基于改进SURF的图像配准关键算法研究  

Research on the Key Algorithm for Image Matching Based on Improved SURF

作　　者：张开玉[1] 梁凤梅[1] 

机构地区：[1]太原理工大学信息工程学院,太原030024

出　　处：《科学技术与工程》2013年第10期2875-2879,共5页Science Technology and Engineering

基　　金：山西省科技产业化环境建设项目(2010061023)资助

摘　　要：针对图像特征匹配算法维数高,实时性低的缺陷,研究了SURF特征匹配算法,并对其进行了改进。首先在图像的尺度空间中提取SURF特征点,并生成扩展的特征描述向量。然后建立KD-Tree特征结构,采用BBF查询机制进行最近邻查询实现特征点快速匹配。实验结果表明,SURF算法进行特征检测的时间是SIFT算法的1/3;使用BBF进行特征匹配,匹配速度提高了2—3倍。With the defect of the high dimensions and low real time in image matching algorithm, the algorithm based on SURF is studied and improved. Firstly extract the feature points from the scale space of the image, and create extended feature vectors, then build KD-Tree feature structure, match the interest points by using BBF searching method. Experimental result express that, the time of detecting features by SURF algorithm is one third of SIIT algorithm; The matching speed by using BBF searching method is increased 2--3 times.

关 键 词：图像匹配Hessian矩阵 KD树 BBF 最近邻搜索 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]