视觉导航特征点匹配及误匹配剔除算法  

Algorithms of Feature Matching and Mismatched Points Eliminating in Vision Navigation

作　　者：徐超[1] 沈晓蓉[1] 罗宇锋[2] 范耀祖[1] 

机构地区：[1]北京航空航天大学自动化科学与电气工程学院,北京100083 [2]河南理工大学,河南焦作454000

出　　处：《计算机仿真》2010年第4期236-239,共4页Computer Simulation

摘　　要：飞行器视觉导航中图像的复杂性对特征点匹配提出了很高的要求。根据图像HSV色彩空间各分量固有的稳定性提出了一种新的特征点不变向量生成方法,以基于欧氏距离的最近邻准则作为特征点的相似度量将该算法提取的特征应用于图像特征点匹配;为了降低特征点描述向量的维数,提高匹配的实时性,采用了主成分分析(PCA)方法。针对误匹配问题,提出了一种利用惯性导航系统的输出信息进行误匹配特征点检测的方法。最后,通过实验证明,所提出的色彩匹配方法可以提高匹配的准确率,并且通过将PCA方法与上述方法结合不仅可以保持匹配的准确性还能降低计算的复杂度;所提出的误匹配检测方法可以较好的剔除误匹配点并能满足实时性的要求,这为视觉导航提供了一种可靠性更高的特征点匹配方法。The complexity of images in vision navigation has brought great challenge to feature points matching method. The paper put forward a new local invariant descriptor which utilized the HSV （Hue, Saturation, and Val- ue） color space information of images, and applied the method to feature points matching on the criterion of nearest neighbor based on Euclidean distance. In order to reduce the dimensions of the descriptor vectors the principal compo- nent analysis （PCA） was used to propose a new method to detect the mismatched points with the output of inertial navigation system. Finally, the experiments verified the validity of the matching method and pointed out that the method combined with PCA can not only keep the accuracy but also reduce the computation eomplieaey. The mis- matched points eliminating method can detect the mismatched ones effectively and fulfill the requirement of real - time calculation well. The paper provides a more reliable points matching method for vision navigation.

关 键 词：视觉导航 特征点匹配 误匹配点 惯性导航 主成分分析 

分 类 号：V249.325[航空宇航科学与技术—飞行器设计]