基于图像区域相关的景象匹配概率与精度研究  

Research on Acquisition Probability and Matching Precision of Scene Matching Systems Based on Area Correlation of the Image

作　　者：王刚[1] 段晓君[1] 王正明[1] 

机构地区：[1]国防科技大学理学院,长沙410073

出　　处：《宇航学报》2009年第3期1237-1242,1286,共7页Journal of Astronautics

基　　金：国家自然科学基金资助项目(60572136);航天支撑技术基金项目(2006-HT-GFKD-02)

摘　　要：景象匹配精度与正确匹配概率是决定景象匹配导航系统精度的重要指标,而图像独立象元数的大小是决定这两个精度指标的主要参数之一。分析了传统求取独立象元数方法存在的不足,并针对这些不足引入了图像区域相关的概念,建立了基于图像区域相关求解图像独立象元数的方法。理论推导和仿真分析表明,建立的求取独立象元数的方法合理,结果准确,速度较快,通过独立象元数计算得到的正确匹配概率符合仿真结果。该方法可以用于景象匹配辅助导航系统性能的预测及精确分析。The number of independent pixels （NIP） is one of the primary parameters determining scene matching precision and acquisition probability which are two important indexes for precision of scene matching navigation system. In view of deficiencies of previous methods for calculating NIP, this paper introduces the concept of image area correlation and further establishes a new method for calculating NIP based on area correlation. Theoretical deduction and simulation analysis show that the proposed method is reasonable, accurate and speedy, and results drawn from it conform to engineering background. Therefore, conclusions in this paper can be applied to prediction and precision analysis of the capability of scene matching aided navigation system.

关 键 词：独立象元 景象匹配 区域相关 匹配精度 正确匹配概率 

分 类 号：V448.2[航空宇航科学与技术—飞行器设计]