计算机技术与发展2017，Vol.27Issue(10)：201-204,4.DOI:10.3969/j.issn.1673-629X.2017.10.043
一种基于图像配准的钢绳芯抽动识别技术
A Technique of Recognition on Twitch of Steel Cord Based on Image Registration
摘要
Abstract
Regarding the characteristics of the X-ray image of the steel cord conveyor belt,a method of detecting the twitch of the steel cord based on image registration is proposed,in which the SURF feature points of the current image and reference image are extracted based on principles and general means of image registration and SURF feature,and then the error matched points are removed with ran-dom sample consensus algorithm ( RANSAC) . The image transforming parameters could be gained according to the matched points as the source data,and the reference image is in a geographical transform. The difference matrix between the reference image and current image is gained,and the twitching corresponding to the larger-value area in the difference matrix could be detected and the shifting situation be ascertained. The contrast experiment has been conducted for verification on its feasibility and effectiveness. Experimental results show that compared with the normal detection approach,it has more applicability and has decreased the additional errors and detected more effec-tively.关键词
X光图像/图像配准/SURF特征点/RANSAC算法/抽动检测Key words
X-ray image/image registration/SURF point/RANSAC algorithm/shift detection分类
信息技术与安全科学引用本文复制引用
沈海洋,陈瑞,焦良葆..一种基于图像配准的钢绳芯抽动识别技术[J].计算机技术与发展,2017,27(10):201-204,4.基金项目
江苏省自然科学基金(BK20141389) （BK20141389）
南京工程学院博士基金项目(ZKJ201305) （ZKJ201305）
