吴晓萍, 徐涵秋, 蒋乔灵. GF-1、GF-2与Landsat-8卫星多光谱数据的交互对比[J]. 武汉大学学报 ( 信息科学版), 2020, 45(1): 150-158. DOI: 10.13203/j.whugis20190084
引用本文: 吴晓萍, 徐涵秋, 蒋乔灵. GF-1、GF-2与Landsat-8卫星多光谱数据的交互对比[J]. 武汉大学学报 ( 信息科学版), 2020, 45(1): 150-158. DOI: 10.13203/j.whugis20190084
WU Xiaoping, XU Hanqiu, JIANG Qiaoling. Cross-Comparison of GF-1, GF-2 and Landsat-8 OLI Sensor Data[J]. Geomatics and Information Science of Wuhan University, 2020, 45(1): 150-158. DOI: 10.13203/j.whugis20190084
Citation: WU Xiaoping, XU Hanqiu, JIANG Qiaoling. Cross-Comparison of GF-1, GF-2 and Landsat-8 OLI Sensor Data[J]. Geomatics and Information Science of Wuhan University, 2020, 45(1): 150-158. DOI: 10.13203/j.whugis20190084

GF-1、GF-2与Landsat-8卫星多光谱数据的交互对比

Cross-Comparison of GF-1, GF-2 and Landsat-8 OLI Sensor Data

  • 摘要: 基于两组敦煌定标场的GF-1 PMS2、GF-2 PMS1与Landsat-8 OLI(operational land imager)同步影像,对3种传感器之间表观反射率数据的一致性进行交互对比。结果表明,两颗国产高分卫星的表观反射率数据具有很高的一致性,但它们与Landsat-8数据之间存在着一定的差异,其中尤以近红外波段的差异最为明显。在可见光波段,3种传感器的信号强度关系为:GF-2 PMS1 > GF-1 PMS2 > Landsat-8 OLI,在近红外波段的关系为:Landsat-8 OLI > GF-1 PMS2 > GF-2 PMS1。试验结果表明,所提出的数据转换模型能够有效地降低两种国产高分卫星传感器与Landsat-8 OLI在近红外波段的明显差异,且当选择与被转换影像地物比例一致性较高的模型进行转换时,转换精度能够进一步提高。

     

    Abstract: This paper aims at an analysis on the consistency of the top of atmosphere (TOA) reflectance among GF-1 PMS2, GF-2 PMS1 and Landsat-8 operational land imager(OLI) sensor data based on two synchronous image pairs. The result shows that TOA reflectance of GF-1 PMS2 and GF-2 PMS1 sensors has a high degree of agreement. Nevertheless, this paper also finds that TOA reflectance of either GF-1 PMS2 or GF-2 PMS1 data is less consistent with that of Landsat-8 OLI data, especially in the near-infrared band. In general, the rank of TOA reflectance in the blue, green and red bands of three sensors data is as follows:GF-2 PMS1 > GF-1 PMS2 > Landsat-8 OLI, while the relationship in the near-infrared band is:Landsat-8 OLI > GF-1 PMS2 > GF-2 PMS1. The coversion models among the three sensors data were obtained through regression analysis. The validation shows that the conversion equations can significantly reduce the difference in the near-infrared band among the three sensors. It is also found that when the image to be converted has similar land cover types and proportions with the image on which the conversion model was developed, the conversion accuracy can be improve.

     

/

返回文章
返回