[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-2790":3},{"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":23,"tags":25,"view_count":31,"doi":32,"paper":33,"created_at":61},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。",null,"GIScience & Remote Sensing","2026-09-16T00:00:00Z","论文",10,false,81,{"impact":17,"substance":18,"depth":17,"authority":19,"freshness":20,"relevant":21,"comment":22},18,22,14,9,1,"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。",[24],{"name":10,"url":6},[26,27,28,29,30],"农业人工智能","深度学习","遥感","土地覆盖","高分辨率制图",0,"10.1080\u002F15481603.2026.2726002",{"doi":32,"openalex_id":34,"authors":35,"venue":10,"cited_by_count":31,"oa_url":6,"card":54,"direction":58,"ingested_from":60},"W7213283296",[36,39,42,45,48,50,52],{"name":37,"orcid":38},"Xiangyu Nie","https:\u002F\u002Forcid.org\u002F0009-0001-5095-6401",{"name":40,"orcid":41},"Cong Lin","https:\u002F\u002Forcid.org\u002F0000-0001-5386-7343",{"name":43,"orcid":44},"Wei Zhang","https:\u002F\u002Forcid.org\u002F0000-0001-8162-9422",{"name":46,"orcid":47},"Hong Fang","https:\u002F\u002Forcid.org\u002F0000-0003-3707-0910",{"name":49,"orcid":9},"Zhen Dong",{"name":51,"orcid":9},"Sicong Liu",{"name":53,"orcid":9},"Zhaohui Xue",{"tldr":55,"method":56,"finding":57,"direction":58,"opportunity":59},"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。","农业遥感与作物表型","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。","openalex","2026-09-17T23:30:34.952106Z"]