[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-2527":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":24,"tags":26,"view_count":32,"doi":33,"paper":34,"created_at":62},2527,"Extracting Summer-Harvested Crops in the Baojixia Irrigation District Using CycleGAN and Transfer Learning","https:\u002F\u002Fdoi.org\u002F10.3390\u002Frs18183155","Remote sensing-based mapping of crop planting structures in irrigation districts plays a vital role in forecasting regional production and optimizing water resource allocation. However, optical satellite imagery is limited by insufficient spatial resolution when applied to fragmented farmland landscapes, while unmanned aerial vehicle (UAV) imagery is limited by spatial coverage and high data processing costs. To address these bottlenecks, this study proposed a cross-scale collaborative extraction framework utilizing the CycleGAN network and transfer learning to map summer harvest crops (winter wheat and rapeseed) in the Baojixia Irrigation District for the year 2023. First, multiple semantic segmentation models—including U-Net, DeepLabv3+, SegFormer, and HRNet—were evaluated on a joint satellite–UAV dataset, with U-Net selected as the optimal backbone. Next, CycleGAN was introduced to perform style translation from the UAV domain to the satellite domain. This step generated high-fidelity, satellite-like images that preserve UAV-derived high-resolution spatial details, which were subsequently used to pre-train the U-Net backbone, significantly reducing the labor of manual annotation. Finally, the model was fine-tuned with real satellite images to achieve precise crop extraction. Results indicated that this framework accelerates model convergence and improves segmentation accuracy. The proposed method achieved an mIoU of 85.09%, an mPA (Recall) of 91.63%, a Precision of 91.97%, an Accuracy of 93.57%, and an F1-Score of 91.80%, outperforming the baseline U-Net model by 2.98%, 2.00%, 1.66%, 1.52%, and 1.83%, respectively. By successfully transferring high-resolution prior knowledge into the satellite feature space, this study provides a cost-effective and highly accurate solution for crop identification in complex agricultural landscapes, breaking the spatial limitations of UAV remote sensing.","基于遥感的灌区作物种植结构制图对于区域产量预测与水资源优化配置具有重要作用。然而，光学卫星影像在应用于破碎化农田景观时受限于空间分辨率不足，而无人机（UAV）影像则受限于空间覆盖范围和数据获取成本高。为解决这些瓶颈问题，本研究提出了一种利用CycleGAN网络和迁移学习的跨尺度协同提取框架，用于绘制2023年宝鸡峡灌区夏收作物（冬小麦和油菜）分布图。首先，在卫星—无人机联合数据集上评估了多种语义分割模型——包括U-Net、DeepLabv3+、SegFormer和HRNet——并选择U-Net作为最优主干网络。其次，引入CycleGAN进行从无人机域到卫星域的风格转换。该步骤生成了高保真、类卫星影像，同时保留了无人机来源的高分辨率空间细节，随后用于预训练U-Net主干网络，显著减少了人工标注的工作量。最后，利用真实卫星影像对模型进行微调，以实现精确的作物提取。结果表明，该框架加速了模型收敛并提高了分割精度。所提方法达到了85.09%的mIoU、91.63%的mPA（召回率）、91.97%的精确率、93.57%的总体精度和91.80%的F1分数，分别优于基线U-Net模型2.98%、2.00%、1.66%、1.52%和1.83%。通过成功将高分辨率先验知识迁移至卫星特征空间，本研究为复杂农业景观中的作物识别提供了一种经济高效且高精度的解决方案，突破了无人机遥感的空间局限性。",null,"Remote Sensing","2026-09-14T00:00:00Z","论文",10,false,78,{"impact":17,"substance":18,"depth":19,"authority":20,"freshness":21,"relevant":22,"comment":23},16,22,18,14,8,1,"提出CycleGAN跨尺度协同框架，将无人机高分辨率先验迁移至卫星影像，实现灌溉区夏收作物高精度提取，方法新颖、指标扎实，对农业遥感监测有实用价值。",[25],{"name":10,"url":6},[27,28,29,30,31],"智慧农业","遥感","作物识别","迁移学习","灌溉区",0,"10.3390\u002Frs18183155",{"doi":33,"openalex_id":35,"authors":36,"venue":10,"cited_by_count":32,"oa_url":6,"card":55,"direction":59,"ingested_from":61},"W7212518427",[37,39,41,43,45,47,50,52],{"name":38,"orcid":9},"Zili Chen",{"name":40,"orcid":9},"Zhilong Gao",{"name":42,"orcid":9},"Zefeng Jia",{"name":44,"orcid":9},"Pengjie Pan",{"name":46,"orcid":9},"Wen Gao",{"name":48,"orcid":49},"Jun Zhang","https:\u002F\u002Forcid.org\u002F0000-0001-6972-6828",{"name":51,"orcid":9},"Zijie Niu",{"name":53,"orcid":54},"Dongyan Zhang","https:\u002F\u002Forcid.org\u002F0000-0003-3509-7482",{"tldr":56,"method":57,"finding":58,"direction":59,"opportunity":60},"用CycleGAN和迁移学习融合无人机与卫星影像，提取宝鸡峡灌区夏收作物。","CycleGAN风格迁移+U-Net迁移学习，联合卫星-无人机数据集。","框架提升分割精度，mIoU达85.09%，优于基线U-Net。","农业遥感与作物表型","可探索跨尺度迁移学习在更多作物和区域的应用，降低标注成本。","openalex","2026-09-15T23:30:20.433643Z"]