[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-2670":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":66},2670,"A knowledge-guided machine learning framework for cross-scale wheat harvest monitoring via sample augmentation","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.rse.2026.115671","Accurate monitoring of wheat harvest is crucial for precision agriculture and ensuring food security. However, rapid changes in land surface composition during the harvest period in intensive agricultural regions make it difficult to obtain sufficiently high-confidence ground samples, limiting the performance and generalization of data-driven remote sensing methods. Therefore, this study proposes a Knowledge-Guided Machine Learning (KGML) framework that integrates multi-satellite Earth observation data (PlanetScope, Sentinel-2, and MODIS) to monitor harvest from the field to regional scales. Ground data were collected using vehicle-mounted cameras and smartphones during the 2023 and 2024 wheat harvest periods. The results showed that combining spectral knowledge rules with a Random Forest model (regional accuracy >0.80) generated numerous high-confidence augmented samples from PlanetScope imagery. The augmented dataset was used to train a Hybrid CNN-Transformer-LSTM (HCTL) model with two pathways: Sentinel-2 classification for field-level harvest mapping (overall accuracy = 0.93) and MODIS regression for sub-pixel harvest fraction estimation, which showed high agreement with PlanetScope-derived harvest fractions (R 2 = 0.97, RMSE = 0.07, rRMSE = 0.15). The harvest dates derived from the MODIS harvest fraction time series showed high consistency with field observations (R 2 = 0.82, RMSE = 1.30 days). This framework provides an effective solution for wheat harvest monitoring by bridging the gap between limited ground-truth data and multi-scale satellite observations, thereby supporting food security assessments and informed agricultural management decisions.","准确监测小麦收获对精准农业和保障粮食安全至关重要。然而，在集约化农业区域，收获期地表组成的快速变化使得获取足够高置信度的地面样本变得困难，限制了数据驱动遥感方法的性能和泛化能力。因此，本研究提出了一种知识引导机器学习（KGML）框架，集成多卫星地球观测数据（PlanetScope、Sentinel-2和MODIS），实现从田块到区域尺度的收获监测。地面数据通过车载摄像头和智能手机在2023年和2024年小麦收获期采集。结果表明，将光谱知识规则与随机森林模型相结合（区域精度>0.80），可从PlanetScope影像中生成大量高置信度增强样本。利用该增强数据集训练了混合CNN-Transformer-LSTM（HCTL）模型，该模型包含两条路径：Sentinel-2分类用于田块尺度收获制图（总体精度=0.93），MODIS回归用于亚像元收获比例估算，其结果与PlanetScope-derived收获比例高度一致（R²=0.97，RMSE=0.07，rRMSE=0.15）。由MODIS收获比例时间序列提取的收获日期与田间观测结果高度一致（R²=0.82，RMSE=1.30天）。该框架通过弥合有限地面真值数据与多尺度卫星观测之间的差距，为小麦收获监测提供了有效解决方案，从而支持粮食安全评估和农业管理决策。",null,"Remote Sensing of Environment","2026-09-15T00:00:00Z","论文",10,false,87,{"impact":17,"substance":18,"depth":19,"authority":20,"freshness":21,"relevant":22,"comment":23},22,23,18,15,9,1,"提出知识引导机器学习框架，融合多源卫星数据实现田块到区域尺度的跨尺度小麦收获监测，方法新颖、精度可靠，对精准农业与粮食安全评估有实质参考价值。",[25],{"name":10,"url":6},[27,28,29,30,31],"智慧农业","农业人工智能","粮食安全","遥感监测","小麦收获",0,"10.1016\u002Fj.rse.2026.115671",{"doi":33,"openalex_id":35,"authors":36,"venue":10,"cited_by_count":32,"oa_url":6,"card":59,"direction":63,"ingested_from":65},"W7213296259",[37,40,42,44,46,48,50,52,55,57],{"name":38,"orcid":39},"Mingchao Shao","https:\u002F\u002Forcid.org\u002F0000-0003-2619-4272",{"name":41,"orcid":9},"Chongya Jiang",{"name":43,"orcid":9},"Jingwei An",{"name":45,"orcid":9},"Haokai Zhu",{"name":47,"orcid":9},"Yue Li",{"name":49,"orcid":9},"Xia Yao",{"name":51,"orcid":9},"Tao Cheng",{"name":53,"orcid":54},"Hengbiao Zheng","https:\u002F\u002Forcid.org\u002F0009-0008-4778-0450",{"name":56,"orcid":9},"Weixing Cao",{"name":58,"orcid":9},"Yan Zhu",{"tldr":60,"method":61,"finding":62,"direction":63,"opportunity":64},"提出知识引导机器学习框架，用样本增强实现田块到区域尺度的冬小麦收获监测。","融合PlanetScope、Sentinel-2、MODIS与车载相机地面数据，","增强样本训练的HCTL模型田块分类精度0.93，区域收获比例R²=0.97，收获日期误差约1.3天。","农业遥感与作物表型","可迁移至其他作物收获监测，并探索知识规则自动化构建与跨区域泛化能力。","openalex","2026-09-16T23:30:30.474537Z"]