摘要
为提高铁鞋与车轮之间相对位移的检测精度,提出了一种基于改进DOG边缘检测算子的铁鞋与车轮之间相对位移的检测方法。该方法将相机安装在车轮旁侧的油箱上,首先对铁鞋图像进行预处理,其次采用改进DOG边缘检测算子获得铁鞋轮廓,对铁鞋轮廓进行特征点提取得到铁鞋尾部特征点,最后结合相机标定得到铁鞋与车轮的相对位移。实验结果表明,该方法实现了列车铁鞋与车轮之间相对位移的非接触检测,且误差在1%以内。改变了目前人工现场检查防溜铁鞋工作状态的传统方法,减轻了人工作业强度,该方法的提出为后续进一步开展列车运行安全状态监测与评价研究奠定了基础。
To improve the detection accuracy of relative displacement between skate and wheel,a detection method of relative displacement between iron shoe and wheel based on improved DOG edge detection operator is proposed.In this method,the camera is installed on the oil tank beside the wheel.Firstly,the image of the iron shoe is preprocessed.Secondly,the improved DOG edge detection operator is used to obtain the outline of the iron shoe.Secondly,the feature points of the tail of the iron shoe are extracted from the outline of the iron shoe.Finally,the relative displacement of the iron shoe and the wheel is obtained by combining the camera calibration.The experimental results show that the method realizes the non-contact detection of the relative displacement between the train shoe and wheel,and the error is within 1%.It has changed the traditional method of checking the working condition of anti slide shoes on the spot and reduced the intensity of manual work.This method has laid a foundation for further research on monitoring and evaluation of train operation safety condition.
作者
钱荣威
马增强
许丹丹
周涵
QIAN Rong-wei;MA Zeng-qiang;XU Dan-dan;ZHOU Han(School of Electrical and Electronic Engineering,Shijiazhuang Tiedao University,Shijiazhuang 050043,China;State Key Laboratory of Mechanical Behavior and System Safety of Traffic Engineering Structures,Shijiazhuang Tiedao University,Shijiazhuang 050043,China)
出处
《组合机床与自动化加工技术》
北大核心
2021年第3期94-97,共4页
Modular Machine Tool & Automatic Manufacturing Technique
基金
国家自然科学基金项目(12072207)
全国大学生创新创业项目(201912668009)
广东省教育厅特色创新项目(2019KTSCX240)
广东省创新创业训练项目(S202012668014)
全国大学生创新创业项目(201910107019)。
关键词
车载相机
DOG算子
图像处理
边缘检测
相机标定
vehicle camera
DOG algorithm
image processing
edge detection
camera calibration