[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-3560":3,"related-3560":52},{"id":4,"title":5,"url":6,"summary":7,"summary_zh":8,"content":8,"source_name":9,"source_url":6,"published_at":10,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":14,"score_detail":15,"sources":23,"tags":25,"search_phrases":30,"slug":33,"view_count":34,"doi":35,"paper":36,"created_at":51},3560,"Integrating spectroscopy and remote sensing to assess soil salinization under treated wastewater irrigation: a case study in northern Tunisia","https:\u002F\u002Fdoi.org\u002F10.1007\u002Fs11600-026-02020-1","Integrating spectroscopy and remote sensing to assess soil salinization under treated wastewater irrigation: a case study in northern Tunisia。Acta Geophysica",null,"Acta Geophysica","2026-09-24T00:00:00Z","论文",10,false,64,{"impact":16,"substance":17,"depth":18,"authority":19,"freshness":20,"relevant":21,"comment":22},8,18,16,13,9,1,"核心期刊论文，将光谱与遥感结合用于再生水灌溉下土壤盐渍化评估，方法有参考价值但属区域案例，公共影响有限。",[24],{"name":9,"url":6},[26,27,28,29],"遥感","土壤盐渍化","再生水灌溉","光谱监测",[31,32],"突尼斯 再生水灌溉 土壤盐渍化","光谱 遥感 土壤盐分监测","突尼斯再生水灌溉土壤盐渍化-3560",0,"10.1007\u002Fs11600-026-02020-1",{"doi":35,"openalex_id":37,"authors":38,"venue":9,"cited_by_count":34,"oa_url":8,"card":8,"direction":49,"ingested_from":50},"W7214208065",[39,41,44,46],{"name":40,"orcid":8},"Ines Smaani",{"name":42,"orcid":43},"Sonia Gannouni","https:\u002F\u002Forcid.org\u002F0000-0002-9083-832X",{"name":45,"orcid":8},"Salma Boussen",{"name":47,"orcid":48},"Noamen Rebaï","https:\u002F\u002Forcid.org\u002F0000-0001-6168-853X","农业遥感与作物表型","openalex","2026-09-26T23:30:31.596176Z",{"total":53,"page":21,"page_size":53,"items":54},6,[55,86,128,169,209,271],{"id":56,"title":57,"url":58,"summary":59,"summary_zh":8,"content":8,"source_name":60,"source_url":8,"published_at":61,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":62,"score_detail":63,"sources":67,"tags":69,"search_phrases":73,"slug":76,"view_count":34,"doi":8,"paper":77,"created_at":85},2855,"UAV无人机高光谱图像土壤盐度制图(湿度校正)——MDPI Agronomy","https:\u002F\u002Fwww.mdpi.com\u002F2073-4395\u002F16\u002F18\u002F1812","研究评估了6种光谱变换方法(原始反射率Ref、一阶导数FDR、PDS、OSC、FDR+PDS、FDR+OSC),结合3种机器学习算法(KNN、SVR、MLP)。进一步开发了集成这些基础学习者的Stacking集成模型,以提高湿度干扰下土壤盐度反演的精度。结果表明,Stacking模型在评估模型中达到最高的精度和稳定性。FDR+OSC-Stacking组合实现最佳验证性能,R²p=0.87,RMSEP=0.67 mS·cm⁻¹,RPD=2.93。FDR+OSC-Stacking组合成功应用于UAV高光谱图像,用于EC1:5的空间制图。来自吉林大学。","MDPI Agronomy","2026-09-15T00:00:00Z",75,{"impact":18,"substance":64,"depth":65,"authority":19,"freshness":16,"relevant":21,"comment":66},21,17,"方法组合新颖、验证指标扎实的无人机高光谱盐分制图研究，属细分领域实质进展，值得精选。",[68],{"name":60,"url":58},[70,71,72,26,27],"智慧农业","无人机","机器学习",[74,75],"土壤盐渍化 智慧农业 机器学习 无人机","土壤盐渍化 智慧农业","土壤盐渍化智慧农业机器学习无人机-2855",{"doi":8,"openalex_id":8,"authors":78,"venue":8,"cited_by_count":34,"oa_url":8,"card":79,"direction":49,"ingested_from":84},[],{"tldr":80,"method":81,"finding":82,"direction":49,"opportunity":83},"用无人机高光谱结合Stacking集成模型实现湿度干扰下的土壤盐度制图。","6种光谱变换与KNN、SVR、MLP及Stacking集成，基于UAV高光谱数据","FDR+OSC-Stacking最优，R²p=0.87、RMSEP=0.67 mS·cm⁻¹、RPD","可探索多时相\u002F多传感器融合与迁移学习，提升不同湿度与区域下盐度反演泛化性。","agent","2026-09-18T00:03:30.822732Z",{"id":87,"title":88,"url":89,"summary":90,"summary_zh":91,"content":8,"source_name":92,"source_url":89,"published_at":93,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":94,"score_detail":95,"sources":98,"tags":100,"search_phrases":104,"slug":107,"view_count":34,"doi":108,"paper":109,"created_at":127},3554,"Spatial Estimation of Carbon Sequestration Using Remote Sensing and Its Application in Carbon-credit Assessment: A Comprehensive Review","https:\u002F\u002Fdoi.org\u002F10.9734\u002Fjgeesi\u002F2026\u002Fv30i91125","Carbon sequestration is fundamental to climate-change mitigation because terrestrial and aquatic ecosystems store carbon in vegetation, biomass, soils, and the Earth’s crust. Reliable estimation of carbon stocks and their temporal and spatial changes is essential for the formulation of climate policy, ecosystem management, carbon accounting, and carbon-credit assessment. Conventional field measurements provide detailed information but are constrained by sampling requirements, cost, labour, spatial heterogeneity, and limitations in continuous monitoring. Remote sensing provides a complementary approach by enabling spatially continuous observations of vegetation characteristics, land-cover change, biomass, and other carbon-related variables across different spatial and temporal scales. Advances in multispectral, hyperspectral, synthetic aperture radar, LiDAR, thermal, and high-resolution satellite observations have extended the capability for carbon estimation. Their integration with geographic information systems, field observations, statistical techniques, and machine-learning algorithms enables precise spatial assessment of above-ground biomass, below-ground biomass, soil organic carbon, forest carbon, agricultural carbon, mangrove carbon, and blue-carbon resources. However, uncertainties associated with sensor characteristics, field-data quality, model selection, spatial variability, temporal dynamics, and scaling remain acknowledged limitations. Converting spatial carbon estimates into carbon credits requires consideration of baseline conditions, additionality, permanence, leakage, uncertainty, verification, and measurement, reporting, and verification requirements. This review critically examines remote sensing technologies, carbon-estimation models, spatial mapping approaches, machine-learning techniques, carbon-accounting frameworks, and their integration into carbon-credit assessment. It places particular emphasis on multisource data fusion, emerging Earth-observation technologies, artificial intelligence, uncertainty assessment, and digital MRV systems. Major methodological, technological, and policy gaps are identified, along with opportunities for developing transparent, spatially explicit, scientifically robust, and scalable carbon-credit assessment frameworks.","碳固存是减缓气候变化的基础，因为陆地和水生生态系统将碳储存在植被、生物量、土壤和地壳中。可靠估算碳储量及其时空变化，对于制定气候政策、生态系统管理、碳核算和碳信用评估至关重要。传统野外测量可提供详细信息，但受采样要求、成本、劳动力、空间异质性以及连续监测局限性等因素制约。遥感提供了一种互补方法，能够在不同时空尺度上对植被特征、土地覆盖变化、生物量及其他碳相关变量进行空间连续观测。多光谱、高光谱、合成孔径雷达、激光雷达、热红外和高分辨率卫星观测的进步，拓展了碳估算能力。这些技术与地理信息系统、野外观测、统计技术和机器学习算法的集成，使得对地上生物量、地下生物量、土壤有机碳、森林碳、农业碳、红树林碳和蓝碳资源进行精确空间评估成为可能。然而，与传感器特性、野外数据质量、模型选择、空间变异性、时间动态和尺度转换相关的不确定性仍是公认的局限。将空间碳估算转化为碳信用，需要考虑基线条件、额外性、永久性、泄漏、不确定性、核查以及测量、报告和核查要求。本综述批判性地审视了遥感技术、碳估算模型、空间制图方法、机器学习技术、碳核算框架及其在碳信用评估中的集成。特别强调了多源数据融合、新兴地球观测技术、人工智能、不确定性评估和数字化MRV系统。识别了主要的方法学、技术和政策差距，以及开发透明、空间明确、科学稳健且可扩展的碳信用评估框架的机遇。","Journal of Geography Environment and Earth Science International","2026-09-25T00:00:00Z",67,{"impact":96,"substance":17,"depth":18,"authority":96,"freshness":20,"relevant":21,"comment":97},12,"系统综述遥感碳汇估算与碳信用评估的方法、不确定性与数字MRV进展，对农业碳汇与碳交易信息化有参考价值，但属综述类论文，公共影响有限。",[99],{"name":92,"url":89},[72,26,101,102,103],"碳汇","碳信用","数字MRV",[105,106],"遥感 碳汇 估算","碳信用 评估 遥感","遥感碳汇估算-3554","10.9734\u002Fjgeesi\u002F2026\u002Fv30i91125",{"doi":108,"openalex_id":110,"authors":111,"venue":92,"cited_by_count":34,"oa_url":89,"card":121,"direction":49,"ingested_from":50},"W7214392294",[112,114,117,119],{"name":113,"orcid":8},"Aishwarya Desai",{"name":115,"orcid":116},"Himalaya Ganachari","https:\u002F\u002Forcid.org\u002F0000-0002-6157-9242",{"name":118,"orcid":8},"Sangita Shinde",{"name":120,"orcid":8},"Sachinkumar Nandgude",{"tldr":122,"method":123,"finding":124,"direction":125,"opportunity":126},"综述遥感碳储量估算技术及其在碳信用评估中的应用与挑战。","综述多光谱、高光谱、SAR、LiDAR等遥感与GIS、机器学习融合方法。","遥感可空间连续估算多类碳库，但不确定性及MRV要求制约碳信用转化。","农业绿色发展与碳","可研究多源数据融合与AI驱动的数字MRV，降低碳信用评估不确定性。","2026-09-26T23:30:30.002197Z",{"id":129,"title":130,"url":131,"summary":132,"summary_zh":133,"content":8,"source_name":134,"source_url":131,"published_at":10,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":135,"score_detail":136,"sources":140,"tags":142,"search_phrases":146,"slug":149,"view_count":34,"doi":150,"paper":151,"created_at":168},3515,"Soil mapping and fertilizer optimization for precision agriculture using artificial intelligence","https:\u002F\u002Fdoi.org\u002F10.1007\u002Fs13198-026-03435-1","Soil mapping and fertilizer optimization for precision agriculture using artificial intelligence。International Journal of Systems Assurance Engineering and Management","基于人工智能的精准农业土壤制图与肥料优化。《国际系统保障工程与管理杂志》","International Journal of Systems Assurance Engineering and Management",62,{"impact":96,"substance":137,"depth":138,"authority":19,"freshness":16,"relevant":21,"comment":139},14,15,"论文探讨AI用于土壤制图与施肥优化，属智慧农业细分方向，但摘要信息有限、影响面偏窄，暂不建议进入每日精选。",[141],{"name":134,"url":131},[70,143,144,26,145],"农业人工智能","精准施肥","土壤制图",[147,148],"土壤制图 人工智能 精准施肥","精准农业 肥料优化 AI","土壤制图人工智能精准施肥-3515","10.1007\u002Fs13198-026-03435-1",{"doi":150,"openalex_id":152,"authors":153,"venue":134,"cited_by_count":34,"oa_url":8,"card":162,"direction":166,"ingested_from":50},"W7214144821",[154,157,159],{"name":155,"orcid":156},"Neetu Mittal","https:\u002F\u002Forcid.org\u002F0000-0002-2012-0523",{"name":155,"orcid":158},"https:\u002F\u002Forcid.org\u002F0000-0001-6923-0013",{"name":160,"orcid":161},"Pradeepta Kumar Sarangi","https:\u002F\u002Forcid.org\u002F0000-0003-3827-6208",{"tldr":163,"method":164,"finding":165,"direction":166,"opportunity":167},"利用人工智能进行土壤制图和肥料优化，以支持精准农业。","人工智能方法，用于土壤制图与肥料优化。","AI可提升土壤制图与肥料优化的精准性，促进精准农业。","农业人工智能与决策模型","可探索多源数据融合与实时决策模型，提升肥料推荐的自适应性和可解释性。","2026-09-25T23:30:49.869002Z",{"id":170,"title":171,"url":172,"summary":173,"summary_zh":174,"content":8,"source_name":175,"source_url":172,"published_at":176,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":177,"score_detail":178,"sources":181,"tags":183,"search_phrases":187,"slug":190,"view_count":34,"doi":191,"paper":192,"created_at":208},3509,"Integrating multi-criteria decision-making and remote sensing for wildfire susceptibility assessment of the Lower Ziz Valley oases in Morocco","https:\u002F\u002Fdoi.org\u002F10.1007\u002Fs44288-026-00735-8","Oasis ecosystems in North Africa are ecologically and economically vital landscapes. Yet they are increasingly threatened by wildfires driven by intensifying climate change and anthropogenic pressures. Despite this urgency, spatially explicit wildfire susceptibility assessments for arid oasis environments remain scarce. This study develops and validates a Fire Susceptibility Index (FSI) for the Lower Ziz Valley oases in southeastern Morocco using an integrated Multi-Criteria Decision-Making (MCDM) and remote sensing framework. Six conditioning factors were selected: Land Surface Temperature (LST), Normalized Difference Vegetation Index (NDVI), Normalized Difference Moisture Index (NDMI), proximity to roads and settlements, and land cover classified via a U-Net deep learning model (overall accuracy: 95%). Factor weights were determined through the Analytic Hierarchy Process (AHP; CR = 0.09) and aggregated using Weighted Linear Combination (WLC). Land cover (w = 0.30) and NDVI (w = 0.25) emerged as dominant controls, with anthropogenic proximity factors jointly accounting for 31% of the index weight. FSI values ranged from 0.30 to 0.91 (mean = 0.67), with 42.7% of the oasis area (~ 1,707 ha) classified as high susceptibility (FSI ≥ 0.7). Validation against the August 2021 fire event (318 ha) demonstrated strong predictive performance: 96.5% of burned pixels fell within high-risk zones, and mean FSI was significantly higher for burned (0.72) than unburned (0.67) pixels ( p \u003C 0.001), with an AUC of 0.703. Corroborating validation against 60 historical fire records (2009–2024) showed that 98% of events occurred in high or extreme susceptibility zones. The resulting FSI map provides a robust spatial basis for prioritizing fuel management interventions and fire prevention strategies. Its integration into a dynamic geodatabase could further support the development of an early warning system (EWS) for operational wildfire prediction in these irreplaceable oasis landscapes.","北非的绿洲生态系统是具有重要生态和经济价值的景观。然而，日益加剧的气候变化和人为压力所驱动的野火正对其构成越来越大的威胁。尽管形势紧迫，针对干旱绿洲环境的空间显式野火易发性评估仍然匮乏。本研究采用多准则决策（MCDM）与遥感相结合的综合框架，为摩洛哥东南部下济兹河谷绿洲开发并验证了火灾易发性指数（FSI）。选取了六个条件因子：地表温度（LST）、归一化差异植被指数（NDVI）、归一化差异水分指数（NDMI）、道路与居民点邻近度，以及通过U-Net深度学习模型分类的土地覆盖（总体精度：95%）。因子权重通过层次分析法（AHP；CR = 0.09）确定，并采用加权线性组合（WLC）进行聚合。土地覆盖（w = 0.30）和NDVI（w = 0.25）成为主导控制因子，人为邻近因子合计占指数权重的31%。FSI值范围为0.30至0.91（均值 = 0.67），其中42.7%的绿洲面积（约1,707公顷）被归类为高易发性（FSI ≥ 0.7）。针对2021年8月火灾事件（318公顷）的验证表明模型具有较强的预测性能：96.5%的过火像元落在高风险区内，过火像元的FSI均值（0.72）显著高于未过火像元（0.67）（p \u003C 0.001），AUC为0.703。对60条历史火灾记录（2009–2024年）的佐证验证显示，98%的事件发生在高或极高易发性区域内。由此生成的FSI图为优先开展可燃物管理干预和火灾预防策略提供了可靠的空间依据。将其整合至动态地理数据库可进一步支持面向这些不可替代的绿洲景观的业务化野火预测预警系统（EWS）的开发。","Discover Geoscience","2026-09-22T00:00:00Z",73,{"impact":96,"substance":179,"depth":17,"authority":19,"freshness":16,"relevant":21,"comment":180},22,"方法扎实、验证充分的遥感野火易发性研究，对干旱绿洲农业防灾有参考价值，但属区域性案例，公共影响有限。",[182],{"name":175,"url":172},[143,184,26,185,186],"防灾减灾","野火风险","绿洲农业",[188,189],"摩洛哥 Ziz河谷 绿洲 野火","遥感 野火易发性 评估","摩洛哥Ziz河谷绿洲野火-3509","10.1007\u002Fs44288-026-00735-8",{"doi":191,"openalex_id":193,"authors":194,"venue":175,"cited_by_count":34,"oa_url":202,"card":203,"direction":49,"ingested_from":50},"W7213984877",[195,198,200],{"name":196,"orcid":197},"Rachid Ouachoua","https:\u002F\u002Forcid.org\u002F0000-0002-3194-7585",{"name":199,"orcid":8},"Hamid Benssi",{"name":201,"orcid":8},"Najib Kadiri","https:\u002F\u002Flink.springer.com\u002Fcontent\u002Fpdf\u002F10.1007\u002Fs44288-026-00735-8.pdf",{"tldr":204,"method":205,"finding":206,"direction":49,"opportunity":207},"构建遥感与多准则决策融合的绿洲野火易发性指数，并验证其预测精度。","AHP加权线性组合，结合LST、NDVI、NDMI、道路居民点及U-Net土地覆","42.7%绿洲面积高易发，96.5%过火区落于高风险区，AUC达0.703。","可延伸至动态地理数据库与实时早期预警系统，融合气象与人类活动数据。","2026-09-25T23:30:35.200699Z",{"id":210,"title":211,"url":212,"summary":213,"summary_zh":214,"content":8,"source_name":215,"source_url":212,"published_at":216,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":217,"score_detail":218,"sources":220,"tags":222,"search_phrases":226,"slug":229,"view_count":34,"doi":230,"paper":231,"created_at":270},3507,"OCO‐3 Meets ECOSTRESS: Insights Into Ecosystem Diurnal Water‐Use Efficiency From Co‐Located Solar‐Induced Fluorescence and Thermal Observations","https:\u002F\u002Fdoi.org\u002F10.1029\u002F2026gl124105","Abstract Terrestrial ecosystem regulation of carbon and water fluxes is critical for constraining climate–biosphere feedbacks but remains poorly quantified across space and time. Here, we evaluate whether co‐located observations from NASA's Orbiting Carbon Observatory‐3 (OCO‐3) and the ECOsystem Spaceborne Thermal Radiometer Experiment on Space Station (ECOSTRESS) can reproduce ecosystem water use efficiency (WUE) dynamics observed at FLUXNET sites across temporal scales, vegetation types, climates, and drought conditions. We created the ECOCO3 data set, which harmonizes OCO‐3 and ECOSTRESS observations in space and time. ECOCO3 captures broad seasonal and diurnal carbon and water flux patterns including midday drought responses. Sampling sensitivity analysis shows that ECOCO3 is primarily limited by available sample size for distinguishing vegetation and climate driven differences in WUE. Our findings highlight both the promise and limitations of remote sensing for resolving sub‐daily carbon–water coupling.","陆地生态系统对碳通量和水通量的调节对于约束气候–生物圈反馈至关重要，但在空间和时间尺度上仍缺乏充分的量化。在此，我们评估了来自NASA轨道碳观测站-3（OCO-3）和空间站生态系统星载热辐射计实验（ECOSTRESS）的同位观测能否在时间尺度、植被类型、气候条件和干旱状况下重现FLUXNET站点观测到的生态系统水分利用效率（WUE）动态。我们创建了ECOCO3数据集，该数据集在空间和时间上协调了OCO-3和ECOSTRESS的观测。ECOCO3能够捕捉广泛的季节性和日间碳通量与水通量模式，包括正午干旱响应。采样敏感性分析表明，ECOCO3主要受限于可用样本量，难以区分植被和气候驱动的WUE差异。我们的研究结果既凸显了遥感在解析亚日尺度碳–水耦合方面的前景，也揭示了其局限性。","Geophysical Research Letters","2026-09-23T00:00:00Z",81,{"impact":17,"substance":179,"depth":17,"authority":138,"freshness":16,"relevant":21,"comment":219},"NASA两颗卫星协同观测提升生态系统碳水耦合监测能力，方法新颖、数据可靠，对农业遥感与水资源管理有参考价值。",[221],{"name":215,"url":212},[223,26,224,225,101],"农业遥感","水资源利用","生态监测",[227,228],"OCO-3 ECOSTRESS 水分利用效率","ECOCO3 数据集 碳水通量","OCO-3ECOSTRESS水分利用效率-3507","10.1029\u002F2026gl124105",{"doi":230,"openalex_id":232,"authors":233,"venue":215,"cited_by_count":34,"oa_url":264,"card":265,"direction":49,"ingested_from":50},"W7214122316",[234,237,240,243,246,249,252,255,258,261],{"name":235,"orcid":236},"Zoe Pierrat","https:\u002F\u002Forcid.org\u002F0000-0002-6726-2406",{"name":238,"orcid":239},"Thomas P. Kurosu","https:\u002F\u002Forcid.org\u002F0000-0003-2555-7780",{"name":241,"orcid":242},"Abhishek Chatterjee","https:\u002F\u002Forcid.org\u002F0000-0002-3680-0160",{"name":244,"orcid":245},"Joshua B. Fisher","https:\u002F\u002Forcid.org\u002F0000-0003-4734-9085",{"name":247,"orcid":248},"Margaret C. Johnson","https:\u002F\u002Forcid.org\u002F0000-0003-1481-9706",{"name":250,"orcid":251},"Le Kuai","https:\u002F\u002Forcid.org\u002F0000-0001-6406-1150",{"name":253,"orcid":254},"Kaniska Mallick","https:\u002F\u002Forcid.org\u002F0000-0002-2735-930X",{"name":256,"orcid":257},"Nicholas Cody Parazoo","https:\u002F\u002Forcid.org\u002F0000-0002-4424-7780",{"name":259,"orcid":260},"Benjamin C. Wiebe","https:\u002F\u002Forcid.org\u002F0000-0002-9325-1540",{"name":262,"orcid":263},"Kerry A. Cawse-Nicholson","https:\u002F\u002Forcid.org\u002F0000-0002-0510-4066","https:\u002F\u002Fonlinelibrary.wiley.com\u002Fdoi\u002Fpdfdirect\u002F10.1029\u002F2026GL124105",{"tldr":266,"method":267,"finding":268,"direction":49,"opportunity":269},"融合OCO-3与ECOSTRESS观测评估生态系统日间水分利用效率动态。","构建ECOCO3数据集，结合FLUXNET站点验证与采样敏感性分析。","ECOCO3能捕捉季节与日间碳-水通量模式，但样本量限制其区分植被与气候差异。","可探索多源遥感融合提升亚日尺度碳水耦合估算精度，并扩展至农田生态系统。","2026-09-25T23:30:33.881463Z",{"id":272,"title":273,"url":274,"summary":275,"summary_zh":276,"content":8,"source_name":277,"source_url":274,"published_at":10,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":278,"score_detail":279,"sources":281,"tags":283,"search_phrases":287,"slug":290,"view_count":34,"doi":291,"paper":292,"created_at":315},3504,"WaterFormer-MS: A multi-stream deep learning framework for surface water segmentation from satellite imagery","https:\u002F\u002Fdoi.org\u002F10.24850\u002Fj-tyca-18-03-08","Accurate mapping of surface water bodies is essential for dam monitoring, irrigation planning, and sustainable water resource management. Conventional methods based on spectral indices, such as NDWI, can face challenges in complex environments due to spectral ambiguity, seasonal variability, and urban or vegetated surfaces. Deep learning approaches such as U-Net and DeepLabv3+ have improved water-body segmentation; however, accurately delineating fine boundaries and detecting small water bodies remain challenging. To address these limitations, this study proposes WaterFormer-MS, a multi-stream deep learning framework for surface water segmentation from Sentinel-2 imagery. The architecture integrates spectral, geometric, and contextual information through dedicated processing streams and combines the extracted features using a convolutional fusion module enhanced with CBAM attention. This design aims to improve feature representation and support accurate water-body delineation. The proposed model was trained and evaluated using Sentinel-2 imagery from a publicly available Kaggle dataset containing diverse land-cover conditions, including urban, vegetated, and agricultural environments. At 300 epochs, WaterFormer-MS achieved an overall accuracy of 97.4%, a precision of 93.2%, a recall of 94.2%, an F1-score of 93.7%, and an IoU of 89.7%. Within the experimental framework, WaterFormer-MS achieved higher values than the other proposed variants. Comparisons with conventional NDWI-based approaches and published deep learning methods suggest competitive performance; however, these comparisons should be interpreted cautiously because datasets, preprocessing procedures, training protocols, and evaluation settings may differ. The results support the potential of the proposed framework for surface water segmentation and environmental monitoring applications in diverse remote sensing scenarios.","准确绘制地表水体分布对于大坝监测、灌溉规划和水资源可持续管理至关重要。基于NDWI等光谱指数的传统方法在复杂环境中可能因光谱模糊性、季节性变化以及城市或植被覆盖地表而面临挑战。U-Net和DeepLabv3+等深度学习方法改善了水体分割效果，但精确描绘精细边界和检测小型水体仍然具有挑战性。为解决这些局限，本研究提出WaterFormer-MS，一种用于Sentinel-2影像地表水分割的多流深度学习框架。该架构通过专门的处理流整合光谱、几何和上下文信息，并利用经CBAM注意力增强的卷积融合模块将提取的特征进行融合。该设计旨在改善特征表征并支持精确的水体描绘。所提模型使用来自公开Kaggle数据集的Sentinel-2影像进行训练和评估，该数据集包含城市、植被和农业环境等多种土地覆盖条件。在300个训练轮次时，WaterFormer-MS取得了97.4%的总体精度、93.2%的精确率、94.2%的召回率、93.7%的F1分数和89.7%的IoU。在实验框架内，WaterFormer-MS取得了高于其他所提变体的数值。与传统基于NDWI的方法及已发表的深度学习方法进行比较，结果表明其性能具有竞争力；然而，这些比较应谨慎解读，因为数据集、预处理流程、训练方案和评估设置可能存在差异。研究结果支持所提框架在多样化遥感场景中用于地表水分割和环境监测应用的潜力。","Tecnología y Ciencias del Agua",74,{"impact":138,"substance":179,"depth":17,"authority":12,"freshness":20,"relevant":21,"comment":280},"提出多流深度学习框架用于卫星影像地表水提取，精度指标明确，对灌溉规划与水资源监测有参考价值，但属方法类论文，产业影响有限。",[282],{"name":277,"url":274},[143,284,26,285,286],"Sentinel-2","水资源管理","地表水体提取",[288,289],"WaterFormer-MS 地表水 分割","Sentinel-2 水体提取 深度学习","WaterFormer-MS地表水分割-3504","10.24850\u002Fj-tyca-18-03-08",{"doi":291,"openalex_id":293,"authors":294,"venue":277,"cited_by_count":34,"oa_url":274,"card":310,"direction":49,"ingested_from":50},"W7214228825",[295,298,301,304,307],{"name":296,"orcid":297},"Ayoub Benali","https:\u002F\u002Forcid.org\u002F0000-0002-8989-755X",{"name":299,"orcid":300},"Yesma Bendaha","https:\u002F\u002Forcid.org\u002F0000-0002-1347-856X",{"name":302,"orcid":303},"Malika Seddik Bouchouicha","https:\u002F\u002Forcid.org\u002F0009-0001-6092-784X",{"name":305,"orcid":306},"Zakaria Bellahcene","https:\u002F\u002Forcid.org\u002F0000-0003-4353-7871",{"name":308,"orcid":309},"Rachid Rimani","https:\u002F\u002Forcid.org\u002F0000-0002-1923-4312",{"tldr":311,"method":312,"finding":313,"direction":49,"opportunity":314},"提出多流深度学习框架WaterFormer-MS，从Sentinel-2影像中精准分割地表水体。","多流架构融合光谱、几何与上下文特征，CBAM注意力卷积融合，Sentinel-2","300轮训练下总体精度97.4%、IoU 89.7%，优于NDWI及对比深度学习方法。","可探索多流框架在农业灌溉水体与小型水塘精细提取中的迁移，并引入时序Sentinel-2提升季节鲁棒性。","2026-09-25T23:30:31.338437Z"]