摘要
提出了一种激光与视觉复合的火炮内膛膛线检测方法。该方法采用激光位移传感器对火炮膛线直径进行定量检测,获取火炮膛线截面位移数据后通过RANSAC算法排除火炮阴阳线交错处与内膛表面弊病所产生的异点,对得到的阴线、阳线数据进行拟合与优化,计算出膛线直径。采用HOG特征结合SVM分类器对膛内锈斑等缺陷进行检测识别,完成整个膛线检测;完成了膛内爬行机器人样机制作,搭建了膛线检测系统,膛线直径和缺陷检测结果验证了该算法的有效性。
Aiming at the problems of poor rifling detection environment and high detection accuracy,a laser and vision composite rifling detection method was proposed.This method used a laser displacement sensor to quantitatively detect the diameter of the artillery rifle,obtained the displacement data of the rifle section of the artillery,and eliminated the abnormal points caused by the intersection of groove and land lines of the artillery and the defects of the inner bore surface through the RANSAC algorithm.In order to detect defects such as rust spots in the bore,the HOG feature combined with the SVM classifier was used to detect and identify them,and the entire rifle detection was completed by fitting and optimization,calculating the diameter of the rifling.The prototype production of the crawling robot was completed,and the rifling detection system was built.To verify the effectiveness of the method,the rifle diameter and defect detection were carried out,and the experimental results verified the effectiveness of the algorithm.
作者
王磊
常精丰
张嘉易
郝永平
WANG Lei;CHANG Jingfeng;ZHANG Jiayi;HAO Yongping(Weapon Science and Technology Research Center,Shenyang Ligong University,Shenyang 110159,China)
出处
《兵器装备工程学报》
CSCD
北大核心
2021年第7期222-227,共6页
Journal of Ordnance Equipment Engineering
基金
沈阳市中青年科技创新人才支持计划项目(RC200537)。
