基于改进SURF算法的机器人识别匹配方法  

Robots Recognition and Matching Method Based on Improved SURF Algorithm

作　　者：王一璋 王亚刚[1] WANG Yi-zhang;WANG Ya-gang(College of Optical Electrical Information and Computer Engineering,University of ShangHai for Science and Technology,Shanghai 200093,China)

机构地区：[1]上海理工大学光电信息与计算机工程学院,上海200093

出　　处：《软件导刊》2018年第10期14-17,共4页Software Guide

基　　金：国家自然科学基金项目(61074087)

摘　　要：由于传统SURF匹配算法选取大量不符合预期的特征点,增加了后期匹配运算时间,导致不能满足工业级应用快速性的要求。提出一种改进的SURF算法,首先对摄像头获取的目标图像进行均值滤波处理,然后选择合理阈值、运用Canny算子对获取的目标图像进行边缘检测,再通过Hessian矩阵获取图像局部最值,并利用SURF算法对边缘图像进行匹配。仿真结果表明,该SURF算法在应用于工业机器人目标识别匹配时,既能减少匹配时间,又可以提高匹配准确度。Traditional SURF matching algorithm tends to pick a large number of feature points that do not meet expectations, which prolongs the time of matching operation later, leading to the inability to meet the requirements of rapidity in industrial ap plications. To solve this problem, an improved SURF algorithm was proposed in this article. Firstly, the target image acquired by the camera is subiected to mean filtering processing, then the reasonable threshold is selected to use the Canny operator to perfotto edge detection on the acquired target image. Furthermore the local maximum value of the image is obtained by Hessian matrix, and the edge image is matched by SURF algorithm. Experimental results show that the improved SURF algorithm can not only reduce the matching time but also improve the matching accuracy when it is applied to target recognition and matching of industrial robots.

关 键 词：CANNY边缘检测 SURF算法 图像匹配 目标识别 

分 类 号：TP31[自动化与计算机技术—计算机软件与理论]