首页 | 本学科首页   官方微博 | 高级检索  
     检索      

超分辨率图像重构技术在航天器相对位置测量中的应用
引用本文:杜小平,赵继广,曾德贤.超分辨率图像重构技术在航天器相对位置测量中的应用[J].光学技术,2006,32(2):248-252.
作者姓名:杜小平  赵继广  曾德贤
作者单位:装备指挥技术学院,航天装备系,北京,101416
摘    要:以两颗航天器间相对位置参数测量为研究对象,确定了基于单目计算机视觉及目标特征的测量方案,提出了基于超分辩率图像重构技术的航天器间相对位置的高精度测量方法。分析讨论了图像传感器的观察模型、图像传感器及光学测量系统的点扩散函数、配准方法、重构原理及方法。利用地面试验验证了该方法的正确性。试验结果表明:该方法与基于单帧图像及目标特征的方法、基于插值的方法相比较,在测量精度及稳定性等方面都有了较大的提高;与基于单帧图像及目标特征的方法相结合,不仅可以保证在不同作用距离上相对位置的高精度测量,而且具有较高的测量稳定性和可靠性。

关 键 词:航天器  相对位置  计算机视觉  超分辨率图像重构
文章编号:1002-1582(2006)02-0248-05
收稿时间:2005/2/24
修稿时间:2005年2月24日

The application of the technology of the super-resolution image reconstruction in the relative position measure between spacecrafts
DU Xiao-ping,ZHAO Ji-guang,ZENG De-xian.The application of the technology of the super-resolution image reconstruction in the relative position measure between spacecrafts[J].Optical Technique,2006,32(2):248-252.
Authors:DU Xiao-ping  ZHAO Ji-guang  ZENG De-xian
Abstract:Regarding relative position parameters measure between spacecrafts as research object,the measure scheme based on monocular computer vision and target feature was determined,the testing method with high precision of relative position between spacecrafts based on super-resolution image reconstruction was put forward.The observe model of image sensor,the point spread function of image sensor and the optical measure system,image registration method,the theory and method of image reconstruction are analyzed and discussed.The results of ground prototype testing system verify the correctness of this method.Test result shows that compared with the method based on single frame image and the method of interpolation,this method greatly improve precision and stability.Combined with the method based on single image and target feature,this method can not only guarantee the relative position measure of high accuracy at different effective distance,but also ensure the higher stability and reliability.
Keywords:spacecraft  relative position  computer vision  super-resolution image reconstruction
本文献已被 CNKI 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号