基于机器视觉的受电弓滑板磨耗检测系统研究与实现
Research and Implementation of a Machine Vision Based Wear Detection System for Pantograph Skateboards
曾光 1黄杨灵 1佟景泉 1武文星 1黄健盛1
作者信息
- 1. 广东交通职业技术学院,广东广州 510630
- 折叠
摘要
为满足城轨列车受电弓滑板磨耗病害的自动化检测需求,提高受电弓滑板磨耗病害检测效率,设计了一种基于机器视觉的受电弓滑板磨耗检测系统,应用改进型Canny边缘检测算法,获取受电弓滑板边缘图像,通过霍夫变换等技术,提取与拟合滑板磨耗边缘,获取滑板的磨耗信息,然后分析判断磨耗数据,最终实现对弓网磨耗情况的非接触式精准自动检测.结果表明,在正向平行拍摄角度下,系统测量结果与人工测量结果高度吻合,准确率达到91.6%,可满足实际检修需求.
Abstract
In order to meet the automation detection requirements for the wear and tear of pantograph slide plates on urban rail trains and improve the detection efficiency of pantograph slide plate wear and tear,a machine vision based wear and tear detection system for pantograph slide plates is designed.The improved Canny edge detection algorithm is applied to obtain the edge image of the pantograph slide plate.Through techniques such as Hough transform,the wear and tear edge of the slide plate is extracted and fitted,and the wear information of the slide plate is obtained.Then,the wear data is analyzed and judged,and finally,non-contact accurate automatic detection of pantograph mesh wear is achieved.The results show that under the forward parallel shooting angle,the system measurement results are highly consistent with the manual measurement results,with an accuracy rate of 91.6%,which can meet the actual maintenance needs.
关键词
机器视觉/弓网/磨耗/边缘检测/图像处理Key words
machine vision/bow net/wear/edge detection/image processing引用本文复制引用
基金项目
广东省普通高校特色创新项目(2021KTSCX223)
广东省普通高校青年创新人才类项目(2021KQNCX176)
2021年广东省科技创新战略专项资金项目(pdjh2022b0854)
2023年广东省科技创新战略专项资金项目(pdjh202360852)
出版年
2024
NETL
NSTL
万方数据