SARINS组合导航中基于SURF的鲁棒景象匹配算法.docVIP

SAR/INS组合导航中基于SURF的鲁棒景象匹配算法 熊智 陈方(南京航空航天大学南京 210016) 在SAR/INS组合导航系统中，由于合成孔径雷达采用正侧视成像工作方式，会引起SAR图像的严重变形，而且获取的SAR图像还可能存在严重的斑点噪声。为了适应SAR图像的几何畸变和高斑点噪声，需要提取出的图像特征具有较高的鲁棒性。本文提出了基于SURF的导航用鲁棒景象匹配算法，算法首先针对惯性组合导航的工作特点，对SURF特征匹配进行了改进和优化设计，然后用RANSAC方法过滤掉错误和低精度的匹配点，最后，进行最小二乘精确匹配获取航向和位置偏差信息。分析了算法对SAR图像的适应性，抗斑点噪声性能，匹配精度以及实时性，并与基于SIFT特征的景象匹配算法进行了对比。结果表明算法性能优越，在匹配适应性、鲁棒性、匹配精度及匹配速度方面都优于SIFT算法，可以满足SAR/INS组合导航系统匹配修正的高性能要求。 惯性组合导航系统；景象匹配辅助导航；SURF；SIFT；图像匹配 中图法分类号: TP391.41 　 文献标识码: A XIONG Zhi， CHEN Fang， WANG Dan， LIU Jian-ye (Navigation Research Center, Nanjing University of Aeronautics and Astronautics, Nanjing 210016,China) Abstract: In the SAR/INS integrated navigation system, as the use of positive side-looking synthetic aperture radar (SAR) imaging work methods,SAR images can cause a serious distortion,and SAR images maybe existing serious speckle noises.Image features must have high robustness in order to adapt speckle noise of SAR image. In this paper, we propose the robust scene matching algorithm based on SURF for navigation. Firstly, we optimize the process of SURF feature matching especially for the inertial integrated navigation. Then, we adopted RANSAC algorithm to remove the false and low precision matching points. Finally, the least square algorithm is applied for getting both the aircraft position errors and course deviation. In the experiment, we analyze adaptability of the algorithm for the different SAR images, performance of speckle noise resistance, matching accuracy and real-time performance. The experiment result demonstrates the superiority of algorithm in matching speed, accuracy, adaptability and robustness in contrast with scene matching algorithm based on SIFT fetures, so it can meet the high performance needs for matching navigation in the SAR/INS integrated navigation system. Key words: inertial integrated navigation system; scene matching aided navigation; speeded-up robust features; scale-invariant feature transform; image matching 1 引言 目前很多飞行器