基于因子分解和光束法平差从航空序列图像恢复三维结构  

3D Structure from the Aerial Sequential Image Based on Factorization and Bundle Adjustment

作　　者：邱志强[1] 陆宏伟[1] 张小虎[1] 于起峰[1] 

机构地区：[1]国防科技大学航天与材料工程学院,湖南长沙410073

出　　处：《国防科技大学学报》2004年第4期46-50,共5页Journal of National University of Defense Technology

摘　　要：从航空序列图像恢复目标三维结构是航空摄影测量的基本问题。针对现有算法中精度与计算量的矛盾,提出一种基于因子分解与光束法平差的新算法。作为因子分解与光束法平差的桥梁,提出基于"R T"模式共线方程的概念,详细推导了其线性化公式。仿真实验和真实图像实验表明,新算法精度高、稳定性好,可应用于航空测绘、虚拟现实、军事侦察等工程实践。3D structure from aerial image sequence is acquiring much attention in aerial photogrammetry. A novel method based on factorization and bundle adjustment,as well as its principle and implementation is presented. Being the bridge between factorization and bundle adjustment, the notion 'R-T' mode based collinearity equation is clarified as well as its linearization equations. Synthetic and real image test show the new method gives an accurate and robust result, and it can be applied to aerial surveying and mapping, virtual reality and military reconnaissance, etc.

关 键 词：航空图像 序列图像 因子分解 光束法平差 三维结构 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]