【摘要】 遥感技术已成为农业信息提取的重要手段。为探究不同遥感数据源下的农作物精细化识别与分类，选取广西壮族自治区贺州市八步区东融供港蔬菜产业示范区为研究区，基于Planet、GF6 WFV、Landsat 8 OLI影像数据，利用支持向量机分类算法，对研究区豆杯、学斗、青仔、尖叶菜心、芥蓝等不同农作物进行识别与提取，通过类别可分离性、总体分类精度、Kappa系数、光谱变化、成图效果等几个方面对提取效果进行评价，结果表明GF6 WFV影像是研究区农作物识别与提取的最佳数据源。更多还原

【Abstract】 Remote sensing technology has become an important means of extracting agricultural information. In order to explore the identification and classification of crops of different remote sensing data sources, we selected Babu District Dongrong vegetable industry demonstration zone supplied to HongKong as the research area. Base on Planet, GF6 WFV, Landsat 8 OLI remote sensing images, we used the support vector machine method to identify and extract different crops of the tip leaves of bean, Xuedou, Qingzai, vegetable heart of pointed leaf, cabbage mustard. We also evaluated the extraction effect through class separability, overall classification, Kappa coefficient, spectral variation and mapping effect. Results showed that GF6 WFV images were the best resource for crop recognition and extraction in the study area.更多还原