[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-3185":3,"related-3185":70},{"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,"search_phrases":32,"slug":35,"view_count":36,"doi":37,"paper":38,"created_at":69},3185,"Generating Annual 10 m Land Cover Maps for 37 Chinese Metropolises Using Google Satellite Embeddings","https:\u002F\u002Fdoi.org\u002F10.3390\u002Frs18183233","Accurate urban land cover information is essential for monitoring urbanization and environmental change, yet the complexity and heterogeneity of urban landscapes remain challenging for high-resolution remote sensing classification. This study used Google Satellite Embeddings as the core feature input, combined with ensemble learning and spatiotemporal post-processing, to generate annual 10 m land cover maps for 37 Chinese metropolises from 2017 to 2024. Results showed that the weighted ensemble model achieved an overall accuracy (OA) of 81.99%, 2.13% higher than the best individual model, while spatiotemporal post-processing further increased OA to 82.59%. Compared with existing land cover products such as FROM-GLC10, the proposed product achieved 4.20% higher OA. The ablation experiment showed that the embedding representation was the primary source of performance improvement. Under the same ensemble learning framework, Satellite Embedding features outperformed conventional remote sensing features by 6.54% in OA. SHapley Additive exPlanations (SHAP) analysis identified A16, A36, A51, A61, and A22 as the key embedding dimensions, while Pearson correlation analysis further revealed associations between the key dimensions and spectral, radar, texture, and topographic information. Change analysis indicated concurrent urban expansion, ecological recovery, and agricultural land contraction from 2017 to 2024, with cropland-to-forest and cropland-to-impervious-surface conversions being the most prominent transition pathways. Overall, Satellite Embeddings provide effective feature representations for annual land cover mapping and long-term change analysis in complex metropolitan environments.","准确的城市土地覆盖信息对于监测城市化与环境变化至关重要，然而城市景观的复杂性和异质性仍对高分辨率遥感分类构成挑战。本研究以Google Satellite Embeddings作为核心特征输入，结合集成学习与时空后处理，生成了2017—2024年中国37个大城市逐年10 m土地覆盖图。结果表明，加权集成模型的总体精度（OA）达到81.99%，比最优单一模型高2.13%，时空后处理进一步将OA提升至82.59%。与FROM-GLC10等现有土地覆盖产品相比，所提产品的OA高出4.20%。消融实验表明，嵌入表征是性能提升的主要来源。在相同集成学习框架下，Satellite Embedding特征的OA比传统遥感特征高6.54%。SHapley Additive exPlanations（SHAP）分析识别出A16、A36、A51、A61和A22为关键嵌入维度，Pearson相关分析进一步揭示了关键维度与光谱、雷达、纹理和地形信息之间的关联。变化分析表明，2017—2024年间城市扩张、生态恢复与农用地收缩同时发生，其中耕地转为林地和耕地转为不透水面的转换路径最为突出。总体而言，Satellite Embeddings为复杂大城市环境中的逐年土地覆盖制图和长期变化分析提供了有效的特征表征。",null,"Remote Sensing","2026-09-20T00:00:00Z","论文",10,false,81,{"impact":17,"substance":18,"depth":19,"authority":20,"freshness":21,"relevant":22,"comment":23},18,22,19,14,8,1,"方法新颖、数据规模大且结论可靠，对城市扩张与耕地变化监测有实质参考价值，但属学术论文而非政策或产业事件，适合作为专业精选。",[25],{"name":10,"url":6},[27,28,29,30,31],"遥感","城市扩张","土地覆盖","Google卫星嵌入","耕地变化",[33,34],"Google Satellite Embeddings 土地覆盖","中国大都市 10米土地覆盖制图","GoogleSatelliteEmbeddings土地覆盖-3185",0,"10.3390\u002Frs18183233",{"doi":37,"openalex_id":39,"authors":40,"venue":10,"cited_by_count":36,"oa_url":6,"card":62,"direction":66,"ingested_from":68},"W7213897500",[41,44,47,50,52,54,56,59],{"name":42,"orcid":43},"Yu Wang","https:\u002F\u002Forcid.org\u002F0000-0002-1825-1241",{"name":45,"orcid":46},"Han Liu","https:\u002F\u002Forcid.org\u002F0000-0002-9386-2464",{"name":48,"orcid":49},"Li Wang","https:\u002F\u002Forcid.org\u002F0000-0001-5538-4337",{"name":51,"orcid":9},"Lingling Sang",{"name":53,"orcid":9},"Lili Wang",{"name":55,"orcid":9},"Caisheng Zhao",{"name":57,"orcid":58},"Tengyun Hu","https:\u002F\u002Forcid.org\u002F0009-0001-5514-2167",{"name":60,"orcid":61},"Xuecao Li","https:\u002F\u002Forcid.org\u002F0000-0002-6942-0746",{"tldr":63,"method":64,"finding":65,"direction":66,"opportunity":67},"利用谷歌卫星嵌入生成中国37个大都市2017-2024年10米年度土地覆盖图。","谷歌卫星嵌入特征+集成学习+时空后处理，对比FROM-GLC10并做消融与SHA","集成模型总体精度82.59%，嵌入特征比传统特征高6.54%，揭示城市扩张与生态恢复并存。","农业遥感与作物表型","可探索卫星嵌入在耕地变化监测与农业用地精细分类中的迁移能力及跨城市泛化性。","openalex","2026-09-22T23:30:25.425152Z",{"total":71,"page":22,"page_size":71,"items":72},6,[73,111,160,202,232,258],{"id":74,"title":75,"url":76,"summary":77,"summary_zh":78,"content":9,"source_name":79,"source_url":76,"published_at":80,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":81,"score_detail":82,"sources":86,"tags":88,"search_phrases":92,"slug":95,"view_count":36,"doi":96,"paper":97,"created_at":110},3196,"Geospatial and machine learning techniques for spatiotemporal analyses of urban dynamics in the Coimbatore city, South India","https:\u002F\u002Fdoi.org\u002F10.3846\u002Fjeelm.2026.28264","Rapid urbanization has drastically changed the land use and environmental conditions in Indian cities and need to be monitored continuously for sustainable urban planning. The study used Landsat satellite images of the years 2001, 2013 and 2023 to examine the spatiotemporal urban dynamics of the Coimbatore city of South India. The impacts of urban sprawl on the environment were evaluated from Land Use\u002F Land Cover (LULC), Land Surface Temperature (LST) and spectral indices (NDVI, NDWI and NDBI). The built-up land increased by 1.21% (2001–2013) and 1.67% (2013–2023) and agricultural land decreased by 1.62% (2013–2023). The LULC classification had an Overall Accuracy of 95.76% with a Kappa coefficient of 0.95. The Kappa coefficient of 0.96 indicates that the ANN-CA model has high predictive reliability in predicting the LULC scenario in 2031. The results show that the continued expansion of cities leads to an increase in land surface temperature and a decrease in vegetation cover. This has major implications for sustainable urban development and contributes to SDG 11 and SDG 13.","快速城市化极大地改变了印度城市的土地利用和环境状况，需要持续监测以支持可持续城市规划。本研究利用2001年、2013年和2023年的Landsat卫星影像，考察了印度南部哥印拜陀市的时空城市动态。研究从土地利用\u002F土地覆盖（LULC）、地表温度（LST）和光谱指数（NDVI、NDWI和NDBI）方面评估了城市蔓延对环境的影响。建设用地在2001—2013年间增加了1.21%，在2013—2023年间增加了1.67%；农业用地在2013—2023年间减少了1.62%。LULC分类的总体精度为95.76%，Kappa系数为0.95。Kappa系数0.96表明，ANN-CA模型在预测2031年LULC情景方面具有较高的预测可靠性。结果表明，城市持续扩张导致地表温度升高和植被覆盖减少。这对可持续城市发展具有重要影响，并有助于实现可持续发展目标11和可持续发展目标13。","Journal of Environmental Engineering and Landscape Management","2026-09-21T00:00:00Z",62,{"impact":21,"substance":17,"depth":83,"authority":84,"freshness":21,"relevant":22,"comment":85},15,13,"基于Landsat多时相遥感与ANN-CA模型的印度城市扩张研究，方法规范、数据翔实，对农业用地变化与遥感监测有参考价值，但属境外区域案例，公共影响有限。",[87],{"name":79,"url":76},[89,27,90,91,28],"可持续发展","土地利用","地表温度",[93,94],"Coimbatore 城市扩张 遥感","LULC LST NDVI 印度城市","Coimbatore城市扩张遥感-3196","10.3846\u002Fjeelm.2026.28264",{"doi":96,"openalex_id":98,"authors":99,"venue":79,"cited_by_count":36,"oa_url":76,"card":104,"direction":109,"ingested_from":68},"W7213862636",[100,102],{"name":101,"orcid":9},"Nagamani Singamuthu",{"name":103,"orcid":9},"Elangovan Krishnan",{"tldr":105,"method":106,"finding":107,"direction":66,"opportunity":108},"基于Landsat影像与机器学习分析印度哥印拜陀市2001-2023年城市扩张及其环境效应。","Landsat影像、LULC分类、NDVI\u002FNDWI\u002FNDBI指数、ANN-CA","建设用地增加、农地减少，城市扩张导致地表温度上升、植被覆盖下降。","可结合多源遥感与深度学习提升城市扩张预测精度，并探究其对周边农业用地的长期影响。","农业人工智能与决策模型","2026-09-22T23:30:43.496849Z",{"id":112,"title":113,"url":114,"summary":115,"summary_zh":116,"content":9,"source_name":117,"source_url":114,"published_at":118,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":15,"score_detail":119,"sources":122,"tags":124,"search_phrases":128,"slug":131,"view_count":36,"doi":132,"paper":133,"created_at":159},2790,"High-resolution land cover mapping from coarse labels via a noisy label learning-guided cross-scale framework","https:\u002F\u002Fdoi.org\u002F10.1080\u002F15481603.2026.2726002","High-resolution remote sensing images (HRSIs) provide essential data support for land cover mapping, where deep learning has shown great promise. However, deep learning-based methods rely on abundant high-quality annotations, while low-resolution coarse labels are difficult to use directly in HRSIs training. In this paper, a novel noisy label learning-guided cross-scale framework (NL-CSF) is proposed, which is designed to achieve high-resolution land cover mapping from coarse labels. First, a spectral-based label mask filtering strategy is developed to preliminarily optimize coarse labels. Then, an adaptive noise evaluation scheme is introduced that assigns loss weights based on the noise differences between image and label patches in the training set. Finally, we design a cross-scale transfer Transformer (CSTT) model based on the vision Transformer (ViT) architecture, and the training process is guided by a noise-weighted loss function. Two cross-scale datasets are utilized to evaluate the performance of NL-CSF in multiple spatial scale differences (10 m to 3 m, 3 m to 0.5 m, and 10 m to 0.5 m). Experimental results demonstrate that NL-CSF improves overall accuracy (OA) by at least 7%, 6%, and 4% across the three cross-scale tasks in the first dataset, and by at least 2%, 9%, and 4% in the second dataset, respectively, compared with existing methods. Furthermore, the proposed framework is applied to cross-scale mapping across Jianye District of Nanjing (urban), Sheyang County of Yancheng (agricultural), and the Yellow River Delta of Dongying (wetland), leveraging a low-resolution land cover product and high-resolution PlanetScope images to generate more precise land cover maps. These results demonstrate the effectiveness of the proposed framework in mitigating the impact of noisy coarse labels and generating reliable high-resolution land cover maps.","高分辨率遥感影像(high-resolution remote sensing images, HRSIs)为土地覆盖制图提供了重要的数据支撑，深度学习在此领域展现出巨大潜力。然而，基于深度学习的方法依赖于大量高质量标注，而低分辨率粗标签难以直接用于高分辨率遥感影像训练。本文提出了一种新的噪声标签学习引导的跨尺度框架(noisy label learning-guided cross-scale framework, NL-CSF)，旨在从粗标签实现高分辨率土地覆盖制图。首先，提出了一种基于光谱的标签掩膜过滤策略，对粗标签进行初步优化。然后，引入了一种自适应噪声评估方案，根据训练集中影像块与标签块之间的噪声差异分配损失权重。最后，基于视觉Transformer(vision Transformer, ViT)架构设计了跨尺度迁移Transformer(cross-scale transfer Transformer, CSTT)模型，并以噪声加权损失函数引导训练过程。利用两个跨尺度数据集评估NL-CSF在多种空间尺度差异(10 m至3 m、3 m至0.5 m、10 m至0.5 m)下的性能。实验结果表明，与现有方法相比，NL-CSF在第一个数据集的三个跨尺度任务中总体精度(overall accuracy, OA)分别至少提升7%、6%和4%，在第二个数据集中分别至少提升2%、9%和4%。此外，将所提框架应用于南京建邺区(城市)、盐城射阳县(农业)和东营黄河三角洲(湿地)的跨尺度制图，利用低分辨率土地覆盖产品和高分PlanetScope影像生成更精确的土地覆盖图。这些结果证明了所提框架在减轻噪声粗标签影响和生成可靠高分辨率土地覆盖图方面的有效性。","GIScience & Remote Sensing","2026-09-16T00:00:00Z",{"impact":17,"substance":18,"depth":17,"authority":20,"freshness":120,"relevant":22,"comment":121},9,"提出噪声标签学习引导的跨尺度框架，用低分辨率粗标签生成高分辨率土地覆盖图，精度提升显著，对农业遥感监测有实用价值。",[123],{"name":117,"url":114},[125,126,27,29,127],"农业人工智能","深度学习","高分辨率制图",[129,130],"农业人工智能 高分辨率制图 土地覆盖 深度学习","农业人工智能 高分辨率制图","农业人工智能高分辨率制图土地覆盖深度学习-2790","10.1080\u002F15481603.2026.2726002",{"doi":132,"openalex_id":134,"authors":135,"venue":117,"cited_by_count":36,"oa_url":114,"card":154,"direction":66,"ingested_from":68},"W7213283296",[136,139,142,145,148,150,152],{"name":137,"orcid":138},"Xiangyu Nie","https:\u002F\u002Forcid.org\u002F0009-0001-5095-6401",{"name":140,"orcid":141},"Cong Lin","https:\u002F\u002Forcid.org\u002F0000-0001-5386-7343",{"name":143,"orcid":144},"Wei Zhang","https:\u002F\u002Forcid.org\u002F0000-0001-8162-9422",{"name":146,"orcid":147},"Hong Fang","https:\u002F\u002Forcid.org\u002F0000-0003-3707-0910",{"name":149,"orcid":9},"Zhen Dong",{"name":151,"orcid":9},"Sicong Liu",{"name":153,"orcid":9},"Zhaohui Xue",{"tldr":155,"method":156,"finding":157,"direction":66,"opportunity":158},"提出噪声标签学习引导的跨尺度框架，用低分辨率粗标签生成高分辨率土地覆盖图。","谱掩膜过滤粗标签、自适应噪声评估加权损失、基于ViT的跨尺度迁移Transfor","在多个跨尺度任务上总体精度提升2%-9%，并在城市、农业、湿地场景生成更精确土地覆盖图。","可探索将粗标签跨尺度学习用于作物精细分类与长时序农情监测，降低高精度标注依赖。","2026-09-17T23:30:34.952106Z",{"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":174,"tags":176,"search_phrases":178,"slug":181,"view_count":36,"doi":182,"paper":183,"created_at":201},2432,"Multi-annual sub-pixel land cover mapping with TESSERA latent embeddings","https:\u002F\u002Fdoi.org\u002F10.31223\u002Fx59j7c","Mapping land cover in highly heterogeneous landscapes is challenging, and classifications have inherent limitations where the spatial resolution of remotely sensed data exceeds the size of small objects. Sub-pixel land cover fraction maps based on medium-resolution optical remote sensing data like Landsat or Sentinel-2 overcome this limitation but are often not consistently available across multiple years because creating temporally robust and comparable multi-class fraction models can be challenging. Latent embeddings from geospatial foundation models are promising as they represent high-dimensional, large-scale general-purpose spectral, temporal, and structural features derived from Earth Observation data that are largely independent of available high-quality observations at specific points in time. We here assessed the usability of TESSERA latent embeddings for machine-learning regression-based land cover fraction models and compared the performance to more conventional spectral-temporal metrics and spline coefficients as input features. We also assessed the suitability of these inputs to train temporally transferred and generalized fraction models. We found that TESSERA latent embeddings are a well-suitable input to land cover fraction mapping, performing slightly better than spectral-temporal-metrics-based models in many cases, both being notably outperformed by spline coefficients (MAE across classes and years 11.58 vs. 11.64 vs. 9.45). Spline coefficients also performed best in all temporally generalized and many temporally transferred models. However, models trained with TESSERA latent embeddings demonstrate the highest consistency of transferred models, highlighting their capability to handle data gaps. We suggest that TESSERA latent embeddings are a valuable input to fraction mapping where data availability across years is highly variable, but spline coefficients generally perform best when sufficient high-quality observations are available.","在高度异质的景观中绘制土地覆盖图具有挑战性，而且当遥感数据的空间分辨率超过小对象尺寸时，分类存在固有局限。基于中分辨率光学遥感数据（如Landsat或Sentinel-2）的亚像元土地覆盖比例图克服了这一局限，但往往无法在多个年份间一致获取，因为构建时间上稳健且可比的多类比例模型可能颇具挑战。地理空间基础模型的潜在嵌入（latent embeddings）前景广阔，因为它们代表了从地球观测数据中提取的高维、大规模通用光谱、时间和结构特征，这些特征在很大程度上独立于特定时间点可获取的高质量观测。我们在此评估了TESSERA潜在嵌入用于基于机器学习回归的土地覆盖比例模型的可用性，并将其性能与更传统的光谱-时间指标（spectral-temporal metrics）和样条系数（spline coefficients）作为输入特征进行了比较。我们还评估了这些输入用于训练时间迁移和广义化比例模型的适用性。我们发现，TESSERA潜在嵌入是土地覆盖比例制图的合适输入，在许多情况下略优于基于光谱-时间指标的模型，而两者的表现均明显不及样条系数（跨类别和年份的MAE分别为11.58、11.64和9.45）。样条系数在所有时间广义化模型和许多时间迁移模型中也表现最佳。然而，使用TESSERA潜在嵌入训练的模型在迁移模型间表现出最高的一致性，凸显了其处理数据缺口的能​​力。我们建议，在跨年份数据可用性高度可变的情况下，TESSERA潜在嵌入是比例制图的宝贵输入，但当有足够的高质量观测可用时，样条系数通常表现最佳。","OpenAlex","2026-09-11T00:00:00Z",70,{"impact":170,"substance":171,"depth":172,"authority":170,"freshness":21,"relevant":22,"comment":173},12,21,17,"评估TESSERA地理空间基础模型潜嵌入用于多年份亚像元土地覆盖制图，方法新颖、结论可靠，对农业遥感监测有参考价值，但属细分方法研究，影响面有限。",[175],{"name":166,"url":163},[125,27,177,29],"基础模型",[179,180],"农业人工智能 土地覆盖 基础模型 遥感","农业人工智能 土地覆盖","农业人工智能土地覆盖基础模型遥感-2432","10.31223\u002Fx59j7c",{"doi":182,"openalex_id":184,"authors":185,"venue":9,"cited_by_count":36,"oa_url":195,"card":196,"direction":66,"ingested_from":68},"W7212309976",[186,189,191,193],{"name":187,"orcid":188},"Franz Schug","https:\u002F\u002Forcid.org\u002F0000-0003-1534-5610",{"name":190,"orcid":9},"David Klehr",{"name":192,"orcid":9},"Jari Mahler",{"name":194,"orcid":9},"David Frantz","https:\u002F\u002Feartharxiv.org\u002Frepository\u002Fobject\u002F14912\u002Fdownload\u002F25925\u002F",{"tldr":197,"method":198,"finding":199,"direction":66,"opportunity":200},"评估TESSERA潜嵌入用于多年度亚像元土地覆盖制图，并与光谱时序指标和样条系数比较。","使用TESSERA潜嵌入、光谱时序指标和样条系数作为特征，训练机器学习回归模型预","样条系数精度最高，但TESSERA潜嵌入在时间迁移模型中一致性最好，适合数据缺失场景。","可探索将地理空间基础模型潜嵌入用于多年度作物覆盖比例制图，解决数据不均下的时序泛化问题。","2026-09-14T23:30:27.289645Z",{"id":203,"title":204,"url":205,"summary":206,"summary_zh":9,"content":9,"source_name":207,"source_url":9,"published_at":208,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":209,"score_detail":210,"sources":213,"tags":215,"search_phrases":219,"slug":222,"view_count":36,"doi":9,"paper":223,"created_at":231},3250,"农机化研究2026(11)：基于多源感知与语义分割的果树行识别与路径跟踪控制——郑州工业应用技术学院许洋洋等 DeepLabv3+ F1达0.94","http:\u002F\u002Fwww.qikanvip.com\u002Fqkml\u002F166268.html","农机化研究2026年第11期发表郑州工业应用技术学院许洋洋、柴亚珂、王莹、牛瑞利团队论文：针对果园中裸露土壤、遮蔽物干扰和冠层阴影等因素导致导航路径难以稳定提取的问题，提出融合多源感知与语义分割的自走式植保机果树识别与路径控制方法。首先利用无人机获取多光谱影像并生成正射影像（DOM）与数字表面模型（DSM），计算归一化差异绿度指数（NDGI），增强果树冠层与背景的可分性；然后采用U-Net\u002FDeepLabv3+语义分割网络在保留空间上下文信息的同时实现果树行\u002F背景端到端分割。田间对比试验结果表明，DeepLabv3+整体分割F1可达0.94，每样区生成导航线断裂1.2次，且转向工况下横向误差RMSE为0.11 m，显著优于传统阈值分割和SVM方法。","农机化研究","2026-09-22T00:00:00Z",79,{"impact":211,"substance":18,"depth":17,"authority":84,"freshness":13,"relevant":22,"comment":212},16,"多源遥感与语义分割结合的果园导航研究，F1达0.94且横向误差0.11米，方法新颖、数据扎实，具备行业参考价值。",[214],{"name":207,"url":205},[216,125,217,27,218],"智慧农业","农机导航","果园植保",[220,221],"郑州工业应用技术学院 果树行识别","DeepLabv3 果园导航 路径跟踪","郑州工业应用技术学院果树行识别-3250",{"doi":9,"openalex_id":9,"authors":224,"venue":9,"cited_by_count":36,"oa_url":9,"card":225,"direction":66,"ingested_from":230},[],{"tldr":226,"method":227,"finding":228,"direction":66,"opportunity":229},"提出融合无人机多光谱与语义分割的果树行识别与路径跟踪方法，实现植保机自主导航。","无人机多光谱DOM\u002FDSM与NDGI，U-Net\u002FDeepLabv3+语义分割，","DeepLabv3+分割F1达0.94，导航线断裂1.2次，转向横向误差RMSE 0.11 m，优于","可探索多光谱与DSM特征融合的轻量化分割模型，并迁移至多树种、多季节果园导航。","agent","2026-09-23T00:04:33.496233Z",{"id":233,"title":234,"url":235,"summary":236,"summary_zh":9,"content":9,"source_name":237,"source_url":9,"published_at":208,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":238,"score_detail":239,"sources":241,"tags":243,"search_phrases":246,"slug":249,"view_count":36,"doi":9,"paper":250,"created_at":257},3246,"Geospatial Artificial Intelligence in Precision Agriculture: A Systematic Review（精准农业中的地理空间人工智能：系统综述）","https:\u002F\u002Fwww.mdpi.com\u002F3043-1204\u002F1\u002F1\u002F4","MDPI AI Precis. Agric. 1(1) 4 发表系统综述：综述2019-2026年Scopus、Web of Science及补充检索文献，考察精准农业中的应用、多模态数据集成与决策支持。综述表明研究范式从孤立的制图与预测转向集成工作流，整合卫星与无人机影像、田间传感器、气象、土壤与管理记录。报告效益包括产量预测改善、压力与病害更早检测、土壤制图更精细、灌溉与投入品更精准。然而，产量、环境和经济效益仍依赖当地条件和运行验证。核心发现是：地理参考数据本身不能确保真正的地理空间AI；可靠工作流必须解决空间依赖性、尺度不匹配、可迁移性和不确定性。可解释性和融入农场运营对可执行推荐同样重要。采用仍受数据互操作性、连接性、可负担性、隐私和技术能力限制。","MDPI AI in Precision Agriculture",82,{"impact":17,"substance":18,"depth":17,"authority":20,"freshness":13,"relevant":22,"comment":240},"系统综述梳理2019-2026年GeoAI在精准农业的集成工作流与落地瓶颈，结论扎实、时效性强，对智慧农业技术路线有参考价值。",[242],{"name":237,"url":235},[216,125,27,244,245],"决策支持","多模态数据",[247,248],"GeoAI 精准农业 系统综述","地理空间人工智能 精准农业","GeoAI精准农业系统综述-3246",{"doi":9,"openalex_id":9,"authors":251,"venue":9,"cited_by_count":36,"oa_url":9,"card":252,"direction":66,"ingested_from":230},[],{"tldr":253,"method":254,"finding":255,"direction":66,"opportunity":256},"系统综述2019-2026年地理空间AI在精准农业的应用、多模态数据集成与决策支持。","系统综述Scopus、Web of Science等文献，分析多模态数据集成与决","地理参考数据不等于地理空间AI，可靠工作流须解决空间依赖、尺度、可迁移性与不确定性。","可研究跨尺度空间依赖建模与可迁移性评估，提升模型在不同农场条件下的泛化能力。","2026-09-23T00:04:33.172821Z",{"id":259,"title":260,"url":261,"summary":262,"summary_zh":263,"content":9,"source_name":264,"source_url":261,"published_at":208,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":265,"score_detail":266,"sources":268,"tags":270,"search_phrases":275,"slug":278,"view_count":36,"doi":279,"paper":280,"created_at":297},3191,"Spatial prediction of soil organic carbon stocks in Sudanese clay soils using regression kriging","https:\u002F\u002Fdoi.org\u002F10.3389\u002Fsjss.2026.16733","Soil organic carbon (SOC) stocks are a critical component of terrestrial carbon pools, influencing soil quality, agricultural productivity, and climate change mitigation. This study aimed to map and improve spatial estimation of SOC stocks in Sudan’s Blue Nile clay soils using regression kriging (RK). The model integrated 554 spatially unique soil profiles with nine environmental covariates: precipitation, temperature, relative humidity, normalized difference vegetation index (NDVI), land use\u002Fcover, bare soil index (BSI), digital elevation model (DEM), LS-factor, and aspect. Spectral indices were derived from Landsat 9 imagery (April 2024), while climate and terrain data were obtained from CHIRPS\u002FWorldClim and SRTM (30 m). RK performance was robust, with spatial cross-validation R 2 = 0.72, RMSE = 8.4 Mg C ha −1 (29% of mean observed stock), and mean bias = −0.8 Mg C ha −1 . Predicted SOC stocks (0–30 cm) ranged from 12.4 to 51.2 Mg C ha −1 (mean 28.6 Mg C ha −1 ). NDVI, clay content, and topographic wetness index were the most influential predictors. Agricultural lands exhibited the highest stocks (51.2 Mg C ha −1 ), while bare lands had the lowest (14.2 Mg C ha −1 ). This study (1) applies spatially explicit validation for SOC mapping in Sudan’s Blue Nile region, (2) harmonizes legacy and contemporary soil data using equivalent soil mass correction, and (3) provides high-resolution SOC maps for climate-resilient agricultural planning. Findings support soil carbon management and climate mitigation in semi-arid regions.","土壤有机碳（SOC）储量是陆地碳库的重要组成部分，影响土壤质量、农业生产力及气候变化减缓。本研究旨在利用回归克里金（RK）方法对苏丹青尼罗河黏土区SOC储量进行制图并改进其空间估算。该模型整合了554个空间独立土壤剖面与9个环境协变量：降水、温度、相对湿度、归一化植被指数（NDVI）、土地利用\u002F覆盖、裸土指数（BSI）、数字高程模型（DEM）、LS因子和坡向。光谱指数源自Landsat 9影像（2024年4月），气候与地形数据分别来自CHIRPS\u002FWorldClim和SRTM（30 m）。RK表现稳健，空间交叉验证R²=0.72，RMSE=8.4 Mg C ha⁻¹（为实测储量均值的29%），平均偏差=−0.8 Mg C ha⁻¹。预测SOC储量（0–30 cm）范围为12.4–51.2 Mg C ha⁻¹（均值28.6 Mg C ha⁻¹）。NDVI、黏粒含量和地形湿度指数是最具影响力的预测因子。农地储量最高（51.2 Mg C ha⁻¹），裸地最低（14.2 Mg C ha⁻¹）。本研究（1）对苏丹青尼罗河地区SOC制图采用空间显式验证，（2）利用等效土壤质量校正协调历史与当代土壤数据，（3）为气候韧性农业规划提供高分辨率SOC图。研究结果支持半干旱地区的土壤碳管理与气候减缓。","Spanish Journal of Soil Science",68,{"impact":21,"substance":171,"depth":172,"authority":84,"freshness":120,"relevant":22,"comment":267},"基于554个土壤剖面与多源遥感协变量的回归克里金制图研究，方法规范、验证充分，对半干旱区土壤碳管理与气候适应型农业规划有参考价值，但属区域性学术成果，公共影响有限。",[269],{"name":264,"url":261},[271,272,27,273,274],"农业遥感","气候变化","土壤碳汇","数字土壤制图",[276,277],"苏丹青尼罗河 土壤有机碳 回归克里金","Landsat 9 土壤有机碳 空间预测","苏丹青尼罗河土壤有机碳回归克里金-3191","10.3389\u002Fsjss.2026.16733",{"doi":279,"openalex_id":281,"authors":282,"venue":264,"cited_by_count":36,"oa_url":261,"card":291,"direction":296,"ingested_from":68},"W7213971196",[283,285,287,289],{"name":284,"orcid":9},"Faroug A.H. Jadalla",{"name":286,"orcid":9},"Kolapo O. Oluwasemire",{"name":288,"orcid":9},"Abd Elmagid A. Elmobarak",{"name":290,"orcid":9},"Mohammed A. M. Mohammed Zein",{"tldr":292,"method":293,"finding":294,"direction":66,"opportunity":295},"用回归克里金结合多源环境协变量预测苏丹青尼罗河粘土区土壤有机碳储量。","554个土壤剖面与9个环境协变量，Landsat 9、CHIRPS\u002FWorldC","模型R²=0.72，NDVI、粘土含量和地形湿度指数影响最大，农地碳储量最高。","可引入时序遥感与机器学习提升半干旱区SOC动态预测，并耦合农业管理措施评估固碳潜力。","数字乡村与农业信息化","2026-09-22T23:30:31.028178Z"]