Please wait a minute...
img

官方微信

遥感技术与应用  2021, Vol. 36 Issue (1): 165-175    DOI: 10.11873/j.issn.1004-0323.2021.1.0165
遥感应用     
Sentinel⁃2A MSI 和Landsat 8 OLI两种传感器多光谱信息的交互对比
徐光志1,2(),徐涵秋1,2()
1.福州大学环境与资源学院,空间数据挖掘与信息共享教育部重点实验室,福建 福州 350116
2.福州大学遥感信息工程研究所,福建省水土流失遥感监测评价重点实验室,福建 福州 350116
Cross-comparison of Sentinel-2A MSI and Landsat 8 OLI Multispectral Information
Guangzhi Xu1,2(),Hanqiu Xu1,2()
1.Institute of Remote Sensing Information Engineering,Fuzhou University,College of Environment and Resources,Fuzhou University,Fuzhou 350116,China
2.Institude of Remote Sensing Information Engineering,Fujian Provincial Key Laboratory of Remote Sensing of Soil Erosion,Fuzhou 350116,China
 全文: PDF(5698 KB)   HTML
摘要:

在遥感对地观测中,受卫星发射、过空时间以及大气等因素的影响,单颗卫星获取的影像难以满足长时间序列的观测需求。因此,定量研究不同卫星平台传感器数据之间的关系是非常必要的。对Sentinel-2A MSI和Landsat-8 OLI传感器之间的定量关系进行了研究,基于两种传感器的3对同日过空的无云影像对,采用样区均值法和全试验区法对二者对应波段的光谱信息进行逐一对比,探求它们之间的定量关系。研究结果表明:Sentinel-2A MSI和Landsat 8 OLI的表观反射率数据总体比较一致,两种方法获得的R2均值分别为0.89 (全试验区法)和0.99 (样区均值法)。但二者也存在着差异,表现在Sentinel-2A MSI的表观反射率总体要比Landsat 8 OLI高约5%,且在不同的波段的表现有所不同。分析表明:二者之间的差异与其光谱响应函数、光谱范围以及试验区土地覆盖类型的不同有关。通过回归分析获得了两种传感器各对应波段数据的转换方程。验证结果表明,转换方程可以显著提高Sentinel-2A MSI和Landsat 8 OLI数据之间的一致性,为二者之间的协同使用提供了可行的方法。

关键词: Sentinel?2A MSILandsat?8 OLI表观反射率交互对比定标    
Abstract:

The images provided by an individual satellite are difficult to meet the requirement for a long time series observation due to the factors such as satellite operation duration, satellite passage time, atmosphere condition and others. Therefore, it is necessary to quantitatively study the relationship between different satellite sensor data for their collaborative use for a long time series earth observation. In this paper, the quantitative relationship between Sentinel-2A MSI and Landsat-8 OLI sensors’ data was studied. Based on three synchronous, cloud-free image pairs of the two sensors, the corresponding bands of the two sensors were compared band-by-band by using the Region Of Interest (ROI) method and the whole test area method to explore the quantitative relationship between them. The results show that the Top Of Atmospheric (TOA) reflectance data of Sentinel-2A MSI and Landsat-8 OLI are generally consistent, and the mean R2 values obtained via the two methods are 0.89 (whole test area method) and 0.99 (ROI method), respectively. However, the differences between the two sensors’ data have also been revealed. The total TOA reflectance of Sentinel-2A MSI is generally about 5% higher than that of Landsat-8 OLI, which, however, is different in different bands. The analysis shows that this difference is due to the two sensors’ differences in spectral response function and spectral range, as well as the difference in land cover type of the test areas. By regression analysis, the data conversion equations of each band between the two sensors are obtained. The validation results show that the conversion equations can significantly improve the consistency of Sentinel-2A MSI and Landsat-8 OLI data, providing a feasible method for the collaborative use of the two sensors’ data.

Key words: Sentinel-2A MSI    Landsat 8 OLI    TOA reflectance    Cross comparison    Calibration
收稿日期: 2020-02-04 出版日期: 2021-04-13
ZTFLH:  TP75  
基金资助: 国家自然科学基金项目(31971639)
通讯作者: 徐涵秋     E-mail: xuguangzhi223@163.com;hxu@fzu.edu.cn
作者简介: 徐光志(1994-),男,安徽安庆人,主要从事环境与资源遥感应用研究。E?mail:xuguangzhi223@163.com
服务  
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章  
徐光志
徐涵秋

引用本文:

徐光志,徐涵秋. Sentinel⁃2A MSI 和Landsat 8 OLI两种传感器多光谱信息的交互对比[J]. 遥感技术与应用, 2021, 36(1): 165-175.

Guangzhi Xu,Hanqiu Xu. Cross-comparison of Sentinel-2A MSI and Landsat 8 OLI Multispectral Information. Remote Sensing Technology and Application, 2021, 36(1): 165-175.

链接本文:

http://www.rsta.ac.cn/CN/10.11873/j.issn.1004-0323.2021.1.0165        http://www.rsta.ac.cn/CN/Y2021/V36/I1/165

图1  Landsat 8 OLI与Sentinel-2A MSI交互对比的影像对 (RGB: 8A,4,3波段)
影像对传感器日期时间太阳天顶角/°太阳方位角/°
克拉玛依Landsat 8 OLI2019-07-2613:08:1830.29139.40
Sentinel-2A MSI13:17:0128.02149.06
敦煌Landsat 8 OLI2019-08-1012:26:2030.81137.10
Sentinel-2A MSI12:27:1128.70143.67
哈密Landsat 8 OLI2019-06-1412:31:2525.50135.27
Sentinel-2A MSI12:37:0123.08143.52
表1  同步影像对参数
Sentinel-2A MSILandsat 8 OLI
波段号波段名空间分辨率/m波长范围/nm波段号波段名空间分辨率/m波长范围/nm
B1深蓝60433~453B1深蓝30435~451
B2蓝光10458~523B2蓝光30452~512
B3绿光10543~578B3绿光30533~590
B4红光10650~680B4红光30636~673
B8A近红外20855~875B5近红外30851~879
B11中红外1201 565~1 655B6中红外1301 566~1 651
B12中红外2202 100~2 280B7中红外2302 107~2 294
表2  Sentinel-2A MSI和Landsat 8 OLI对应波段的特性
图2  4类ROI样区 (方框中为样区,RGB: 8A,4,3波段)
图3  2个试验区的Landsat 8 OLI和Sentinel-2A MSI的表观反射率散点对比图
对应波段Landsat 8 OLISentinel-2A MSIME/%RMSER2
最小值最大值均值最小值最大值均值
深蓝0.119 20.254 70.163 60.122 50.259 70.173 05.730.011 10.991 6
蓝光0.098 20.256 70.152 10.096 20.268 30.157 83.720.008 70.997 2
绿光0.089 00.290 80.155 70.093 10.292 50.163 95.250.009 10.997 9
红光0.056 50.363 00.156 50.056 10.369 10.165 75.890.012 60.998 2
近红外0.159 00.646 30.400 60.167 00.659 40.413 63.260.013 60.999 6
中红外10.140 70.456 30.237 90.149 50.458 40.251 15.550.014 10.997 4
中红外20.049 40.408 50.168 70.052 90.407 30.177 55.230.010 40.998 6
表3  基于ROI对比的Sentinel-2A MSI与Landsat 8 OLI的统计特征

对应

波段

Landsat 8 OLISentinel-2A MSIME/%RMSER2
最小值最大值均值最小值最大值均值
克拉玛依深蓝0.118 40.190 00.123 80.121 40.166 20.126 72.340.004 30.620 3
蓝光0.095 60.192 00.103 00.094 90.204 20.101 5-1.360.003 50.821 7
绿光0.084 00.218 90.096 10.088 00.212 00.100 24.270.005 80.818 3
红光0.053 20.261 70.065 20.053 70.254 50.065 30.150.006 90.813 4
近红外0.246 50.655 20.574 70.252 50.666 00.587 02.140.017 90.909 0
中红外10.131 60.538 70.165 40.142 00.495 50.174 65.560.010 90.891 9
中红外20.043 30.821 60.063 40.047 80.850 00.067 25.990.009 30.856 7
敦煌深蓝0.114 60.522 20.201 90.136 10.417 60.216 27.080.014 80.932 8
蓝光0.094 60.556 90.198 70.097 70.655 10.210 76.040.012 60.950 4
绿光0.091 30.621 70.212 20.086 90.670 10.223 55.330.012 20.948 1
红光0.057 10.676 90.237 30.063 40.759 80.253 46.780.017 20.942 5
近红外0.035 00.721 40.256 30.038 90.781 00.270 75.620.015 30.962 4
中红外10.031 10.695 10.303 30.043 40.712 20.320 05.510.017 40.965 3
中红外20.024 60.894 30.269 80.030 10.672 00.282 94.860.014 10.959 8
表4  基于全试验区法的Sentinel-2A MSI与Landsat 8 OLI的统计特征
图4  基于试验影像建立的Landsat 8 OLI和Sentinel-2A MSI转换方程的拟合结果
对应波段RMSEME/%
模拟前模拟后变化/%模拟前模拟后变化/倍
深蓝0.007 90.004 0-48.554.67-0.924.08
蓝光0.006 00.002 8-52.663.40-0.2910.72
绿光0.007 00.005 1-26.993.75-1.401.68
红光0.009 10.005 6-38.854.38-1.571.79
近红外0.014 20.009 9-30.783.99-0.903.43
中红外10.013 30.006 8-48.984.86-0.597.23
中红外20.007 70.007 5-3.173.13-1.690.85
表5  哈密验证影像模拟转换前后的统计特征值对比
图5  Sentinel-2A MSI和Landsat 8 OLI的光谱响应函数
图6  Sentinel-2A MSI不同空间分辨率波段的像元数差异
波段原分辨率均值重采样为30 m分辨率均值偏差(%)
深蓝0.124 80.124 7-0.08
蓝光0.098 30.098 30.00
绿光0.095 60.095 60.00
红光0.058 40.058 3-0.17
近红外0.651 10.649 6-0.23
中红外10.180 60.180 2-0.22
中红外20.060 30.060 1-0.33
均值0.181 30.181 0-0.15
表6  Sentinel-2A MSI影像重采样对表观反射率的影响
图7  日本名古屋Landsat 8 OLI与Sentinel-2A MSI影像对(RGB: 8A,4,3波段)

对应

波段

Landsat-8 OLISentinel-2A MSIME/%RMSER2
最小值最大值均值最小值最大值均值
全试验区法深蓝0.110 20.206 10.117 80.112 30.151 50.119 01.020.001 70.840 0
蓝光0.085 50.195 90.092 90.080 00.239 10.089 0-4.200.004 20.782 2
绿光0.067 30.195 00.078 80.064 80.237 40.081 83.810.003 70.788 0
红光0.039 30.195 00.047 00.036 60.260 80.045 5-3.190.003 00.744 9
近红外0.118 40.608 20.348 60.136 10.600 50.361 93.820.021 80.914 5
中红外10.060 70.352 30.133 90.037 60.329 10.145 48.590.013 40.937 8
中红外20.024 10.276 70.053 80.029 40.246 80.058 07.810.005 30.925 1
表7  日本名古屋Landsat 8 OLI与 Sentinel-2A MSI的林地统计特征对比
1 Chander G, Coan M J, Scaramuzza P L. Evaluation and Comparison of the IRS-P6 and the Landsat Sensors[J]. IEEE Transactions on Geoscience and Remote Sensing, 2007, 46(1): 209-221. doi: 10.1109/TGRS.2007.907426.
doi: 10.1109/TGRS.2007.907426
2 Xu H, Zhang T. Assessment of Consistency in Forest-dominated Vegetation Observations between ASTER and Landsat ETM+ Images in Subtropical Coastal Areas of Southeastern China[J]. Agricultural and Forest Meteorology, 2013, 168: 1-9. doi: 10.1016/j.agrformet.2012.08.012.
doi: 10.1016/j.agrformet.2012.08.012
3 Song Junwei, Zhang Youjing, Li Xinchuan, et al. Comparison between GF-1 and Landsat 8 Images in Land Cover Classification[J].Progress in Geography,2016,35(2):255-263.
3 宋军伟,张友静,李鑫川,等.基于GF-1与Landsat-8影像的土地覆盖分类比较[J]. 地理科学进展,2016,35(2):255-263.
4 Mancino G, Ferrara A, Padula A, et al. Cross-comparison between Landsat 8 (OLI) and Landsat 7 (ETM+) Derived Vegetation Indices in a Mediterranean Environment[J]. Remote Sensing, 2020, 12(2): 291. doi: 10.3390/rs12020291.
doi: 10.3390/rs12020291
5 She X, Zhang L, Cen Y, et al. Comparison of the Continuity of Vegetation Indices Derived from Landsat 8 OLI and Landsat 7 ETM+ Data Among Different Vegetation Types[J]. Remote Sensing, 2015, 7(10): 13485-13506. doi: 10.3390/rs71013485.
doi: 10.3390/rs71013485
6 Li P, Jiang L, Feng Z. Cross-comparison of Vegetation Indices Derived from Landsat 7 Enhanced Thematic Mapper Plus (ETM+) and Landsat 8 Operational Land Imager (OLI) Sensors[J]. Remote Sensing, 2014, 6(1): 310-329. doi: 10.3390/rs6010310.
doi: 10.3390/rs6010310
7 Silvestri M, Romaniello V, Hook S, et al. First Comparisons of Surface Temperature Estimations between ECOSTRESS, ASTER and Landsat 8 over Italian Volcanic and Geothermal Areas[J]. Remote Sensing, 2020, 12(1): 184. doi: 10.3390/rs12010184.
doi: 10.3390/rs12010184
8 Chastain R, Housman I, Goldstein J, et al. Empirical Cross Sensor Comparison of Sentinel-2A and 2B MSI, Landsat 8 OLI, and Landsat 7 ETM+ Top of Atmosphere Spectral Characteristics over the Conterminous United States[J]. Remote Sensing of Environment, 2019, 221: 274-285. doi: 10.1016/j.rse.2018.11.012.
doi: 10.1016/j.rse.2018.11.012
9 Barsi J A, Alhammoud B, Czapla-Myers J, et al. Sentinel-2A MSI and Landsat 8 OLI Radiometric Cross Comparison over Desert Sites[J]. European Journal of Remote Sensing, 2018,51(1):822-837. doi: 10.1080/22797254.2018.1507613.
doi: 10.1080/22797254.2018.1507613
10 Arekhi M, Goksel C, Balik Sanli F, et al. Comparative Evaluation of the Spectral and Spatial Consistency of Sentinel-2 and Landsat 8 OLI Data for Igneada Longos Forest[J]. ISPRS International Journal of Geo-Information, 2019, 8(2): 56. doi: 10.3390/ijgi8020056.
doi: 10.3390/ijgi8020056
11 Zhang H K, Roy D P, Yan L, et al. Characterization of Sentinel-2A and Landsat 8 Top of Atmosphere, Surface, and Nadir BRDF Adjusted Reflectance and NDVI Differences[J]. Remote Sensing of Environment, 2018, 215: 482-494. doi: 10.1016/j.rse.2018.04.031.
doi: 10.1016/j.rse.2018.04.031
12 Mandanici E, Bitelli G. Preliminary Comparison of Sentinel-2 and Landsat 8 Imagery for a Combined Use[J]. Remote Sensing, 2016, 8(12): 1014. doi: 10.3390/rs8121014.
doi: 10.3390/rs8121014
13 Zhong Hanxiao, Bian Jinhu, Li Ainong. Radiometric Consistency between Landsat 8 OLI and Sentinel-2 MSI Imagery in Mountainous Terrain[J]. Remote Sensing Technology and Application, 2018, 33(3): 428-438.
13 钟函笑, 边金虎, 李爱农. Landsat-8 OLI与Sentinel-2 MSI山区遥感影像辐射一致性研究[J]. 遥感技术与应用, 2018, 33(3): 428-438.
14 Zhang Yi, Xia Jianxin, Du Lei, et al. Comparison and Collaborative Application of Water Index based on Landsat 8 and Sentinel-2A Data: A Case Study in Heihe River Basin, Northwest China[J]. Journal of Basic Science and Engineering, 2020, 28(6): 1361-1368.
14 张戈, 夏建新, 杜蕾, 等. Landsat 8和Sentinel-2A的水体指数比较与协同应用——以黑河流域典型区为例[J]. 应用基础与工程科学学报, 2020, 28(6):1361-1368.
15 Wang Q, Li J, Jin T, et al. Comparative Analysis of Landsat 8, Sentinel-2, and GF-1 Data for Retrieving Soil Moisture over Wheat Farmlands[J]. Remote Sensing, 2020, 12(17): 2708. doi: 10.3390/rs12172708.
doi: 10.3390/rs12172708
16 Teillet P M, Barker J L, Markham B L, et al. Radiometric Cross-calibration of the Landsat 7 ETM+ and Landsat 5 TM Sensors based on Tandem Data Sets[J]. Remote Sensing of Environment, 2001, 78(1): 39-54. doi: 10.1016/S0034-4257(01)00248-6.
doi: 10.1016/S0034-4257(01)00248-6
17 USGS. Landsat 8 Data Users Handbook [DB/OL].https://prd-wret.s3-us-west-2.amazonaws.com/assets/palladium/production/atoms/files/LSDS-1574_L8_Data_Users_Handbook-v5.0.pdf, 2016.
18 ESA. Sentinel-2 User Handbook[DB/OL]. https://Sentinel.esa.int/documents/247904/685211/ Sentinel-2_User_Handbook, 2015.
19 Jiang Qiaoling, Xu Hanqiu. Cross-Comparison between GF-1PMS2 and GF-2 PMS2 Sensor Data[J]. Remote Sensing Technology and Application, 2018, 33(6):1084-1094.
19 蒋乔灵, 徐涵秋. GF-1 PMS2与GF-2 PMS2传感器数据的交互对比[J]. 遥感技术与应用, 2018, 33(6): 1084-1094.
20 Li Chunhua, Xu Hanqiu, Chen Licong. Cross-Comparison between ASTER and Landsat 7 ETM+ Multispectral Imagery[J]. Spectroscopy and Spectral Analysis, 2010, 30(9): 2518-2524.
20 李春华,徐涵秋,陈荔聪. ASTER和Landsat-7 ETM+两种多光谱传感器影像的交互对比[J]. 光谱学与光谱分析, 2010, 30(9): 2518-2524.
21 Liu Q, Yu T, Gao H. Radiometric Cross-calibration of GF-1 PMS Sensor with a New BRDF Model[J]. Remote Sensing. 2019, 11(6): 707. doi: 10.3390/rs11060707.
doi: 10.3390/rs11060707
22 Chen Y, Sun K, Li D, et al. Radiometric Cross-calibration of GF-4 PMS Sensor based on Assimilation of Landsat 8 OLI Images[J]. Remote Sensing. 2017, 9(8): 811. doi: 10.3390/rs9080811.
doi: 10.3390/rs9080811
23 Li Juan, Feng Lian, Pang Xiaoping. Comparison of the Cross-calibration Methods between Image-based and RTM-BRDF for GF-1 Images[J]. Acta Geodaetica et Cartographica Sinica, 2017, 46(7):882-890.
23 李娟,冯炼,庞小平. 针对GF-1遥感影像的基于影像与基于辐射传输模型的两种交叉定标方法比较[J]. 测绘学报, 2017, 46(7):882-890.
[1] 谷松岩,郭杨,窦芳丽,吴琼,卢乃锰. 风云三号微波大气探测载荷辐射定标[J]. 遥感技术与应用, 2021, 36(1): 141-154.
[2] 李娇阳,王振占,谷松岩,张升伟. 星载微波大气湿度探测仪再定标共性技术分析[J]. 遥感技术与应用, 2019, 34(6): 1212-1220.
[3] 邓翔,阎敬业,吴季,蓝爱兰. AgileDARN雷达内定标的方法与实现[J]. 遥感技术与应用, 2019, 34(6): 1221-1226.
[4] 张宇飞,蒋兴伟,马超飞. 基于现场观测的星载雷达高度计定标方法进展[J]. 遥感技术与应用, 2019, 34(6): 1190-1196.
[5] 王振占,李娇阳. 风云三号卫星微波探测仪定标原理及算法[J]. 遥感技术与应用, 2019, 34(6): 1197-1204.
[6] 谷松岩,王振占,卢乃锰,杨忠东,李靖,张升伟,杨磊,吴雪宝. 风云三号微波湿度计航空校飞辐射定标原理及数据分析[J]. 遥感技术与应用, 2019, 34(6): 1205-1211.
[7] 庞博, 马灵玲, 刘耀开, 王宁, 赵永光, 韩启金. 陆地卫星光学载荷地基自动辐射定标与验证分析[J]. 遥感技术与应用, 2019, 34(1): 146-154.
[8] 蒋乔灵, 徐涵秋. GF-1 PMS2与GF-2 PMS2传感器数据的交互对比[J]. 遥感技术与应用, 2018, 33(6): 1084-1094.
[9] 牛明慧,陈福春. 基于月球反射的遥感仪器在轨定标方法[J]. 遥感技术与应用, 2018, 33(2): 337-341.
[10] 王振占,董帅,殷晓斌,陆浩,李彬. 有源微波冷噪声源的原理及其应用分析[J]. 遥感技术与应用, 2017, 32(2): 247-254.
[11] 龚循武,郭伟,万珺之. 基于有源定标器的HY-2高度计后向散射系数在轨定标方法研究[J]. 遥感技术与应用, 2017, 32(2): 374-379.
[12] 陈怡羽,郭强. 中短波红外恒星定标点扩散函数求解及精度分析[J]. 遥感技术与应用, 2017, 32(1): 158-165.
[13] 王功雪,蒋玲梅,武胜利,刘晓敬,郝诗睿. FY-3B与FY-3C/MWRI交叉定标及雪深算法应用[J]. 遥感技术与应用, 2017, 32(1): 49-56.
[14] 陈卫英,刘高峰,何嘉恺,潘莉. 降低微波成像仪接收机非线性偏差方法研究[J]. 遥感技术与应用, 2017, 32(1): 121-125.
[15] 徐曦煜,刘和光,杨双宝. 星载雷达高度计沙漠散射特性及定标方法研究[J]. 遥感技术与应用, 2016, 31(5): 893-899.