一种应用于双目测距的SURF优化算法研究

Research on a SURF Optimization Algorithm for Binocular Ranging

针对传统SURF算法在图像匹配过程中,存在计算量大、算法精度低的问题,提出一种应用于双目测距系统的优化SURF算法。该优化算法可对预处理后的图像进行特征点检测和描述子构建,然后用FLANN算法进行初步匹配,再用最大和最小距离差值的阈值比筛选出匹配点对,进而利用半随机抽样PROSAC算法进行最后匹配。实验结果表明,优化后的SURF算法与传统SURF算法和加入RANSAC算法相比,误匹配率显著降低,分别从46.4%、11.8%降至6.4%。该算法在提高匹配精度方面优势明显,将其应用于目标测距系统中,最小测量误差仅为0.26%,能够满足精确测距要求。

In order to solve the problems of large amount of computation and low precision of the traditional SURF algorithm in the process of image matching,this paper proposes an optimized SURF algorithm applied to binocular ranging system.This optimization algorithm can detect feature points and construct descriptors of preprocessed images.Then the FLANN algorithm is used for preliminary matching,and then the threshold ratio of the maximum and minimum distance difference is used to screen out the matching point pairs,and then the semi-random sampling PROSAC algorithm is used for final matching.Experimental results show that compared with the traditional SURF algorithm and RANSAC algorithm,the mismatching rate of the optimized SURF algorithm is significantly reduced from 46.4%,11.8%to 6.4%,respectively.The algorithm has obvious advantages in improving the matching accuracy.When applied to the target ranging system,the minimum measurement error is only 0.26%,which can meet the requirements of accurate ranging.