On-orbit MTF estimation and restoration of GF-2 satellite image
WANG Zhizhong1,2, ZHANG Qingjun1
1. China Institute Aerospace Science and Technology, Beijing 100086, China;
2. China Centre of Earth Resource Satellite Data and Application, Beijing 100094, China
调制传递函数（modulation transfer function，MTF）不仅是监测遥感卫星光学相机在轨运行情况和性能的有效手段，也是对卫星图像进行复原处理的重要参数。利用刀刃法对高分二号（GF-2）卫星全色相机进行MTF在轨测量。在计算MTF的过程中，使用汉明窗对截取后的线扩展函数（line spread function，LSF）曲线进行处理，以抑制截取过程所造成的频谱泄露。此外，还对截取后LSF曲线的两端补0，扩展LSF曲线的长度，以提高傅里叶变换时的采样频率。实验结果表明，使用本文刀刃法计算所得的MTF采样密度比传统方法提高了5倍，使得MTF曲线更为平滑，提高了MTF的数值精度，有利于后续的图像复原处理。基于传统刀刃法和本文刀刃法计算所得的MTF测量结果，采用维纳滤波法分别对GF-2全色图像进行复原处理研究。结果表明，对2种方法得到的MTF测量结果的维纳滤波均可明显提高图像的清晰度和边缘细节信息；但使用本文方法得到的MTF结果可使复原后的图像在对比度、边缘能量和平均梯度等关键指标上均优于使用传统方法得到的MTF结果。
Modulation transfer function(MTF) is not only an efficient method for monitoring the on-orbit satellite operation status and performance but also an important parameter which is often used to restore satellite image. In this paper, the knife edge method was used to measure the on-orbit MTF of GF-2 satellite panchromatic camera. During the calculation of MTF, Hamming window was used in the process of clipping the line spread function(LSF) in order to restrain the leak of frequency spectrum. Besides, the authors expanded the LSF with zeros so as to improve the sampling frequency during the Fourier transform. The experimental results show that the sampling density of MTF using the knife edge method proposed in this paper is 5 times more than that of the traditional method and the MTF curve is also smoother compared with that of the traditional method. Therefore, the more accurate MTF matrix value can be obtained to improve the performance of image restoration by this method. In this paper, the Wiener filtering method was used to restore the GF-2 satellite panchromatic image with the MTF matrix value. The experimental results also show that the MTF matrices computed by the two methods can all improve the clearness and object edge information of the image. However, the image restored by the MTF matrix values of the authors' method is superior to that of the traditional method in such characteristics as contrast, edge energy and average gradient.
[1] 顾行发,李小英,闵祥军,等.CBERS-02卫星CCD相机MTF在轨测量及图像MTF补偿[J].中国科学E辑工程与材料科学,2005,35(s1):26-40. Gu X F,Li X Y,Min X J,et al.In flight MTF monitoring and compensation for CCD camera on CBERS-02[J].China Science Series E Engineering & Materials Science,2005,35(s1):26-40.
[2] 王先华,乔延利,易维宁.卫星光学相机MTF在轨检测方法研究[J].遥感学报,2007,11(3):318-322. Wang X H,Qiao Y L,Yi W N.In-flight measurement of satellite optical camera MTF[J].Journal of Remote Sensing,2007,11(3):318-322.
[3] 蔡新明.基于卫星遥感图像的MTF计算和分析[D].南京:南京理工大学,2007:15-18. Cai X M.Calculation and Analysis of MTF Based on Satellite Remote Sensing Image[D].Nanjing:Nanjing University of Science and Technology,2007:15-28.
[4] 陈强,戴奇燕,夏德深.基于MTF理论的遥感图像复原[J].中国图象图形学报,2006,11(9):1209-1305. Chen Q,Dai Q Y,Xia D S.Restoration of remote sensing images based on MTF theory[J].Journal of Image and Graphics,2006,11(9):1209-1305.
[5] 柴雅琼,冯钟葵,齐东楷,等.改进的MTF遥感影像复原算法研究[J].遥感信息,2012(1):78-83. Chai Y Q,Feng Z K,Qi D K,et al.An improved MTFC restoration algorithm for remote sensing image[J].Remote Sensing Information,2012(1):78-83.
[6] 葛苹,王密,潘俊,等.高分辨率TDI-CCD成像数据的自适应MTF图像复原处理研究[J].国土资源遥感,2010,22(4):23-28.doi:10.6046/gtzyyg.2010.04.06. Ge P,Wang M,Pan J,et al.A study of adaptive MTF image restoration of high resolution TDI-CCD image data[J].Remote Sensing for Land and Resources,2010,22(4):23-28.doi:10.6046/gtzyyg.2010.04.06.
[7] 孟伟,金龙旭,李国宁,等.调制传递函数在遥感图像复原中的应用[J].红外与激光工程,2014,43(5):1690-1696. Meng W,Jin L X,Li G N,et al.Application of MTF in remote sensing image restoration[J].Infrared and Laser Engineering,2014,43(5):1690-1696.
[8] 张凡.基于改进NAS-RIF算法的遥感噪声图像自适应复原[J].国土资源遥感,2015,27(2):105-111.doi:10.6046/gtzyyg.2015.02.17. Zhang F.Self-adaptive restoration for remote sensing noise images based on improved NAS-RIF algorithm[J].Remote Sensing for Land and Resources,2015,27(2):105-111.doi:10.6046/gtzyyg.2015.02.17.
[9] 徐航,李传荣,李晓辉,等.一种优化的刃边法MTF在轨评估算法[J].遥感信息,2012,27(6):10-16. Xu H,Li C R,Li X H,et al.An optimized knife-edge algorithm for on-orbit MTF estimation[J].Remote Sensing Information,2012,27(6):10-16.
2016, Vol. 28 
