摘要
针对ASIFT算法抗大视角变换能力较好,但运算效率低的缺点,提出一种基于ORB的快速大视角图像匹配算法.该算法结合透视变换模型和ORB算法对ASIFT中的仿射变换模型和SIFT算法进行优化,在粗匹配算法获得单应性矩阵的基础上进行精匹配,有效减少了模拟次数,并提高了算法运算效率.实验结果表明,所提出算法具备抗视角变换能力,计算速度比ASIFT算法提高10倍,实时性强,工程使用价值高.
For the problem that the affine scale invariant feature transform（ASIFT） algorithm does well in large viewing image matching but has low computing efficiency, a fast large viewing image matching method based on the oriented FAST and rotated BRIEF（ORB） algorithm is proposed. The improved algorithm combines the perspective transformation model and ORB algorithm to optimize the affine transformation model and SIFT algorithm in the ASIFT algorithm. The refined matching is performed with the homography matrix based on coarse matching, which can reduce the number of simulation and improve the efficiency of the algorithm. The experimental results show that the proposed algorithm has the ability to resist the angle of view, and is 10 times faster than the ASIFT algorithm. Also, it has strong real-time performance and high engineering application value.
出处
《控制与决策》
EI
CSCD
北大核心
2017年第12期2233-2239,共7页
Control and Decision
基金
陆军装备部十三五预研项目(30102080101)
中央高校基本科研业务费专项资金项目(2017209)
航空科学基金项目(20165852052)
关键词
ORB
大视角
透视变换
仿射不变
ORB
large viewing
perspective transformation
affine transformation
