[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-3188":3,"related-3188":74},{"id":4,"title":5,"url":6,"summary":7,"summary_zh":8,"content":9,"source_name":10,"source_url":6,"published_at":11,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":15,"score_detail":16,"sources":18,"tags":20,"primary_topic":9,"secondary_topics":25,"classification":26,"search_phrases":50,"slug":53,"view_count":54,"doi":55,"paper":56,"created_at":73},3188,"Identification of granitic pegmatites based on GF-5 hyperspectral data and LSTM-Transformer model","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.acags.2026.100406","Lithium is a key strategic mineral to support new energy and advanced technology, mainly enriched in granitic pegmatite. The integration of hyperspectral remote sensing and deep learning frameworks has emerged as a leading approach for quantitative identification of pegmatites. However, deep learning algorithms are often limited by the scarcity of labeled samples and the challenge of extracting features from high-dimensional spectral data. In this study, a data augmentation-driven LSTM-Transformer framework is proposed for pegmatite mapping using GF-5 hyperspectral remote sensing data. In view of limited samples, this study constructed an enhanced training data set by integrating field-measured and infrared scanning spectra, simulating real spectral variability and linear mixing to improve data diversity. The proposed LSTM-Transformer architecture combines LSTM’s advantages in local sequence feature modeling and Transformer’s capacity to capture long-range global relationships, thereby enhancing the detection of weakly diagnostic spectral features. The framework was demonstrated in the Jingerquan lithium deposit, Xinjiang, China. The high-probability areas delineated by the proposed model successfully captured several pegmatite dikes documented in previous geological studies. The prediction results were further validated through field sampling. Geochemical analyses of specimens collected from the predicted high-probability target areas show an average Li 2 O grade of 7.86 wt.%, which meets industrial requirements and confirms the reliability of the proposed method. Overall, these results provide the potential of the proposed framework for pegmatite mapping and subsequent mineral exploration under the conditions of limited labeled samples.","锂是支撑新能源与先进技术的关键战略矿产，主要富集于花岗伟晶岩中。高光谱遥感与深度学习框架的融合已成为伟晶岩定量识别的前沿方法。然而，深度学习算法常受限于标注样本稀缺以及高维光谱数据特征提取的难题。本研究提出了一种数据增强驱动的LSTM-Transformer框架，用于基于GF-5高光谱遥感数据的伟晶岩填图。针对样本有限的问题，本研究通过整合野外实测光谱与红外扫描光谱构建了增强训练数据集，模拟真实光谱变异与线性混合以提升数据多样性。所提出的LSTM-Transformer架构结合了LSTM在局部序列特征建模方面的优势与Transformer捕捉长程全局关系的能力，从而增强了对弱诊断光谱特征的检测。该框架在中国新疆镜儿泉锂矿床进行了验证。所提模型圈定的高概率区域成功捕获了以往地质研究中记录的若干伟晶岩脉。预测结果进一步通过野外采样得到验证。从预测高概率目标区采集的标本地球化学分析显示，Li₂O平均品位为7.86 wt.%，满足工业要求，证实了所提方法的可靠性。总体而言，这些结果展示了所提框架在标注样本有限条件下用于伟晶岩填图及后续矿产勘查的潜力。",null,"Applied Computing and Geosciences","2026-09-19T00:00:00Z","论文",10,false,0,{"impact":15,"substance":15,"depth":15,"authority":15,"freshness":15,"relevant":15,"comment":17},"该论文聚焦高光谱遥感与深度学习用于伟晶岩型锂矿识别，属地质矿产勘探领域，与三农、农业信息化、智慧农业主题无关，不建议进入每日精选。",[19],{"name":10,"url":6},[21,22,23,24],"深度学习","遥感监测","高光谱遥感","锂矿勘探",[],{"status":27,"locked":14,"version":28,"method":29,"evidence":30},"needs_review","2026-10-22.2","rules",{"candidates":31,"agricultural_evidence":14,"reason":49},[32,42],{"code":33,"score":34,"supported":35,"title":36,"abstract":38,"tags":41},"I3",19,true,[37],"hyperspectral",[39,40,37],"遥感","remote sensing",[22,23],{"code":43,"score":44,"supported":35,"title":45,"abstract":46,"tags":48},"I2",7,[],[21,47],"deep learning",[21],"农业相关性需要复核",[51,52],"GF-5 高光谱 伟晶岩","新疆 镜儿泉 锂矿","GF-5高光谱伟晶岩-3188",4,"10.1016\u002Fj.acags.2026.100406",{"doi":55,"openalex_id":57,"authors":58,"venue":10,"cited_by_count":15,"oa_url":65,"card":66,"direction":70,"ingested_from":72},"W7213617576",[59,62],{"name":60,"orcid":61},"Zhong Li","https:\u002F\u002Forcid.org\u002F0000-0002-2312-3746",{"name":63,"orcid":64},"Ziye Wang","https:\u002F\u002Forcid.org\u002F0000-0001-6538-5798","https:\u002F\u002Fwww.sciencedirect.com\u002Fscience\u002Farticle\u002Fpii\u002FS259019742600090X\u002Fpdf",{"tldr":67,"method":68,"finding":69,"direction":70,"opportunity":71},"提出数据增强驱动的LSTM-Transformer框架，用GF-5高光谱数据识别花岗伟晶岩。","GF-5高光谱数据，融合实测与红外扫描光谱增强样本，构建LSTM-Transfo","在新疆金儿泉锂矿成功圈定伟晶岩脉，验证区Li2O平均品位7.86 wt.%。","农业遥感与作物表型","可将该高光谱-深度学习框架迁移至农田土壤属性或作物养分反演，解决标签稀缺问题。","openalex","2026-09-22T23:30:25.608490Z",{"total":75,"page":76,"page_size":75,"items":77},6,1,[78,129,187,243,327,378],{"id":79,"title":80,"url":81,"summary":82,"summary_zh":83,"content":9,"source_name":84,"source_url":81,"published_at":85,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":15,"score_detail":86,"sources":88,"tags":90,"primary_topic":92,"secondary_topics":97,"classification":98,"search_phrases":106,"slug":109,"view_count":15,"doi":110,"paper":111,"created_at":128},4831,"Non-linear spectral matching for asbestos detection in urban hyperspectral imagery using dynamic time warping","https:\u002F\u002Fdoi.org\u002F10.34069\u002Fai\u002F2026.88.02.2","Hyperspectral remote sensing has gained importance in applications such as biomedicine, agriculture, and environmental monitoring, where spectral signatures enable material detection using computational methods. This article presents the implementation and evaluation of the Dynamic Time Warping (DTW) method for detecting asbestos in hyperspectral images. The method is compared to the widely used correlation technique. The research follows the CRISP-DM methodology, adapted into four phases: business understanding, data preparation, modeling and evaluation, and deployment. Results show that the DTW method is effective for detecting asbestos in hyperspectral images, though it is less sensitive and less computationally efficient than the correlation method. The study suggests that the DTW method is suitable for identifying materials with non-linear spectral variations and can be applied to detect other materials beyond asbestos. Given the public health relevance of asbestos detection, this method can contribute to more precise identification of asbestos-cement roofs in urban areas. The use of open-source libraries such as Spectral, NumPy, SciPy, and FastDTW for the implementation highlights the accessibility of this approach for academic and industrial applications. This work aims to serve as a reference for future research on the detection of materials using DTW in hyperspectral imaging.","高光谱遥感在生物医学、农业和环境监测等应用中日益重要，其中光谱特征使得利用计算方法进行材料检测成为可能。本文介绍了动态时间规整（DTW）方法在高光谱图像中检测石棉的实现与评估。该方法与广泛使用的相关技术进行了比较。研究遵循CRISP-DM方法，并调整为四个阶段：业务理解、数据准备、建模与评估以及部署。结果表明，DTW方法在高光谱图像中检测石棉是有效的，尽管其灵敏度和计算效率低于相关方法。研究表明，DTW方法适用于识别具有非线性光谱变化的材料，并可应用于检测石棉以外的其他材料。鉴于石棉检测与公共健康的相关性，该方法可有助于更精确地识别城市地区的石棉水泥屋顶。使用Spectral、NumPy、SciPy和FastDTW等开源库进行实现，凸显了该方法在学术和工业应用中的可及性。本研究旨在为未来利用DTW在高光谱成像中进行材料检测的研究提供参考。","Revista Amazonia Investiga","2026-10-10T00:00:00Z",{"impact":15,"substance":15,"depth":15,"authority":15,"freshness":15,"relevant":15,"comment":87},"该论文聚焦城市高光谱遥感石棉检测，属环境与公共健康领域，与三农、农业信息化、智慧农业无直接关联，相关性门槛未通过。",[89],{"name":84,"url":81},[22,23,91],"城市环境监测",{"code":33,"slug":93,"name":94,"path":95,"parent_code":96},"remote-sensing","遥感与农情监测","\u002Ftopic\u002Fdigital-agriculture\u002Fremote-sensing","I",[],{"status":99,"locked":14,"version":28,"method":29,"evidence":100},"classified",{"candidates":101,"agricultural_evidence":35},[102],{"code":33,"score":34,"supported":35,"title":103,"abstract":104,"tags":105},[37],[39,40,37],[22,23],[107,108],"城市环境监测 高光谱遥感 遥感监测","城市环境监测 高光谱遥感","城市环境监测高光谱遥感遥感监测-4831","10.34069\u002Fai\u002F2026.88.02.2",{"doi":110,"openalex_id":112,"authors":113,"venue":84,"cited_by_count":15,"oa_url":81,"card":123,"direction":70,"ingested_from":72},"W7221543948",[114,117,120],{"name":115,"orcid":116},"Gabriel Elías Chanchí Golondrino","https:\u002F\u002Forcid.org\u002F0000-0002-0257-1988",{"name":118,"orcid":119},"Manuel Alejandro Ospina Alarcón","https:\u002F\u002Forcid.org\u002F0000-0003-4510-0753",{"name":121,"orcid":122},"Manuel Saba","https:\u002F\u002Forcid.org\u002F0000-0003-1345-9657",{"tldr":124,"method":125,"finding":126,"direction":70,"opportunity":127},"用动态时间规整（DTW）在高光谱城市影像中检测石棉，并与相关法对比。","基于CRISP-DM流程，用DTW与相关法对比，开源库Spectral、NumP","DTW能有效检测石棉，但灵敏度和计算效率低于相关法，适合非线性光谱变化材料。","可将DTW非线性光谱匹配迁移到作物\u002F土壤等复杂地物识别，并优化计算效率。","2026-10-13T23:30:26.380112Z",{"id":130,"title":131,"url":132,"summary":133,"summary_zh":134,"content":9,"source_name":135,"source_url":132,"published_at":136,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":15,"score_detail":137,"sources":139,"tags":141,"primary_topic":9,"secondary_topics":143,"classification":144,"search_phrases":156,"slug":159,"view_count":160,"doi":161,"paper":162,"created_at":186},4231,"CrossCD-Eval: a unified framework with domain-aware feature normalization for cross-domain remote sensing change detection","https:\u002F\u002Fdoi.org\u002F10.1038\u002Fs41598-026-70914-7","Abstract Change detection (CD) in remote sensing imagery is essential for monitoring urban growth, land-use transformation, and environmental dynamics. However, cross-domain change detection remains challenging due to distribution shifts caused by differences in geographic regions, spatial resolutions, sensor\u002Facquisition conditions, temporal periods, and annotation protocols. This paper introduces CrossCD-Eval, a unified framework for evaluating the generalization of deep learning-based change detection models under in-domain, cross-domain, and few-shot domain adaptation settings. Five recent models, namely GLAFormer 1 , Change-Mamba 2 , EHCTNet 3 , ELGC-Net 4 , and STNet 5 , are evaluated using CLCD, HRSCD, and LEVIR-CD. To reduce representation-level domain mismatch, we propose a Domain-Aware Feature Normalization (DAFN) module that aligns source and target feature statistics as a lightweight feature-level adaptation component. Experimental results show that all models achieve strong in-domain performance but degrade under cross-domain transfer, confirming the impact of domain shift. DAFN improves cross-domain and few-shot adaptation performance, with larger gains observed under more challenging source–target shifts. Repeated 10-shot experiments over five random target-sample selections further show that DAFN improves adaptation stability, reducing IoU standard deviation by up to 22.7%. An additional ablation and baseline-comparison study demonstrates that the observed gains are not solely due to preprocessing. Compared with full preprocessing, AdaBN, adversarial feature alignment, and the instance-normalization-based CrossCDNet baseline, DAFN provides additional feature-level adaptation benefits, especially when combined with strong backbones such as STNet.","摘要 遥感影像中的变化检测（CD）对于监测城市扩张、土地利用转变和环境动态至关重要。然而，由于地理区域、空间分辨率、传感器\u002F采集条件、时间段和标注协议差异所导致的分布偏移，跨域变化检测仍然具有挑战性。本文提出CrossCD-Eval，一个用于评估基于深度学习的变化检测模型在域内、跨域和少样本域适应设置下泛化能力的统一框架。使用CLCD、HRSCD和LEVIR-CD对五个近期模型进行了评估，即GLAFormer 1、Change-Mamba 2、EHCTNet 3、ELGC-Net 4和STNet 5。为减少表示层面的域不匹配，我们提出域感知特征归一化（DAFN）模块，作为轻量级特征级适应组件，对齐源域和目标域的特征统计量。实验结果表明，所有模型均取得较强的域内性能，但在跨域迁移时性能下降，证实了域偏移的影响。DAFN提升了跨域和少样本适应性能，在更具挑战性的源-目标偏移下增益更大。在五次随机目标样本选择上重复进行的10-shot实验进一步表明，DAFN提升了适应稳定性，将IoU标准差最多降低22.7%。额外的消融和基线对比研究表明，所观察到的增益并非仅由预处理所致。与完整预处理、AdaBN、对抗特征对齐以及基于实例归一化的CrossCDNet基线相比，DAFN提供了额外的特征级适应收益，尤其是与STNet等强骨干网络结合时。","Scientific Reports","2026-10-03T00:00:00Z",{"impact":15,"substance":15,"depth":15,"authority":15,"freshness":15,"relevant":15,"comment":138},"属遥感变化检测通用方法研究，未涉及农业场景，与三农及农业信息化主题不相关。",[140],{"name":135,"url":132},[21,22,142],"变化检测",[],{"status":27,"locked":14,"version":28,"method":29,"evidence":145},{"candidates":146,"agricultural_evidence":14,"reason":49},[147,152],{"code":33,"score":148,"supported":35,"title":149,"abstract":150,"tags":151},15,[40],[39,40],[22],{"code":43,"score":44,"supported":35,"title":153,"abstract":154,"tags":155},[],[21,47],[21],[157,158],"CrossCD-Eval 跨域变化检测","遥感变化检测 域自适应","CrossCD-Eval跨域变化检测-4231",3,"10.1038\u002Fs41598-026-70914-7",{"doi":161,"openalex_id":163,"authors":164,"venue":135,"cited_by_count":15,"oa_url":180,"card":181,"direction":70,"ingested_from":72},"W7218559287",[165,168,171,174,177],{"name":166,"orcid":167},"Menna M. Elkholy","https:\u002F\u002Forcid.org\u002F0000-0003-4475-5424",{"name":169,"orcid":170},"Hala M. Ebied","https:\u002F\u002Forcid.org\u002F0000-0001-9843-842X",{"name":172,"orcid":173},"Mohamed F. Tolba","https:\u002F\u002Forcid.org\u002F0000-0003-3104-6418",{"name":175,"orcid":176},"Marwa S. Moustafa","https:\u002F\u002Forcid.org\u002F0000-0003-3805-9668",{"name":178,"orcid":179},"Dina Elsayad","https:\u002F\u002Forcid.org\u002F0000-0003-3934-7637","https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41598-026-70914-7.pdf",{"tldr":182,"method":183,"finding":184,"direction":70,"opportunity":185},"提出统一框架CrossCD-Eval评估跨域遥感变化检测，并设计DAFN模块提升跨域泛化。","在CLCD、HRSCD、LEVIR-CD上评测五种模型，提出域感知特征归一化DA","模型域内表现强但跨域下降，DAFN提升跨域与少样本适应，IoU标准差最多降22.7%。","可将DAFN用于农业跨区域\u002F跨传感器作物变化监测，探索轻量域适应与少样本迁移。","2026-10-05T23:30:25.083737Z",{"id":188,"title":189,"url":190,"summary":191,"summary_zh":192,"content":9,"source_name":193,"source_url":190,"published_at":194,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":15,"score_detail":195,"sources":197,"tags":199,"primary_topic":9,"secondary_topics":202,"classification":203,"search_phrases":220,"slug":223,"view_count":54,"doi":224,"paper":225,"created_at":242},3798,"Debris Flow Susceptibility Mapping Using Explainable Deep Learning in Eastern Hindukush Pakistan","https:\u002F\u002Fdoi.org\u002F10.1002\u002Fesp.70419","ABSTRACT Debris flows are among the most destructive geomorphological hazards in mountainous regions, causing damage to human settlements, infrastructure and ecosystems. The Eastern Hindu Kush region of Pakistan is particularly vulnerable due to steep topography, intense monsoonal precipitation, active tectonics and increasing anthropogenic disturbance. This study develops an integrated debris flow susceptibility assessment framework by combining convolutional neural networks (CNN) with SHapley Additive exPlanations (SHAP). A total of 13 conditioning factors representing topographic, hydrological, geological, environmental and anthropogenic characteristics were prepared using GIS and remote sensing datasets at a spatial resolution of 30 × 30 m. The CNN model demonstrated strong predictive performance, achieving an area under the curve (AUC) of 0.94, overall accuracy (ACC) of 0.86, precision of 0.84, recall of 0.88, an F1‐score of 0.86 and a Kappa coefficient of 0.71. The resulting susceptibility map identified approximately 26.3% of the study area as high to very high susceptibility zones, predominantly concentrated along steep river valleys and highly dissected mountainous terrain. The susceptibility map was further validated by debris flow density analysis, which showed hazard density increasing from 0 events\u002Fkm 2 in very low zones to 0.048 events\u002Fkm 2 in very high zones, with most historical debris flow events concentrated within the very high susceptibility class. SHAP analysis identified slope, NDVI, rainfall, distance to roads and distance to streams as the most influential conditioning factors controlling debris flow occurrence. The proposed CNN–SHAP framework provides a robust and interpretable approach for regional‐scale debris flow susceptibility assessment, with applicability to similar mountainous environments. These findings provide valuable information for hazard mitigation, infrastructure planning and land‐use management in data‐scarce mountain regions.","泥石流是山区最具破坏性的地貌灾害之一，会对人类聚居地、基础设施和生态系统造成损害。巴基斯坦东兴都库什地区由于地形陡峭、季风降水强烈、构造活动活跃以及人为干扰日益加剧，尤为脆弱。本研究通过将卷积神经网络（CNN）与SHapley加法解释（SHAP）相结合，构建了一个综合性的泥石流易发性评估框架。利用GIS和遥感数据集，以30×30 m的空间分辨率制备了代表地形、水文、地质、环境和人为特征的13个 conditioning factors（孕灾因子）。CNN模型表现出较强的预测性能，曲线下面积（AUC）达到0.94，总体精度（ACC）为0.86，精确率为0.84，召回率为0.88，F1分数为0.86，Kappa系数为0.71。所得易发性图识别出约26.3%的研究区域为高至极高易发性区，主要集中于陡峭的河谷和高度切割的山地地形。易发性图进一步通过泥石流密度分析得到验证，结果显示灾害密度从极低区的0 events\u002Fkm²增加到极高区的0.048 events\u002Fkm²，且大多数历史泥石流事件集中在极高易发性类别内。SHAP分析识别出坡度、NDVI、降雨量、距道路距离和距河流距离是控制泥石流发生的最具影响力的孕灾因子。所提出的CNN–SHAP框架为区域尺度泥石流易发性评估提供了一种稳健且可解释的方法，适用于类似的山区环境。这些发现为数据稀缺山区的灾害减缓、基础设施规划和土地利用管理提供了有价值的信息。","Earth Surface Processes and Landforms","2026-09-27T00:00:00Z",{"impact":15,"substance":15,"depth":15,"authority":15,"freshness":15,"relevant":15,"comment":196},"泥石流易发性遥感与深度学习研究，属地质灾害领域，与三农、农业信息化无直接关联，不建议进入每日精选。",[198],{"name":193,"url":190},[21,200,22,201],"SHAP","地质灾害",[],{"status":27,"locked":14,"version":28,"method":29,"evidence":204},{"candidates":205,"agricultural_evidence":14,"reason":49},[206,210,214],{"code":43,"score":13,"supported":35,"title":207,"abstract":208,"tags":209},[47],[],[200,21],{"code":33,"score":44,"supported":35,"title":211,"abstract":212,"tags":213},[],[39,40],[22],{"code":215,"score":160,"supported":35,"title":216,"abstract":217,"tags":219},"I4",[],[218],"GIS",[],[221,222],"东兴都库什 泥石流 易发性","CNN SHAP 泥石流","东兴都库什泥石流易发性-3798","10.1002\u002Fesp.70419",{"doi":224,"openalex_id":226,"authors":227,"venue":193,"cited_by_count":15,"oa_url":236,"card":237,"direction":70,"ingested_from":72},"W7214538761",[228,231,233],{"name":229,"orcid":230},"Kashif Ullah","https:\u002F\u002Forcid.org\u002F0000-0003-2880-0977",{"name":232,"orcid":9},"Xi Chen",{"name":234,"orcid":235},"Tie Liu","https:\u002F\u002Forcid.org\u002F0000-0002-6879-4818","https:\u002F\u002Fonlinelibrary.wiley.com\u002Fdoi\u002Fpdfdirect\u002F10.1002\u002Fesp.70419",{"tldr":238,"method":239,"finding":240,"direction":70,"opportunity":241},"用CNN结合SHAP绘制巴基斯坦东兴都库什泥石流易发性图并解释关键因子。","CNN模型+SHAP，基于GIS与遥感13个30米条件因子。","高易发区占26.3%，坡度、NDVI、降雨、道路与河流距离最关键。","可将CNN-SHAP框架迁移至山区农业灾害风险与耕地安全评估，弥补数据稀缺区研究空白。","2026-09-29T23:30:30.418175Z",{"id":244,"title":245,"url":246,"summary":247,"summary_zh":248,"content":9,"source_name":249,"source_url":246,"published_at":250,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":251,"score_detail":252,"sources":259,"tags":261,"primary_topic":9,"secondary_topics":265,"classification":266,"search_phrases":300,"slug":303,"view_count":160,"doi":304,"paper":305,"created_at":326},3479,"Artificial Intelligence-Enabled Mangrove Ecosystem Monitoring Using Remote Sensing and Environmental Data","https:\u002F\u002Fdoi.org\u002F10.68012\u002Fair.v1i2.252","Mangrove ecosystems play a critical role in coastal protection, carbon sequestration, biodiversity conservation, and climate change mitigation; however, increasing anthropogenic pressures and environmental changes have accelerated mangrove degradation, creating an urgent need for efficient and scalable monitoring approaches. This study aims to develop an Artificial Intelligence-Enabled framework for monitoring mangrove ecosystem conditions by integrating remote sensing imagery with environmental datasets to improve the accuracy and timeliness of ecosystem assessment in tropical coastal regions. The proposed method combines multispectral satellite images, including vegetation indices derived from remote sensing data, with environmental variables such as temperature, precipitation, salinity, and tidal information, which are subsequently processed using a deep learning-based classification model to identify and categorize mangrove health conditions. Experimental evaluation demonstrates that the integration of remote sensing and environmental data significantly enhances model performance compared with approaches relying solely on satellite imagery, achieving high classification accuracy and improving the detection of early signs of ecosystem degradation. The findings further reveal that environmental parameters contribute substantially to distinguishing healthy, moderately degraded, and severely degraded mangrove areas across heterogeneous coastal environments. Consequently, the proposed framework provides an intelligent and reliable decision support tool for environmental monitoring agencies and policymakers while contributing to the development of resilient coastal ecosystem management and sustainable environmental governance in tropical archipelagic regions.","红树林生态系统在海岸防护、碳固存、生物多样性保护及气候变化减缓中发挥着关键作用；然而，日益加剧的人为压力和环境变化加速了红树林退化，亟需高效且可扩展的监测方法。本研究旨在开发一个人工智能驱动的框架，通过整合遥感影像与环境数据集来监测红树林生态系统状况，以提高热带沿海地区生态系统评估的准确性和时效性。所提出的方法将多光谱卫星影像（包括由遥感数据衍生的植被指数）与温度、降水、盐度和潮汐信息等环境变量相结合，随后使用基于深度学习的分类模型进行处理，以识别和分类红树林健康状况。实验评估表明，与仅依赖卫星影像的方法相比，遥感与环境数据的整合显著提升了模型性能，实现了较高的分类精度，并改善了对生态系统退化早期迹象的检测。研究结果进一步揭示，环境参数对于区分异质性沿海环境中健康、中度退化和严重退化的红树林区域具有重要贡献。因此，所提出的框架为环境监测机构和政策制定者提供了一种智能且可靠的决策支持工具，同时有助于热带群岛地区韧性沿海生态系统管理和可持续环境治理的发展。","AI Innovation and Resilience for the Environment (AIR)","2026-09-22T00:00:00Z",76,{"impact":253,"substance":254,"depth":255,"authority":256,"freshness":257,"relevant":76,"comment":258},18,21,17,12,8,"将遥感与环境数据融合的深度学习框架用于红树林健康监测，方法有创新且结论可靠，对沿海生态治理有参考价值，但属细分领域研究，公共影响有限。",[260],{"name":249,"url":246},[262,21,22,263,264],"农业人工智能","生态监测","红树林",[],{"status":27,"locked":14,"version":28,"method":29,"evidence":267},{"candidates":268,"agricultural_evidence":14,"reason":49},[269,275,280,287,294],{"code":43,"score":34,"supported":35,"title":270,"abstract":272,"tags":274},[271],"artificial intelligence",[21,47,271,273],"decision support",[262,21],{"code":33,"score":253,"supported":35,"title":276,"abstract":277,"tags":279},[40],[39,40,278],"multispectral",[22],{"code":281,"score":257,"supported":35,"title":282,"abstract":283,"tags":286},"H6",[],[284,285],"生物多样性","biodiversity",[263,264],{"code":288,"score":75,"supported":35,"title":289,"abstract":290,"tags":293},"H7",[],[291,292],"气候变化","climate change",[],{"code":295,"score":160,"supported":35,"title":296,"abstract":297,"tags":299},"H4",[],[298],"carbon sequestration",[],[301,302],"红树林 遥感 人工智能","红树林生态系统 监测","红树林遥感人工智能-3479","10.68012\u002Fair.v1i2.252",{"doi":304,"openalex_id":306,"authors":307,"venue":249,"cited_by_count":15,"oa_url":319,"card":320,"direction":325,"ingested_from":72},"W7213895268",[308,311,314,316],{"name":309,"orcid":310},"Dirvi Surya Abbas","https:\u002F\u002Forcid.org\u002F0000-0002-7819-2837",{"name":312,"orcid":313},"Asep Sutarman","https:\u002F\u002Forcid.org\u002F0009-0002-1029-7963",{"name":315,"orcid":9},"Ryan Davis",{"name":317,"orcid":318},"Maulana Abbas","https:\u002F\u002Forcid.org\u002F0009-0009-0137-9650","https:\u002F\u002Fjournal.sundarapublishing.com\u002Findex.php\u002Fair\u002Farticle\u002Fdownload\u002F252\u002F142",{"tldr":321,"method":322,"finding":323,"direction":70,"opportunity":324},"融合遥感影像与环境数据，用深度学习构建红树林生态系统健康监测框架。","多光谱卫星影像与植被指数，结合温度、降水、盐度、潮汐等环境变量，训练深度学习分类","融合环境数据显著提升分类精度，能更早识别红树林退化迹象，有效区分不同退化程度。","可探索多源时序遥感与环境数据融合的早期退化预警，并迁移至其他滨海湿地生态系统监测。","智慧农业 \u002F 农业物联网","2026-09-25T23:30:14.373467Z",{"id":328,"title":329,"url":330,"summary":331,"summary_zh":332,"content":9,"source_name":333,"source_url":330,"published_at":11,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":15,"score_detail":334,"sources":336,"tags":338,"primary_topic":341,"secondary_topics":345,"classification":346,"search_phrases":361,"slug":364,"view_count":54,"doi":365,"paper":366,"created_at":377},3015,"Forecasting Urbanization Dynamics On İstanbul’s European Side Using Deep Learning And Extrapolation Techniques","https:\u002F\u002Fdoi.org\u002F10.2478\u002Fjlecol-2026-0037","Abstract Rapid urbanization in metropolitan regions poses significant environmental, social, and infrastructural challenges, necessitating advanced analytical approaches to monitor and predict urban growth. This study investigates the spatio-temporal dynamics of urbanization on the European side of Istanbul from 2013 to 2024 using Landsat 8 imagery and a deep learning (DL)–based Land Cover Classification model integrated within ArcGIS Pro. The U-Net–based pre-trained model generated 15-class Land Use\u002FLand Cover (LULC) maps, which were validated against the Urban Atlas dataset, resulting in high classification accuracies for forest and water classes (PA: 0.84–0.94; UA: 0.87–0.87) and an overall binary urban\u002Fnon-urban accuracy of 87 %, confirming the robustness of the employed DL approach. Spatio-temporal analyses of LULC data were conducted using both Ordinary Least Squares (OLS) and nonlinear regression functions to examine urban growth trends and project future development for 2025, 2026, and 2027. The results indicate a strong linear increase in urbanized areas across most districts, with total developed area on the European side projected to reach approximately 807 km² by 2027, representing a nearly 50% increase compared to 2013. These findings highlight the significant pressure of urban expansion on natural and agricultural lands and emphasize the need for informed planning strategies. By integrating remote sensing, deep learning, and predictive modeling, this study provides actionable insights for sustainable urban development, offering a replicable framework for monitoring rapid urbanization and supporting policy decisions to mitigate environmental and socio-spatial impacts in rapidly growing metropolitan regions.","摘要 大都市区域的快速城市化带来了显著的环境、社会和基础设施挑战，亟需先进的分析方法来监测和预测城市增长。本研究利用Landsat 8影像和集成于ArcGIS Pro中的基于深度学习（DL）的土地覆盖分类模型，研究了2013年至2024年伊斯坦布尔欧洲一侧城市化的时空动态。基于U-Net的预训练模型生成了15类土地利用\u002F土地覆盖（LULC）地图，并依据Urban Atlas数据集进行了验证，森林和水体类别的分类精度较高（生产者精度PA：0.84–0.94；用户精度UA：0.87–0.87），城市\u002F非城市二分类总体精度达87%，证实了所采用深度学习方法稳健可靠。研究采用普通最小二乘法（OLS）和非线性回归函数对LULC数据进行时空分析，以考察城市增长趋势并预测2025年、2026年和2027年的未来发展。结果表明，大多数区域的城市化面积呈显著线性增长，预计到2027年欧洲一侧的总建成区面积将达到约807 km²，较2013年增长近50%。这些发现凸显了城市扩张对自然和农业用地的巨大压力，并强调了科学规划策略的必要性。通过整合遥感、深度学习和预测建模，本研究为可持续城市发展提供了可操作的见解，为监测快速城市化提供了一个可复制的框架，并支持旨在缓解快速增长的都市区域中环境和社会空间影响的政策决策。","Journal of Landscape Ecology",{"impact":15,"substance":15,"depth":15,"authority":15,"freshness":15,"relevant":15,"comment":335},"研究伊斯坦布尔城市扩张与土地覆盖预测，属城市遥感与景观生态领域，与三农、农业信息化、智慧农业无直接关联，不建议进入每日精选。",[337],{"name":333,"url":330},[21,22,339,340],"土地利用","城市扩张",{"code":43,"slug":342,"name":343,"path":344,"parent_code":96},"agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai",[],{"status":99,"locked":14,"version":28,"method":29,"evidence":347},{"candidates":348,"agricultural_evidence":35},[349,353,357],{"code":43,"score":148,"supported":35,"title":350,"abstract":351,"tags":352},[47],[21,47],[21],{"code":33,"score":44,"supported":35,"title":354,"abstract":355,"tags":356},[],[39,40],[22],{"code":281,"score":76,"supported":14,"title":358,"abstract":359,"tags":360},[],[],[339],[362,363],"伊斯坦布尔 城市扩张 遥感","U-Net 土地覆盖分类","伊斯坦布尔城市扩张遥感-3015","10.2478\u002Fjlecol-2026-0037",{"doi":365,"openalex_id":367,"authors":368,"venue":333,"cited_by_count":15,"oa_url":330,"card":372,"direction":70,"ingested_from":72},"W7213670913",[369],{"name":370,"orcid":371},"Gizem Dinç","https:\u002F\u002Forcid.org\u002F0000-0003-2406-604X",{"tldr":373,"method":374,"finding":375,"direction":70,"opportunity":376},"用Landsat 8影像和U-Net深度学习模型分析伊斯坦布尔欧洲侧2013-2024年城市化动态并","Landsat 8影像、ArcGIS Pro中U-Net预训练模型生成15类LU","城市面积呈强线性增长，2027年预计达807 km²，较2013年增长近50%，挤压自然与农业用地。","可借鉴该遥感+深度学习+外推框架，研究快速城市化对城郊农业用地与耕地保护的时空影响。","2026-09-20T23:30:21.335156Z",{"id":379,"title":380,"url":381,"summary":382,"summary_zh":383,"content":9,"source_name":384,"source_url":381,"published_at":385,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":386,"score_detail":387,"sources":392,"tags":394,"primary_topic":9,"secondary_topics":398,"classification":399,"search_phrases":421,"slug":424,"view_count":425,"doi":426,"paper":427,"created_at":447},2776,"Mapping coastal forest retreat using convolutional neural networks and different satellite imagery","https:\u002F\u002Fdoi.org\u002F10.1371\u002Fjournal.pone.0357346","Coastal forests are increasingly threatened by saturated soil and elevated salinity levels resulting from sea level rise, saltwater intrusion, and storm surges. In response to rising salinization and flooding, healthy coastal forests that rely on freshwater (both wetland forests and low-elevation upland forests) are transitioning into landscapes dominated by dead or dying trees, known as ghost forests. Situated among salt-tolerant shrubs and grasses, ghost forests eventually become marshes or open water. Here, our main objective was to quantify the dynamics and pathways of these forest landscape conversions, as well as the factors contributing to the changes, which is vital for understanding the progression of coastal ecosystem degradation and forecasting future changes. We focused first on identifying the best method to track forest landscape change by exploring the role of multiple remote sensing indices (i.e., multispectral, bi-seasonal, topographical, and phenological metrics) in enhancing the performance of deep learning models (convolutional neural networks, CNNs) for land cover classification in the coastal plain of North Carolina using surface reflectance of Landsat 8 and Sentinel-2 images. Then, we used the best available data (Landsat 8) to understand long-term change and identify patterns of land cover change from 1985 to 2021. Our study reveals that incorporating phenology and topographical indices enhances the separability of the ghost forests class from all other vegetation classes. In our assessment, the higher-resolution Sentinel-2 data (F1 Score = 96.3) outperformed Landsat images (F1 score = 93.4) for the 2021 co-available year. However, Landsat remains an important tool used due to its long-term data record. Therefore, we used Landsat to determine that 21% of forests were lost between 1985 and 2021, and that the rate of loss is increasing. Between 2010 and 2021, 23,876 ha of forest were converted to marsh, ghost forest, and shrub, which is 1.5 times higher than the 16,968 ha lost between 1985 and 2010. These conversions from forest to ghost forest and marshes were driven primarily by proximity to the channel, salinity, and the increasing rate of relative sea level rise (RSLR), which are the key environmental drivers of observed changes. By quantifying these changes, we highlight regions most vulnerable to environmental stressors, providing a basis for targeted conservation strategies.","沿海森林正日益受到海平面上升、盐水入侵和风暴潮导致的土壤饱和与盐度升高的威胁。为应对不断加剧的盐渍化和洪水，依赖淡水的健康沿海森林（包括湿地森林和低海拔高地森林）正在转变为以死亡或濒死树木为主的景观，即“幽灵森林”（ghost forests）。幽灵森林与耐盐灌木和草本植物交错分布，最终演变为沼泽或开阔水域。本研究的主要目标是量化这些森林景观转换的动态过程和路径，以及促成这些变化的因素，这对于理解沿海生态系统退化的进程和预测未来变化至关重要。我们首先聚焦于确定追踪森林景观变化的最佳方法，通过探索多种遥感指数（即多光谱、双季节、地形和物候指标）在提升深度学习模型（卷积神经网络，CNNs）土地覆盖分类性能中的作用，研究区域为北卡罗来纳州沿海平原，使用Landsat 8和Sentinel-2影像的地表反射率数据。随后，我们使用最佳可用数据（Landsat 8）来理解长期变化并识别1985年至2021年的土地覆盖变化模式。研究表明，纳入物候和地形指数可增强幽灵森林类别与所有其他植被类别的可分性。在我们的评估中，对于2021年两星共存的年份，较高分辨率的Sentinel-2数据（F1分数 = 96.3）优于Landsat影像（F1分数 = 93.4）。然而，Landsat因其长期数据记录仍是一个重要的使用工具。因此，我们使用Landsat确定1985年至2021年间21%的森林已经消失，且消失速率正在加快。2010年至2021年间，23,876公顷森林转换为沼泽、幽灵森林和灌木，比1985年至2010年间损失的16,968公顷高出1.5倍。这些从森林到幽灵森林和沼泽的转换主要由距河道距离、盐度和相对海平面上升（RSLR）速率的增加所驱动，这些是观测到的变化的关键环境驱动因素。通过量化这些变化，我们突出了最易受环境胁迫影响的区域，为有针对性的保护策略提供了依据。","PLoS ONE","2026-09-15T00:00:00Z",82,{"impact":253,"substance":388,"depth":253,"authority":389,"freshness":390,"relevant":76,"comment":391},23,14,9,"发表于PLoS ONE的研究用CNN结合Landsat与Sentinel-2影像量化海岸森林退化为幽灵森林的过程，方法新颖、数据跨度36年，对沿海生态保护与遥感应用有实质参考价值。",[393],{"name":384,"url":381},[21,22,395,396,397],"海岸防护林","土地覆盖变化","生态退化",[],{"status":27,"locked":14,"version":28,"method":29,"evidence":400},{"candidates":401,"agricultural_evidence":14,"reason":49},[402,406,410,417],{"code":33,"score":13,"supported":35,"title":403,"abstract":404,"tags":405},[],[39,40,278],[22],{"code":43,"score":44,"supported":35,"title":407,"abstract":408,"tags":409},[],[21,47],[21],{"code":411,"score":75,"supported":35,"title":412,"abstract":413,"tags":416},"H2",[],[414,415],"土壤","soil",[],{"code":281,"score":76,"supported":14,"title":418,"abstract":419,"tags":420},[],[],[396],[422,423],"土地覆盖变化 海岸防护林 深度学习 生态退化","土地覆盖变化 海岸防护林","土地覆盖变化海岸防护林深度学习生态退化-2776",2,"10.1371\u002Fjournal.pone.0357346",{"doi":426,"openalex_id":428,"authors":429,"venue":384,"cited_by_count":15,"oa_url":441,"card":442,"direction":325,"ingested_from":72},"W7213246924",[430,433,436,438],{"name":431,"orcid":432},"Tawakalitu Titilayo Tajudeen","https:\u002F\u002Forcid.org\u002F0000-0003-2405-134X",{"name":434,"orcid":435},"Marcelo Ardón","https:\u002F\u002Forcid.org\u002F0000-0001-7275-2672",{"name":437,"orcid":9},"Mirela Tulbure",{"name":439,"orcid":440},"Katherine L. Martin","https:\u002F\u002Forcid.org\u002F0000-0001-6020-9250","https:\u002F\u002Fjournals.plos.org\u002Fplosone\u002Farticle\u002Ffile?id=10.1371\u002Fjournal.pone.0357346&type=printable",{"tldr":443,"method":444,"finding":445,"direction":70,"opportunity":446},"用CNN结合多源遥感指数监测1985-2021年北卡海岸森林退化与幽灵林扩张。","Landsat 8与Sentinel-2影像，多光谱、物候、地形指数，卷积神经网","物候与地形指数提升幽灵林识别；1985-2021年21%森林消失且速率加快，近河道、盐度和海平面上升","可迁移该CNN框架到其他海岸带，融合时序Sentinel数据与海平面上升情景预测幽灵林未来扩张。","2026-09-17T23:30:15.391843Z"]