一种基于卷帘快门CMOS成像的空间相机颤振检测方法  

Jitter Detection Method Based on Rolling Shutter CMOS Imaging for Space Cameras

作　　者：赵薇薇[1] 范城城 王艳[1] 商鼎会[2,3,4] 张永合 尹增山[2,3,4] Zhao Weiwei;Fan Chengcheng;Wang Yan;Shang Dinghui;Zhang Yonghe;Yin Zengshan(Institute of Beijing Remote Sensing Information,Beijing 100192,China;Shanghai Engineering Center for Microsatellites,Shanghai 201210,China;Innovation Academy for Microsatellites of CAS,Shanghai 201210,China;Key Laboratory of Microsatellites of CAS,Shanghai 201210,China)

机构地区：[1]北京市遥感信息研究所,北京100192 [2]上海微小卫星工程中心,上海201210 [3]中国科学院微小卫星创新研究院,上海201210 [4]中国科学院微小卫星重点实验室,上海201210

出　　处：《光学学报》2022年第2期234-241,共8页Acta Optica Sinica

基　　金：国家自然科学基金(42001408);中国科学院青促会项目(2017357);中国科学院国防科技重点实验室基金(CXJJ-19S012);中国科学院战略性先导科技专项(A类);空间科学背景型号项目(XDA15020305)。

摘　　要：星上运动部组件所引起的颤振是当前影响超高分辨率、超高机动成像质量的一个重要因素。针对该信号难以被精确业务化检测识别的问题,提出一种基于卷帘快门CMOS成像的空间相机颤振检测方法,进一步采用高频角位移与高精度参考的数字正摄影像/数字高程模型(DOM/DEM)数据进行效果验证与评价。对2020年发射的某高分型号卫星进行了分析,分析结果表明所设计的检测方法可以有效识别空间相机成像过程中的颤振特性,颤振频率主要集中在156 Hz左右,补偿后的全色影像相对几何精度可提升1.2 pixel左右。The jitter caused by onboard motion components is an important factor affecting the quality of ultra-high-resolution and super-maneuverability imaging. Given that it is difficult to conduct accurate operational detection and identification of the jitter signal, this paper proposes a jitter detection method based on rolling shutter CMOS imaging for space cameras and verifies and evaluates the method with digital orthophoto map/digital elevation model(DOM/DEM) data by using high-frequency angular displacement and high-precision reference. A Gaofen satellite launched in 2020 is discussed as an example. The analysis results show that the proposed detection method can effectively identify the jitter characteristics in the imaging process of space cameras, and the jitter frequency is mainly concentrated at about 156 Hz. The relative geometric accuracy of panchromatic images after compensation is improved by about 1.2 pixel.

关 键 词：遥感 超高分辨率卫星 卷帘快门CMOS 颤振检测 高频角位移 分时成像 

分 类 号：P236[天文地球—摄影测量与遥感]