QQ客服 官方微博 反馈留言

题名	星载相机微振动下图像补偿技术研究
作者	许博谦
学位类别	博士
答辩日期	2015-05
授予单位	中国科学院大学
导师	郭永飞
关键词	星载相机 图像补偿 振动 联合变换相关器 位移矢量测量
其他题名	Study on Image Compensation Technology for Spaceborne Cameras under Micro-vibration circumstances
学位专业	机械电子工程
中文摘要	随着星载相机角分辨率的逐步提高，对来自外界的扰动也越来越敏感。卫星平台的微振动是影响星载相机在轨动态成像质量的重要因素。针对微振动造成的星载相机视轴颤振，采用基于光学补偿方法的图像补偿系统可以直接、有效地补偿卫星平台微振动造成的成像质量退化。 简要介绍了论文所研究的课题背景，以及高分辨力星载相机的发展现状及其成像的主要特点。从图像补偿技术的发展现状出发，比较几种常用的图像补偿方法，并确定本文采用光学补偿方法。 重点研究了微振动对星载相机成像质量的影响。分析了卫星平台微振动的振动源与振动类型，并从中找出对影响星载相机成像质量的关键因素。通过数学建模，推导并仿真来自不同方向、不同类型、不同频率和振幅下的振动对星载相机成像质量的影响。 针对图像补偿技术的前端技术，也是最关键的技术——位移矢量探测方法开展研究。比较多种位移矢量探测方法后，选取联合变换相关器作为星载相机图像补偿位移矢量探测的方法。 提出了采用基于电子学方法实现联合变换相关器，凭借其在体积、重量、功耗以及可靠性等方面的优势，替代传统的光学联合变换相关器。发挥其在可编程方面的优势，改进了联合变换相关算法的计算过程，消除了零级衍射峰的影响，并通过计算机仿真试验验证了改进后算法的性能。 针对传统联合变换相关算法受探测器采样截止频率的局限——仅能测量1.0 pixel量级的相对位移量，改进算法，使其能够测量亚像元级的相对位移。并分别通过计算机仿真试验、半实物仿真实验验证了算法对于亚像元级位移量的测量精度。 解释了“滞后性”对光学补偿方法的影响。通过对比，得出了FPGA是最适合运行联合变换相关算法的处理器的结论。利用现有的实验条件，验证了联合变换相关算法的核心部分——二维FFT运算在FPGA芯片上的可实现性。最后通过仿真试验验证了联合变换相关算法在星载相机图像补偿系统中的作用。
英文摘要	With the developments of the angular resolution of spaceborne cameras, they are getting more and more sensitive to the disturbance of the surrounding environment. The micro-vibration comes from the satellite platform becomes the key factor which affects the imaging quality of these spaceborne cameras. To solve the problem which these micro-vibrations bring, it is necessary to bring in the image compensation system based on optical method, which could compensate the adverse effects directly and effectively. The backgrounds of this paper is briefly introduced, and also the development of modern high resolution spaceborne cameras. A few image compensation methods are compared and the optical compensation method is chosen. The adverse effect which micro-vibrations bring is deeply studied. Starting by analyzing the sources and types of the micro-vibrations, the main source is found which affects the imaging quality of the spaceborne cameras. By mathematical modeling, micro-vibrations effects from different direction, of different type, different frequency and different amplitude are simulated. Movement measurement technology is the front-end technology of image compensation. By comparing a few movement measurement methods, the joint correlation method is chosen to be the appropriate method for spaceborne cameras’ image compensation. A method of electronic joint correlator is introduced, which takes the advantage of volume, weight, power consumption, and also reliability, is a perfect replacement of traditional optical correlators. By reprograming the processor, the calculation process is simplified. The adverse effects are eliminated, and effectively proved by both simulation experiments and semi physical simulation test. To improve the accuracy of the joint correlation algorithm, which could only reach the precision of 1.0 pixel, some improvements have been made and are proved effective by computer simulations and semi physical experiments. The reason of optical compensation can still work under the“Delay Effect”is explained. After series of comparison, it comes to the conclusion of the most appropriate processor for joint correlation algorithm is FPGA chip. The key part of the joint correlation algorithm， 2D FFT is implemented on a FPG chip. The algorithm is simulated as part of a image compensation system, and its effect is proved.
公开日期	2015-12-24
内容类型	学位论文
源URL	[http://ir.ciomp.ac.cn/handle/181722/48930]
专题	长春光学精密机械与物理研究所_中科院长春光机所知识产出
推荐引用方式 GB/T 7714	许博谦. 星载相机微振动下图像补偿技术研究[D]. 中国科学院大学. 2015.

个性服务

查看访问统计

相关权益政策

暂无数据

收藏/分享

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。