Sun Tao, Long Hui, Liu Bao-Cheng, Li Ying
Opt Express. 2015 Jan 26;23(2):1395-410. doi: 10.1364/OE.23.001395.
Given the recent development in high-resolution (HR) optical satellites, the study of both attitude jitter (AJ) detection and compensation has become increasingly essential to improving the radiometric and geometric quality of HR images. A group of HR optical stereo mapping satellites in China, mapping satellite-1 (MS-1) has launched two satellites and will launch one satellite to build a satellite network. The geometric accuracy of the launched MS-1 satellites is greater than 80 m because of the AJ caused by the instability of the platform. AJ detection and compensation are critical issues that must be addressed to improve the accuracy of geo-positioning and mapping before launching a new satellite. The present study employs a method of jitter detection based on short-time asynchronous images to detect MS-1 jitter. The adjacent overlapping areas of an original panchromatic image are used as detection images instead of the traditional multispectral images, and a differential recursion optimal estimation filter is proposed for the optimal estimation and elimination of the gross errors of the registration data procedure, thereby increasing the detection accuracy. The space variant blurring model and viewing angles correction method are employed for the radiometric and geometric jitter compensation of images, respectively. The methods of radiometric objective evaluation indices and geometric checkpoint are then utilised to evaluate the quality of jitter compensation. Finally, the DeZhou regional image (ShanDong province, China) from MS-1 is used as the experimental data. Results for the AJ of MS-1 are analysed and reported for the first time. The assessment results obtained show that both radiometric and geometric qualities greatly increase after the jitter compensation procedure. Thus, the work of this study for jitter detection and compensation effectively addresses the jitter of MS-1 HR optical satellites.
鉴于高分辨率(HR)光学卫星的最新发展,姿态抖动(AJ)检测与补偿的研究对于提高HR图像的辐射和几何质量变得越来越重要。在中国,一组HR光学立体测绘卫星,即测绘卫星-1(MS-1)已发射两颗卫星,并将再发射一颗卫星以构建卫星网络。由于平台不稳定导致的AJ,已发射的MS-1卫星的几何精度大于80米。在发射新卫星之前,AJ检测与补偿是提高地理定位和测绘精度必须解决的关键问题。本研究采用基于短时异步图像的抖动检测方法来检测MS-1的抖动。原始全色图像的相邻重叠区域被用作检测图像,而不是传统的多光谱图像,并提出了一种差分递归最优估计滤波器,用于最优估计和消除配准数据过程中的粗大误差,从而提高检测精度。空间变模糊模型和视角校正方法分别用于图像的辐射和几何抖动补偿。然后利用辐射客观评价指标和几何检查点的方法来评估抖动补偿的质量。最后,将来自MS-1的德州区域图像(中国山东省)用作实验数据。首次分析并报告了MS-1的AJ结果。获得的评估结果表明,在抖动补偿过程之后,辐射和几何质量都有很大提高。因此,本研究中关于抖动检测与补偿的工作有效地解决了MS-1 HR光学卫星的抖动问题。