交叉辐射定标是国际上新近发展起来的一种无场地定标方法,它的应用弥补了场地定标成本较高、定标参数更新周期较长的不足.对于我国2008年发射的环境与灾害监测预报小卫星CCD数据而言,探索交叉辐射定标方法的适用性,对及时发现传感器辐射性能的变化,促进CCD遥感数据的定量化应用具有重要意义.本研究以辐射定标精度较高的Terra/MODIS数据为参考,分别使用光线匹配法(RM)和辐射传输模型方法(RTM)对HJ-1B/CCD1数据进行交叉辐射定标,并与相同条件下进行的场地定标结果比较.实验结果表明,使用这两种方法获取的CCD1的第2、3、4波段的定标结果与场地定标结果差异较小,只有第1波段定标结果与场地定标结果差异相对较大,这证明了交叉辐射定标方法的有效性.另外,虽然RTM方法考虑了参考传感器和待定标传感器光谱响应和观测几何的差异,但是由于RTM方法会受到所使用的6S模型本身的误差以及输入的大气参数、地表参数测量误差的影响,该方法并不总是优于RM方法.%Remote sensor radiometric calibration is a prerequisite for quantitative remote sensing. The accuracy of radiomet ric calibration has a direct impact on the reliability and accuracy of quantitative application of remote sensing data. The radiomet ric performance of onboard sensor will decline with the time's going,therefore,the update of calibration coefficients periodically is very important to those precise remote sensing applications. For the lack of onboard calibration instrument and the expensive vicarious calibration,the updates of calibration coefficients to the small satellite constellation for environment and disaster moni toring( HJ satellites)are usually once a year, which is difficult to satisfy the requirements of quantitative remote sensing. With cross-calibration method,a dedicated sensor can be calibrated on a certain spectral channel against another sensor which has rather accurate calibration coefficients. Cross-calibration method is more economical since stable radiometric site and simultane ous ground measurements are no longer necessary. In this paper,the visible and nearinfrared spectral bands of HJ-1B/CCD1 are cross-calibrated against the corresponding bands of Terra/MODIS. Two types of cross-calibration methods , including RayMatching(RM)method and Radiative Transfer Modeling( RTM) method, are introduced using the HJ-1B/CCD1 and Terra/MODIS images acquired in Dunhuang Gobi desert area tn August,2009. Results were compared between the two cross-calibration methods,and the validity of cross-calibration was proved. At last, cross-calibration uncertainties are analyzed, mainly including dis crepancies of observed geometry and spectral difference between the two sensors. The uncertainty of the cross-calibration is low except for band 1,which have proved the validity of the cross- calibration methods.
展开▼
