基于边缘信息的SIFT图像匹配算法  

SIFT Image Matching Algorithm Based on Edge Information

作　　者：刘克平[1] 李西卫 李岩[1] 于微波[1] LIU Keping;LI Xiwei;LI Yan;YU Weibo(College of Electrical and Electronic Engineering, Changchun University of Technology, Changchun 130012, China)

机构地区：[1]长春工业大学电气与电子工程学院,吉林长春130012

出　　处：《吉林大学学报（信息科学版）》2018年第2期200-206,共7页Journal of Jilin University（Information Science Edition）

基　　金：吉林省省级产业创新专项资金资助项目(2018C038-2);吉林省科技发展计划基金资助项目(20170204067GX);长春市科技发展计划基金资助项目(17DY032)

摘　　要：鉴于传统SIFT(Scale-Invarivant Feature Transform)算法对复杂环境中工具图像匹配的准确率低、速度慢,提出了一种基于边缘信息的SIFT图像匹配算法。采用自适应平滑滤波和改进Otsu算法对Canny算子进行改进,获取工具图像的边缘信息,然后采用形态学运算对边缘进行膨胀和闭运算,去除边缘的鸿沟,最后使用SIFT算法提取稳定特征点实现工具图像匹配。实验结果表明,改进的Canny算子能消除图像中伪边缘对边缘检测的影响,具备较强的自适应性;对于存在尺度变化、缩放、旋转和部分遮挡等情况下的工具图像,改进的算法在匹配准确率和匹配时间等方面优于传统SIFT算法。In view of the low accuracy and slowness of traditional SIFT（Scale-Invarivant Feature Transform）algorithm of tool image matching in complex environment,a SIFT image matching algorithm based on edge information is proposed. The Canny operator is improved by adaptive smoothing and improved Otsu algorithm.The edge information of the tool image is obtained. The edge is expanded and closed by morphological operation,and the edge of the edge is removed. The SIFT algorithm is used to extract the stable feature points to achieve the matching of the tool image. The experimental results show that the improved Canny operator can eliminate the influence of the pseudo edge on the edge detection and have strong adaptability. For the tool image with scale change, scaling, rotation and partial occlusion, the improved algorithm is superior to traditional SIFT algorithm in terms of matching accuracy and matching time. Which is better than the traditional SIFT algorithm.

关 键 词：CANNY算子 形态学 OTSU算法 SIFT算法 图像匹配 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]