],"物联网","internet of things","IoT",[],{"code":55,"score":49,"supported":42,"title":438,"abstract":439,"tags":442},[],[440,441],"数字孪生","digital twin",[27],{"code":48,"score":49,"supported":42,"title":444,"abstract":445,"tags":446},[],[30,52],[30],{"code":448,"score":94,"supported":42,"title":449,"abstract":450,"tags":453},"A2",[],[451,452],"小麦","wheat",[],{"code":114,"score":94,"supported":42,"title":455,"abstract":456,"tags":457},[],[112,115],[],{"code":459,"score":94,"supported":42,"title":460,"abstract":461,"tags":464},"A5",[],[462,463],"马铃薯","potato",[],{"code":466,"score":300,"supported":42,"title":467,"abstract":468,"tags":470},"B2",[],[469],"葡萄",[],[472,473],"热成像 作物水分胁迫","CWSI 灌溉调度","热成像作物水分胁迫-5564","10.14719\u002Fpst.17728",{"doi":475,"openalex_id":477,"authors":478,"venue":396,"cited_by_count":83,"oa_url":393,"card":494,"direction":270,"ingested_from":91},"W7223088921",[479,482,485,488,491],{"name":480,"orcid":481},"R M Jeya","https:\u002F\u002Forcid.org\u002F0009-0001-3621-8533",{"name":483,"orcid":484},"P. C. Prabu","https:\u002F\u002Forcid.org\u002F0000-0002-7237-9481",{"name":486,"orcid":487},"K P Ragunath","https:\u002F\u002Forcid.org\u002F0000-0002-7851-4437",{"name":489,"orcid":490},"P. Kannan","https:\u002F\u002Forcid.org\u002F0000-0001-7003-3535",{"name":492,"orcid":493},"P. M. Shanmugam","https:\u002F\u002Forcid.org\u002F0000-0002-6706-8419",{"tldr":495,"method":496,"finding":497,"direction":89,"opportunity":498},"综述物候期对热成像CWSI阈值的影响，提出物候感知灌溉调度框架。","综述葡萄、小麦、玉米等作物的热遥感CWSI研究，并审计56项AI\u002F热成像研究。","78.6%研究未显式纳入物候，仅5.4%编码物候期；CWSI与产量在生殖期关联最强。","可开展多站点多年物候期特异性CWSI阈值标定与标准化物候定义研究。","2026-10-22T11:20:24.526683Z",{"id":501,"title":502,"url":503,"summary":504,"summary_zh":505,"content":9,"source_name":506,"source_url":503,"published_at":507,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":104,"score_detail":508,"sources":510,"tags":512,"primary_topic":9,"secondary_topics":515,"classification":516,"search_phrases":544,"slug":547,"view_count":83,"doi":548,"paper":549,"created_at":572},5402,"AI‐Enabled RGB Light‐to‐Frequency Sensor for Nondestructive Salinity Stress Monitoring in Medicinal Plants","https:\u002F\u002Fdoi.org\u002F10.1002\u002Faisy.70561","Salinity stress constrains medicinal plant productivity, particularly in coastal and salt affected agroecosystems where rapid, deployable diagnostics are essential. Here, we develop an AI‐enabled RGB light‐to‐frequency proximal sensing platform integrated with Tiny Machine Learning (TinyML) for nondestructive, real‐time salinity discrimination. Plants are exposed to three salinity levels (0, 100, and 200 mM NaCl). RGB‐derived and CIELAB chromatic descriptors are engineered, normalized, and partitioned (75:25) into training and test datasets. ReliefF‐based feature ranking and systematic subset optimization are combined with supervised learning using fivefold cross‐validation, and interpretability is achieved through SHAP analysis. Biomass reductions under saline treatments confirm physiological perturbations underlying spectral variability. Subspace KNN achieved the highest predictive performance (96.28% validation; 95.55% independent test accuracy). Framing model selection as a multiobjective optimization problem identified an optimal 10‐feature subset. Under this constraint, a Wide Neural Network attained 0.904 accuracy with an ultracompact 0.017 MB model size and >246 000 observations s −1 inference speed, demonstrating full TinyML compatibility. SHAP attribution consistently highlighted green intensity, blue channel metrics, and chromatic ratios as dominant predictors. This work establishes a lightweight, interpretable Edge‐AI framework for scalable, real‐time salinity diagnostics in precision agriculture.","盐胁迫制约药用植物的生产力，尤其是在沿海及受盐影响的农业生态系统中，亟需可快速部署的诊断手段。本研究开发了一种基于人工智能的RGB光-频率近端传感平台，并集成微型机器学习（TinyML），用于无损、实时的盐胁迫判别。植物暴露于三个盐浓度水平（0、100和200 mM NaCl）。研究构建了RGB衍生描述符与CIELAB色度描述符，经归一化后按75:25划分为训练集与测试集。将基于ReliefF的特征排序与系统性特征子集优化相结合，采用五折交叉验证进行监督学习，并通过SHAP分析实现模型可解释性。盐处理下的生物量下降证实了光谱变异性背后的生理扰动。子空间KNN取得了最高的预测性能（验证集准确率96.28%；独立测试集准确率95.55%）。将模型选择构建为多目标优化问题，确定了最优的10特征子集。在该约束下，宽神经网络（Wide Neural Network）达到0.904的准确率，模型体积仅0.017 MB，推理速度超过246 000次观测\u002F秒，展现出完全的TinyML兼容性。SHAP归因一致表明，绿色强度、蓝色通道指标及色度比是主要预测因子。本研究为精准农业中可扩展的实时盐胁迫诊断建立了一个轻量、可解释的边缘人工智能框架。","Advanced Intelligent Systems","2026-10-19T00:00:00Z",{"impact":19,"substance":106,"depth":19,"authority":107,"freshness":41,"relevant":22,"comment":509},"该研究提出一种轻量、可解释的边缘AI盐胁迫诊断框架，方法新颖、数据扎实，对精准农业具有参考价值，但属于实验室阶段成果，产业影响有限。",[511],{"name":506,"url":503},[27,28,29,513,514],"盐胁迫监测","药用植物",[],{"status":34,"locked":14,"version":35,"method":36,"evidence":517},{"candidates":518,"agricultural_evidence":42,"reason":543},[519,525,529,533,539],{"code":520,"score":49,"supported":42,"title":521,"abstract":522,"tags":524},"B6",[],[514,523],"medicinal plant",[514],{"code":55,"score":49,"supported":42,"title":526,"abstract":527,"tags":528},[],[143,144],[27],{"code":40,"score":49,"supported":42,"title":530,"abstract":531,"tags":532},[],[427,428],[28],{"code":534,"score":300,"supported":42,"title":535,"abstract":536,"tags":538},"H6",[],[537],"农业生态",[],{"code":60,"score":22,"supported":14,"title":540,"abstract":541,"tags":542},[],[],[29],"多个产业证据相当，需要确认主归属",[545,546],"RGB 光频传感器 盐胁迫","TinyML 药用植物 无损监测","RGB光频传感器盐胁迫-5402","10.1002\u002Faisy.70561",{"doi":548,"openalex_id":550,"authors":551,"venue":506,"cited_by_count":83,"oa_url":503,"card":567,"direction":333,"ingested_from":91},"W7222914402",[552,555,558,561,564],{"name":553,"orcid":554},"Ahmad Ali","https:\u002F\u002Forcid.org\u002F0000-0001-5530-7374",{"name":556,"orcid":557},"Vinie Lee Silva-Alvarado","https:\u002F\u002Forcid.org\u002F0009-0000-5857-3248",{"name":559,"orcid":560},"Arman Heydari","https:\u002F\u002Forcid.org\u002F0009-0002-4705-0729",{"name":562,"orcid":563},"Sandra Sendra","https:\u002F\u002Forcid.org\u002F0000-0001-9556-9088",{"name":565,"orcid":566},"Jaime Lloret","https:\u002F\u002Forcid.org\u002F0000-0002-0862-0533",{"tldr":568,"method":569,"finding":570,"direction":270,"opportunity":571},"开发AI驱动的RGB光频传感器，实现药用植物盐胁迫无损实时监测。","RGB与CIELAB色度特征、ReliefF特征选择、SHAP解释及TinyML","Subspace KNN验证准确率96.28%，10特征Wide NN模型仅0.017MB且推理超2","可探索多作物多胁迫通用色度特征库，及田间光照变化下的模型鲁棒性与长期部署验证。","2026-10-20T23:30:35.809539Z"]