[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-3670":3,"related-3670":60},{"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,"search_phrases":24,"slug":27,"view_count":15,"doi":28,"paper":29,"created_at":59},3670,"Automated pavement marking retroreflectivity estimation via computer vision and machine learning","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.trip.2026.102283","Pavement marking retroreflectivity is essential for nighttime road safety but deteriorates due to environmental and traffic factors. Traditional assessment methods are costly and labor-intensive. This paper introduces a sensor-independent, artificial intelligence driven methodology for predicting pavement-marking retroreflectivity using just Red, Green, and Blue (RGB) smartphone videos. The artificial intelligence contribution is through combining gray level co-occurrence matrix (GLCM) based texture and contrast features with a physics-informed synthetic labeling model that imitates material deterioration and environmental influences, allowing for supervised learning in the absence of specialized sensors. Five machine learning models were trained using eight features derived from video-based image processing and physics-informed synthetic labels. XGBoost achieved the highest internal consistency with the synthetic labeling model (R 2 = 0.98 for yellow and 0.97 for white), demonstrating that machine learning algorithms can effectively approximate complex, physics-based relationships between retroreflectivity and standard RGB video features. These results establish proof-of-concept feasibility; field validation against ground-truth retroreflectivity measurements remains a necessary next step before operational deployment. Future work will validate the model using field measurements and expand the dataset across varied pavement types, weather conditions, and traffic environments to enhance generalizability.","路面标线逆反射性能对夜间道路安全至关重要，但会因环境和交通因素而衰减。传统评估方法成本高且劳动强度大。本文提出了一种与传感器无关、由人工智能驱动的方法，仅使用红、绿、蓝（RGB）智能手机视频即可预测路面标线逆反射性能。其人工智能贡献在于将基于灰度共生矩阵（GLCM）的纹理与对比度特征，与一个模拟材料劣化和环境影响的物理信息合成标注模型相结合，从而在缺乏专用传感器的情况下实现监督学习。研究使用由基于视频的图像处理和物理信息合成标签得到的八个特征训练了五种机器学习模型。XGBoost在与合成标注模型的一致性方面表现最佳（黄色R²=0.98，白色R²=0.97），表明机器学习算法能够有效逼近逆反射性能与标准RGB视频特征之间复杂的、基于物理的关系。这些结果确立了概念验证的可行性；在投入实际应用之前，仍需针对真实逆反射测量值开展现场验证。未来工作将利用现场测量验证模型，并扩展数据集以涵盖不同路面类型、天气条件和交通环境，从而增强泛化能力。",null,"Transportation Research Interdisciplinary Perspectives","2026-09-27T00:00:00Z","论文",10,false,0,{"impact":15,"substance":15,"depth":15,"authority":15,"freshness":15,"relevant":15,"comment":17},"该论文研究路面标线逆反射性能的计算机视觉估算方法，属于道路交通安全与交通基础设施领域，与三农、农业信息化、智慧农业、数字乡村等主题无关，不建议进入每日精选。",[19],{"name":10,"url":6},[21,22,23],"计算机视觉","机器学习","道路安全",[25,26],"计算机视觉 机器学习 道路安全","计算机视觉 机器学习","计算机视觉机器学习道路安全-3670","10.1016\u002Fj.trip.2026.102283",{"doi":28,"openalex_id":30,"authors":31,"venue":10,"cited_by_count":15,"oa_url":50,"card":51,"direction":57,"ingested_from":58},"W7214556100",[32,35,37,40,43,45,48],{"name":33,"orcid":34},"Mark Gervas Ngotonie","https:\u002F\u002Forcid.org\u002F0009-0003-2558-9034",{"name":36,"orcid":9},"Nana Kankam Gyimah",{"name":38,"orcid":39},"Gurcan Comert","https:\u002F\u002Forcid.org\u002F0000-0002-2373-5013",{"name":41,"orcid":42},"Judith Mwakalonge","https:\u002F\u002Forcid.org\u002F0000-0002-7497-6829",{"name":44,"orcid":9},"Said Siuhi",{"name":46,"orcid":47},"Akinbobola Jegede","https:\u002F\u002Forcid.org\u002F0009-0002-4015-6168",{"name":49,"orcid":9},"Ayobami Taiwo","https:\u002F\u002Fwww.sciencedirect.com\u002Fscience\u002Farticle\u002Fpii\u002FS2590198226004483\u002Fpdf",{"tldr":52,"method":53,"finding":54,"direction":55,"opportunity":56},"用手机RGB视频和机器学习预测路面标线逆反射率，无需专用传感器。","GLCM纹理特征+物理信息合成标签，训练五种机器学习模型。","XGBoost与合成标签模型一致性最高（R²=0.98黄\u002F0.97白），验证概念可行。","其他","需实地验证并扩展多路面、天气和交通场景数据集，提升模型泛化能力。","农业遥感与作物表型","openalex","2026-09-28T23:30:27.891115Z",{"total":61,"page":62,"page_size":61,"items":63},6,1,[64,114,160,200,236,264],{"id":65,"title":66,"url":67,"summary":68,"summary_zh":69,"content":9,"source_name":70,"source_url":67,"published_at":11,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":71,"score_detail":72,"sources":79,"tags":81,"search_phrases":86,"slug":89,"view_count":15,"doi":90,"paper":91,"created_at":113},3677,"A stocktake of opportunities and knowledge gaps in advancing digital agriculture in Quebec","https:\u002F\u002Fdoi.org\u002F10.1007\u002Fs43621-026-04697-2","Abstract Sustainable development has become a key priority for the agricultural sector, both globally and in Quebec. Digital agriculture, as an important component of sustainable agricultural systems, represents a promising pathway toward achieving sustainable development goals. However, despite the recognized benefits of digital agriculture, the opportunities, limitations, and research gaps associated with its contribution to sustainable development in Quebec remain insufficiently understood. This study applies a systematic tracking approach to identify applications of digital technologies in Quebec’s agricultural sector and evaluate their potential contributions to provincial sustainable development objectives while highlighting gaps and opportunities for their enhanced deployment. A total of 59 peer-reviewed publications and grey literature reports were identified and analyzed. Overall, 14 categories of digital technologies were documented within Quebec’s agricultural sector. Their representation in the scientific literature was primarily concentrated in three dominant categories: web\u002Fmobile applications and software (31%), machine learning and sensors (14%), and satellite technologies (13%). The identified digital technologies have significant potential to support sustainable development objectives in Quebec agriculture through improved pesticide monitoring, soil health assessment and conservation, fertilizer management optimization, water-use efficiency, and biodiversity conservation. Collectively, these technologies could be combined in 65 distinct ways to support sustainable development goals; however, only 54% of this potential is currently being exploited in Quebec. This highlights substantial opportunities for expanding the role of digital agriculture in addressing sustainability challenges. The study also reveals important knowledge gaps, particularly regarding the social dimensions influencing the diffusion and adoption of digital technologies in agriculture. A limited understanding of the factors affecting Quebec farmers’ adoption decisions remains a major barrier to maximizing the benefits of digital agriculture. Addressing these social and behavioral dimensions will be essential for accelerating the transition toward more sustainable, digitally enabled agricultural systems in Quebec.","摘要 可持续发展已成为全球及魁北克农业部门的重要优先事项。数字农业作为可持续农业系统的重要组成部分，是实现可持续发展目标的一条充满前景的路径。然而，尽管数字农业的益处已得到认可，但其对魁北克可持续发展所作贡献的机遇、局限及研究空白仍未被充分理解。本研究采用系统性追踪方法，识别数字技术在魁北克农业部门中的应用，评估其对省级可持续发展目标的潜在贡献，同时揭示其加强部署的空白与机遇。共识别并分析了59篇同行评审出版物和灰色文献报告。总体而言，魁北克农业部门中记录到14类数字技术。其在科学文献中的呈现主要集中在三个主导类别：网络\u002F移动应用程序与软件（31%）、机器学习与传感器（14%）以及卫星技术（13%）。所识别的数字技术通过改善农药监测、土壤健康评估与保护、肥料管理优化、水资源利用效率以及生物多样性保护，具有支持魁北克农业可持续发展目标的显著潜力。总体而言，这些技术可以65种不同方式组合以支持可持续发展目标；然而，魁北克目前仅开发利用了其中54%的潜力。这凸显了扩大数字农业在应对可持续性挑战方面作用的重大机遇。本研究还揭示了重要的知识空白，特别是影响数字技术在农业中扩散与采纳的社会维度方面。对影响魁北克农民采纳决策的因素了解有限，仍是最大化数字农业效益的主要障碍。解决这些社会与行为维度对于加速魁北克向更可持续、数字化赋能的农业系统转型至关重要。","Discover Sustainability",78,{"impact":73,"substance":74,"depth":75,"authority":76,"freshness":77,"relevant":62,"comment":78},16,22,18,13,9,"基于59篇文献的系统梳理，量化了魁北克数字农业技术类别与可持续目标支撑潜力，指出仅54%潜力被利用及农户采纳的社会维度研究缺口，对区域数字农业政策有参考价值。",[80],{"name":70,"url":67},[82,83,84,22,85],"数字农业","智慧农业","农业遥感","农业可持续发展",[87,88],"魁北克 数字农业","魁北克 农业技术 应用","魁北克数字农业-3677","10.1007\u002Fs43621-026-04697-2",{"doi":90,"openalex_id":92,"authors":93,"venue":70,"cited_by_count":15,"oa_url":67,"card":107,"direction":111,"ingested_from":58},"W7214562996",[94,96,99,102,104],{"name":95,"orcid":9},"Sambiani D. Y. Tindjiete",{"name":97,"orcid":98},"Terence Épule Épule","https:\u002F\u002Forcid.org\u002F0000-0002-5756-382X",{"name":100,"orcid":101},"Daniel Etongo","https:\u002F\u002Forcid.org\u002F0000-0002-8237-0843",{"name":103,"orcid":9},"Changhui Peng",{"name":105,"orcid":106},"Paul Célicourt","https:\u002F\u002Forcid.org\u002F0000-0001-9297-6593",{"tldr":108,"method":109,"finding":110,"direction":111,"opportunity":112},"系统盘点魁北克数字农业技术应用及其对可持续发展的贡献与知识空白。","系统追踪59篇同行评议与灰色文献，归类14类数字技术并评估组合潜力。","技术可65种组合支撑可持续发展目标，但仅54%潜力被利用，社会采纳因素研究不足。","数字乡村与农业信息化","可深入研究农户数字技术采纳的社会行为因素，填补魁北克及类似地区扩散机制的研究空白。","2026-09-28T23:30:36.987716Z",{"id":115,"title":116,"url":117,"summary":118,"summary_zh":119,"content":9,"source_name":120,"source_url":117,"published_at":121,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":122,"score_detail":123,"sources":128,"tags":130,"search_phrases":135,"slug":138,"view_count":15,"doi":139,"paper":140,"created_at":159},3676,"Assessing Land Use, Land Cover Changes, and Urbanization Impacts on Water Stress in West African Watersheds: A Systematic Methodological Review","https:\u002F\u002Fdoi.org\u002F10.3791\u002F72287","Water stress is one of the major environmental and socio-economic constraints in West Africa, where population growth, agricultural expansion, soil sealing, hydroclimatic variability, and rising pressure on surface water and groundwater rapidly transform watershed functioning. Evaluating the impacts of land use and land cover (LULC) changes and urbanization requires methodological approaches that link the spatial dynamics of landscapes to observed or simulated hydrological responses. This review provides a comparative and critical analysis of the main techniques used in the region: instrumented watersheds and in situ observations, controlled field and laboratory experiments, satellite remote sensing and multi-temporal analysis, groundwater-oriented methods, conceptual, semi-distributed, physically based, and integrated surface-subsurface hydrological modeling, statistical and machine-learning approaches, and hybrid methods coupling land-use and climate scenarios. The search and selection process followed a transparent, PRISMA-based protocol and yielded 51 included records, of which 35 are anchored in West Africa. Hydrological modeling driven by remote sensing dominates the corpus, accounting for 15 of the 35 West African records (43%), particularly with SWAT, ACRU, WaSiM, CEQUEAU, SHETRAN, HydroGeoSphere, and ParFlow-CLM; reported Nash-Sutcliffe efficiency (NSE), Kling-Gupta efficiency (KGE), and coefficient of determination generally range from about 0.6 to 0.9, which is satisfactory to very good against standard evaluation guidelines. Statistical and machine-learning studies account for 8 of 35 records, whereas urban- and groundwater-focused studies remain scarce. Moreover, 8 of 35 records originate from a single research network, so several regional conclusions rest on a narrow and partly non-independent evidence base. In situ networks remain essential for process understanding but are constrained by cost and low density; purely satellite-based approaches suffer from cloud cover and spectral confusion in subhumid zones. The most promising advances lie in multi-source, multi-scale frameworks combining targeted instrumentation, satellite time series, integrated hydrological modeling, territorial scenarios, systematic uncertainty analyses, and co-construction with water-management and planning actors.","水资源压力是西非主要的环境和社会经济制约因素之一。在该地区，人口增长、农业扩张、土壤封闭、水文气候变率以及地表水和地下水压力的不断上升，正迅速改变着流域功能。评估土地利用与土地覆盖（LULC）变化及城市化带来的影响，需要将景观的空间动态与观测或模拟的水文响应相联系的方法学途径。本综述对该地区所采用的主要技术进行了比较性和批判性分析：受控流域与实地观测、受控野外与实验室实验、卫星遥感与多时相分析、面向地下水的方法、概念性、半分布式、基于物理过程及地表-地下耦合的综合水文模型、统计与机器学习方法，以及耦合土地利用与气候情景的混合方法。检索与筛选过程遵循透明、基于PRISMA的协议，最终纳入51篇文献，其中35篇以西非为研究区。由遥感驱动的水文建模在文献主体中占主导地位，在35篇西非文献中占15篇（43%），尤以SWAT、ACRU、WaSiM、CEQUEAU、SHETRAN、HydroGeoSphere和ParFlow-CLM为主；所报告的纳什-萨特克利夫效率（NSE）、克林-古普塔效率（KGE）和决定系数通常约在0.6至0.9之间，对照标准评价指南属于满意至很好水平。统计与机器学习研究在35篇文献中占8篇，而聚焦城市和地下水的研究仍然稀少。此外，35篇文献中有8篇来自同一研究网络，因此若干区域性结论建立在狭窄且部分非独立的证据基础之上。实地观测网络对于过程理解仍然不可或缺，但受成本和低密度制约；纯卫星方法在半湿润区则受云覆盖和光谱混淆的影响。最有前景的进展在于多源、多尺度框架，其结合了针对性观测、卫星时间序列、综合水文建模、区域情景、系统不确定性分析，以及与水资源管理和规划主体的共同构建。","Journal of Visualized Experiments","2026-09-25T00:00:00Z",66,{"impact":124,"substance":125,"depth":126,"authority":76,"freshness":124,"relevant":62,"comment":127},8,20,17,"系统综述方法学扎实、数据翔实，但聚焦西非流域且与国内农业信息化关联较弱，仅具方法参考价值。",[129],{"name":120,"url":117},[22,131,132,133,134],"遥感监测","土地利用","水文模型","西非流域",[136,137],"西非流域 土地利用 水文模型","SWAT 遥感 水资源压力","西非流域土地利用水文模型-3676","10.3791\u002F72287",{"doi":139,"openalex_id":141,"authors":142,"venue":120,"cited_by_count":15,"oa_url":9,"card":154,"direction":57,"ingested_from":58},"W7214394296",[143,145,148,150,152],{"name":144,"orcid":9},"Valère-Carin Jofack Sokeng",{"name":146,"orcid":147},"Nakouana Timité","https:\u002F\u002Forcid.org\u002F0000-0002-3894-1450",{"name":149,"orcid":9},"Toto Marc Zahui",{"name":151,"orcid":9},"Kouamé Koffi Fernand",{"name":153,"orcid":9},"Koné Tiémoman",{"tldr":155,"method":156,"finding":157,"direction":57,"opportunity":158},"系统综述西非流域土地利用变化与城市化对水资源压力的评估方法。","PRISMA系统综述，纳入51篇文献，比较遥感、水文模型与机器学习等方法。","遥感驱动水文模型占主导（43%），城市与地下水研究稀缺，证据基础偏窄。","可构建多源多尺度框架，融合实地观测、遥感时序与集成水文模型，并开展不确定性分析与利益相关者协同。","2026-09-28T23:30:31.949583Z",{"id":161,"title":162,"url":163,"summary":164,"summary_zh":165,"content":9,"source_name":166,"source_url":163,"published_at":167,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":168,"score_detail":169,"sources":173,"tags":175,"search_phrases":179,"slug":182,"view_count":15,"doi":183,"paper":184,"created_at":199},3660,"Designing a framework for Sustainable Smart Agriculture with Mobile Robots and CNN","https:\u002F\u002Fdoi.org\u002F10.33423\u002F25102b43","Advances in artificial intelligence, robotics, and computer vision enable intelligent gardening systems that reduce manual monitoring. This study presents a smart gardening framework integrating a mobile robot with convolutional neural network (CNN) flower recognition. The robot captures garden images, while the CNN classifies flower species and provides plant-specific information for future care decisions. Using 50 training images and 50 independent testing images, the model achieved 92.0% training accuracy and 95.0% testing accuracy. The framework connects robotic image acquisition with automated visual recognition and provides a foundation for future irrigation, fertilization, and autonomous garden-management functions.","人工智能、机器人技术和计算机视觉的进步使得智能园艺系统能够减少人工监测。本研究提出了一种智能园艺框架，将移动机器人与卷积神经网络（CNN）花卉识别相结合。机器人采集花园图像，CNN对花卉种类进行分类，并提供植物特定信息，以供未来养护决策使用。使用50张训练图像和50张独立测试图像，该模型达到了92.0%的训练准确率和95.0%的测试准确率。该框架将机器人图像采集与自动视觉识别相连接，并为未来的灌溉、施肥和自主花园管理功能奠定了基础。","Journal of Strategic Innovation and Sustainability","2026-09-28T00:00:00Z",56,{"impact":124,"substance":170,"depth":171,"authority":171,"freshness":13,"relevant":62,"comment":172},14,12,"论文提出移动机器人与CNN花卉识别结合的智能园艺框架，测试准确率95%，但样本仅50张、规模偏小，属细分领域方法探索，公共价值有限。",[174],{"name":166,"url":163},[83,176,21,177,178],"农业人工智能","农业机器人","花卉识别",[180,181],"移动机器人 CNN 花卉识别","智能园艺 自动灌溉","移动机器人CNN花卉识别-3660","10.33423\u002F25102b43",{"doi":183,"openalex_id":185,"authors":186,"venue":166,"cited_by_count":15,"oa_url":9,"card":193,"direction":197,"ingested_from":58},"W7214523094",[187,189,191],{"name":188,"orcid":9},"Ting Zhang",{"name":190,"orcid":9},"Oanh Phan",{"name":192,"orcid":9},"Jiang Lu",{"tldr":194,"method":195,"finding":196,"direction":197,"opportunity":198},"提出移动机器人结合CNN识别花卉的智能园艺框架，测试准确率达95%。","移动机器人采集图像，CNN用50张训练和50张测试图像分类花卉。","CNN花卉识别测试准确率95%，为自动灌溉施肥等管理奠定基础。","智慧农业 \u002F 农业物联网","可扩展至多作物识别与实时决策，并融合物联网实现闭环精准管理。","2026-09-28T23:30:08.100458Z",{"id":201,"title":202,"url":203,"summary":204,"summary_zh":205,"content":9,"source_name":206,"source_url":203,"published_at":121,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":207,"score_detail":208,"sources":211,"tags":213,"search_phrases":216,"slug":219,"view_count":15,"doi":220,"paper":221,"created_at":235},3626,"A Hybrid Model for Crop Yield Prediction Using Recurrent Neural Networks and Explainable Artificial Intelligence","https:\u002F\u002Fdoi.org\u002F10.22194\u002Fjgias\u002F27.2061","Accurate maize yield prediction is essential for ensuring food security and supporting agricultural planning in Kenya. However, the changes in climate and severe weather are posing more challenges to the stability of yield and food security. While advanced machine learning models, such as Long Short–Term Memory (LSTM) networks, offer strong predictive performance, their inherent \"black–box\" nature limits their practical adoption by farmers and policymakers who require interpretable and trustworthy insights. This study developed a transparent predictive model by integrating LSTM with Explainable Artificial Intelligence (XAI). Using county–level maize yield data (2012–2023), the hybrid model was compared against Random Forest and Gradient Boosting baselines. The LSTM model achieved superior performance (R² = 0.8551, RMSE = 0.3715, MAE = 0.2705), compared to Random Forest (R2 = 0.8441, RMSE= 0.3851 and MAE= 0.2856) and Gradient Boosting (R2 = 0.7925, RMSE = 0.4443 and MAE = 0.3234) baselines. XAI analysis, using SHapley Additive explanations (SHAP) and Local Interpretable Model–agnostic Explanations (LIME), identified longitude, latitude, and annual rainfall as key predictors. The model maintained high accuracy while improving interpretability, increasing transparency and trust. This provides actionable insights for farmers and policymakers, supporting evidence–based planning and enhancing resilience in smallholder systems. Keywords: Crop yield prediction, long short–term memory (LSTM), explainable artificial intelligence, precision agriculture, machine learning.","准确的玉米产量预测对于保障肯尼亚粮食安全和支持农业规划至关重要。然而，气候变化和极端天气对产量稳定性和粮食安全构成了更多挑战。尽管长短期记忆（LSTM）网络等先进机器学习模型具有强大的预测性能，但其固有的“黑箱”特性限制了农民和政策制定者的实际采用，因为他们需要可解释且可信的洞见。本研究通过将LSTM与可解释人工智能（XAI）相结合，开发了一个透明的预测模型。利用县级玉米产量数据（2012—2023年），将该混合模型与随机森林和梯度提升基线模型进行了比较。LSTM模型取得了更优的性能（R² = 0.8551，RMSE = 0.3715，MAE = 0.2705），优于随机森林（R2 = 0.8441，RMSE= 0.3851，MAE= 0.2856）和梯度提升（R2 = 0.7925，RMSE = 0.4443，MAE = 0.3234）基线模型。使用SHapley加法解释（SHAP）和局部可解释模型无关解释（LIME）进行的XAI分析，确定了经度、纬度和年降雨量是关键预测因子。该模型在保持高精度的同时提高了可解释性，增强了透明度和信任。这为农民和政策制定者提供了可操作的洞见，支持基于证据的规划，并增强小农系统的韧性。关键词：作物产量预测，长短期记忆（LSTM），可解释人工智能，精准农业，机器学习。","Journal of Global Innovations in Agricultural Sciences",71,{"impact":171,"substance":209,"depth":126,"authority":76,"freshness":124,"relevant":62,"comment":210},21,"将LSTM与可解释AI结合用于肯尼亚县级玉米产量预测，方法新颖、指标详实，对智慧农业有参考价值，但属境外区域研究，公共影响有限。",[212],{"name":206,"url":203},[83,176,22,214,215],"可解释AI","玉米产量预测",[217,218],"LSTM 玉米产量预测","SHAP LIME 农业模型","LSTM玉米产量预测-3626","10.22194\u002Fjgias\u002F27.2061",{"doi":220,"openalex_id":222,"authors":223,"venue":206,"cited_by_count":15,"oa_url":203,"card":229,"direction":233,"ingested_from":58},"W7214446524",[224,226],{"name":225,"orcid":9},"Stephen Gitau Ndung’u",{"name":227,"orcid":228},"Consolata Gakii","https:\u002F\u002Forcid.org\u002F0000-0003-2783-9992",{"tldr":230,"method":231,"finding":232,"direction":233,"opportunity":234},"用LSTM结合可解释AI预测肯尼亚玉米产量，兼顾精度与透明度。","基于2012-2023县级玉米产量数据，LSTM融合SHAP与LIME，对比随机","LSTM精度最高（R²=0.8551），经度、纬度和年降雨量是主要预测因子。","农业人工智能与决策模型","可探索将可解释AI与遥感、气象多源数据融合，提升小农户区域产量预测的可信度与推广性。","2026-09-27T23:31:22.771605Z",{"id":237,"title":238,"url":239,"summary":240,"summary_zh":9,"content":9,"source_name":241,"source_url":9,"published_at":242,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":71,"score_detail":243,"sources":245,"tags":247,"search_phrases":251,"slug":254,"view_count":15,"doi":9,"paper":255,"created_at":263},3602,"Advancing pig lameness detection with a multi-task framework: Leveraging multi-view 3D pose estimation and wavelet convolution（多任务框架推进生猪跛行检测：利用多视角3D姿态估计与小波卷积）","https:\u002F\u002Fm2.mtmt.hu\u002Fapi\u002Fpublication\u002F36981064?&labelLang=hun","Xiang Li、Wang Haidong、Hu Zixuan、Norton Tomas、Jiang Tian、Xue Yueju等在Biosystems Engineering 1537-5110 Vol. 263 Paper: 104393（2026）发表。研究针对生猪跛行检测提出PoseGait-MT多任务分类框架，可同时检测跛行严重程度与受影响肢体。框架从多视角2D视频重建3D骨骼以缓解遮挡与视角依赖问题；提取3D步态时空特征（空间跟踪距离sTRK、头部上下振幅HBA、关节屈曲角JFA），并结合小波卷积（WTConv）增强低频特征提取、抑制高频噪声。实验结果显示：PoseGait-MT在5折交叉验证下对跛行严重程度分类平均准确率94.7%、受影响肢体识别95.7%；独立测试集上分别89%、91.4%。","Biosystems Engineering Vol. 263 2026-09-20","2026-09-20T00:00:00Z",{"impact":75,"substance":74,"depth":75,"authority":170,"freshness":61,"relevant":62,"comment":244},"核心期刊论文，方法新颖且实验数据扎实，对智慧养殖中的动物健康监测有实质参考价值，值得进入每日精选。",[246],{"name":241,"url":239},[83,21,248,249,250],"生猪养殖","3D姿态估计","动物健康监测",[252,253],"PoseGait-MT 生猪跛行检测","多视角3D姿态估计 生猪","PoseGait-MT生猪跛行检测-3602",{"doi":9,"openalex_id":9,"authors":256,"venue":9,"cited_by_count":15,"oa_url":9,"card":257,"direction":233,"ingested_from":262},[],{"tldr":258,"method":259,"finding":260,"direction":233,"opportunity":261},"提出PoseGait-MT多任务框架，用多视角3D姿态与小波卷积同时检测生猪跛行程度和患肢。","多视角2D视频重建3D骨骼，提取步态时空特征，结合小波卷积WTConv。","5折交叉验证跛行程度准确率94.7%、患肢识别95.7%，独立测试集为89%和91.4%。","可探索轻量化实时部署与跨农场泛化，并融合多模态数据提升早期跛行预警能力。","agent","2026-09-27T00:05:18.483996Z",{"id":265,"title":266,"url":267,"summary":268,"summary_zh":269,"content":9,"source_name":270,"source_url":267,"published_at":121,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":271,"score_detail":272,"sources":274,"tags":276,"search_phrases":280,"slug":283,"view_count":15,"doi":284,"paper":285,"created_at":299},3568,"Interpretable multi-index framework for extracting dry-season built-up areas: benchmarking machine learning with Sentinel-2","https:\u002F\u002Fdoi.org\u002F10.4995\u002Fraet.2027.25885","Spectral similarity between built-up surfaces and exposed dry soil significantly constrains built-up area extraction during dry seasons. Although machine-learning techniques can address this challenge by modeling complex spectral relationships, they generally require sufficiently large and representative labeled training datasets. This study therefore investigates a less data-demanding, rule-based multi-index approach for dry-season built-up mapping, aiming to reduce reliance on large labeled training datasets while maintaining effective classification performance. Sentinel-2A imagery acquired in April 2024 was analyzed, and temporal validation was conducted for 2020, 2022, and 2024. Among the tested combinations, the Built-up Area Extraction Index (BAEI), Dry BareSoil Index (DBSI), and Normalized Difference Vegetation Index (NDVI) achieved the highest performance, yielding 95 % overall accuracy and a Kappa coefficient of 0.89. This represents a substantial improvement over singleindex BAEI (80 % overall accuracy; Kappa 0.55), increasing the built-up user’s accuracy from 61 % to 92 %. Multitemporal validation confirmed that the optimized BAEI–DBSI–NDVI decision rules remained stable across the 2020–2024 dry-season images without recalibration, achieving 93–96 % overall accuracy and a built-up F1-score of 84–93 %. Spatial analysis demonstrated robust performance across peripheral and agricultural zones, with moderate variability in dense urban cores due to spectral heterogeneity. To evaluate the operational robustness of the proposed framework, it was benchmarked against machine-learning classifiers, specifically Support Vector Machine and Random Forest (RF). Under the specific conditions of this study, the proposed rule-based method (95 % accuracy) marginally outperformed both SVM (93 %) and RF (92 %), while offering greater transparency and reducing dependency on large, labeled training datasets. Furthermore, feature importance analysis confirmed the critical role of these selected indices in resolving spectral confusion. These findings suggest that the proposed framework offers a transparent and computationally efficient approach for dry-season urban monitoring in tropical coastal environments similar to Visakhapatnam.","建成表面与裸露干土之间的光谱相似性显著制约了旱季建成区提取。尽管机器学习技术可通过建模复杂光谱关系来应对这一挑战，但其通常需要足够大且具有代表性的标记训练数据集。因此，本研究探讨了一种对数据需求较低、基于规则的多指数方法用于旱季建成区制图，旨在减少对大规模标记训练数据集的依赖，同时保持有效的分类性能。研究分析了2024年4月获取的Sentinel-2A影像，并对2020年、2022年和2024年进行了时间验证。在测试的组合中，建成区提取指数（BAEI）、干裸土指数（DBSI）和归一化差异植被指数（NDVI）表现最佳，总体精度达95%，Kappa系数为0.89。相较于单一指数BAEI（总体精度80%；Kappa 0.55），这一结果有显著提升，建成区用户精度从61%提高至92%。多时相验证证实，优化后的BAEI–DBSI–NDVI决策规则在2020—2024年旱季影像上无需重新校准即可保持稳定，总体精度达93%—96%，建成区F1分数为84%—93%。空间分析表明，该方法在外围和农业区域表现稳健，而在密集城市核心区由于光谱异质性存在中等程度变异。为评估所提框架的业务化稳健性，将其与机器学习分类器进行了基准比较，具体为支持向量机（SVM）和随机森林（RF）。在本研究的特定条件下，所提出的基于规则的方法（95%精度）略优于SVM（93%）和RF（92%），同时具有更高的透明性并减少了对大规模标记训练数据集的依赖。此外，特征重要性分析证实了所选指数在解决光谱混淆方面的关键作用。这些发现表明，所提框架为类似维沙卡帕特南的热带沿海环境旱季城市监测提供了一种透明且计算高效的方法。","Revista de teledetección: Revista de la Asociación Española de Teledetección",72,{"impact":171,"substance":209,"depth":126,"authority":76,"freshness":77,"relevant":62,"comment":273},"该研究提出可解释的多指数规则框架，在旱季建成区提取上以更少标注数据达到95%精度并优于SVM\u002FRF，方法新颖、验证充分，对农业遥感与乡村土地利用监测有参考价值。",[275],{"name":270,"url":267},[277,22,278,131,279],"Sentinel-2","NDVI","建成区提取",[281,282],"Sentinel-2 建成区提取","BAEI DBSI NDVI 旱季","Sentinel-2建成区提取-3568","10.4995\u002Fraet.2027.25885",{"doi":284,"openalex_id":286,"authors":287,"venue":270,"cited_by_count":15,"oa_url":267,"card":294,"direction":233,"ingested_from":58},"W7214385607",[288,291],{"name":289,"orcid":290},"Sarah Sejari","https:\u002F\u002Forcid.org\u002F0009-0002-7300-9084",{"name":292,"orcid":293},"Vazeer Mahammood","https:\u002F\u002Forcid.org\u002F0000-0003-1667-5232",{"tldr":295,"method":296,"finding":297,"direction":57,"opportunity":298},"提出可解释多指数规则框架，用Sentinel-2提取旱季建成区，性能优于机器学习。","Sentinel-2A影像，BAEI、DBSI、NDVI组合规则，对比SVM与随","BAEI-DBSI-NDVI规则达95%精度、Kappa 0.89，优于单指数及SVM\u002FRF且无需重","可探索该规则框架向其他气候带与传感器迁移，并结合少量样本的半监督学习提升城市核心区精度。","2026-09-26T23:30:53.200163Z"]