基于二代曲波变换的电气化铁路接触网连接装置旋转双耳不良状态检测
发布时间：2018-11-04 13:53
【摘要】：针对电气化铁路接触网连接装置中旋转双耳耳片状态检测问题,提出一种基于二代曲波变换的非接触式自动检测方法。为降低复杂背景对后期状态检测的干扰,利用快速鲁棒性特征SURF(Speeded-up Robust Features)匹配方法完成旋转双耳模板与待检测旋转双耳局部不变特征点的匹配,实现双耳快速初识别;然后应用Hough变换实现耳片固有特征识别及耳片精确定位;最后利用二代曲波变换的多方向性和各向异性特点分解耳片局部图像,通过曲波系数分布规律及统计特性判别耳片状态。实验表明,基于二代曲波变换的系统方法能在复杂图像中较准确地识别耳片断裂特征,为旋转双耳的状态检测提供一定参考。
[Abstract]:A non-contact automatic detection method based on the second generation Qu Bo transform is proposed to detect the state of rotating binaural ears in the catenary connection device of electrified railway. In order to reduce the interference of complex background to late state detection, the fast robust feature SURF (Speeded-up Robust Features) matching method) is used to match the rotating binaural template with the local invariant feature points of the rotating binaural to be detected. Then the Hough transform is used to realize the recognition of the inherent characteristics of the ear plate and the accurate location of the ear piece. Finally, the multidirectional and anisotropic characteristics of the second generation Qu Bo transform are used to decompose the local images of ear slices, and then the distribution law and statistical characteristics of Qu Bo coefficients are used to judge the state of ear slices. The experimental results show that the system method based on the second generation Qu Bo transform can accurately identify the broken characteristics of ear segments in complex images and provide a certain reference for the detection of rotating binaural states.
【作者单位】： 西安铁路职业技术学院电气工程系;西南交通大学电气工程学院;
【基金】：基金项目:西安铁路职业技术学院(XTZY15G05)
【分类号】：TP391.41;U226.8
本文编号：2310015
[Abstract]:A non-contact automatic detection method based on the second generation Qu Bo transform is proposed to detect the state of rotating binaural ears in the catenary connection device of electrified railway. In order to reduce the interference of complex background to late state detection, the fast robust feature SURF (Speeded-up Robust Features) matching method) is used to match the rotating binaural template with the local invariant feature points of the rotating binaural to be detected. Then the Hough transform is used to realize the recognition of the inherent characteristics of the ear plate and the accurate location of the ear piece. Finally, the multidirectional and anisotropic characteristics of the second generation Qu Bo transform are used to decompose the local images of ear slices, and then the distribution law and statistical characteristics of Qu Bo coefficients are used to judge the state of ear slices. The experimental results show that the system method based on the second generation Qu Bo transform can accurately identify the broken characteristics of ear segments in complex images and provide a certain reference for the detection of rotating binaural states.
【作者单位】： 西安铁路职业技术学院电气工程系;西南交通大学电气工程学院;
【基金】：基金项目:西安铁路职业技术学院(XTZY15G05)
【分类号】：TP391.41;U226.8
【相似文献】
相关期刊论文 前3条
1 王伟旭;刘志刚;栗敏;张桂南;;基于曲波变换的高铁绝缘子状态检测方法[J];电气化铁道;2013年06期
2 朱传武;刘斌;;内燃机车状态检测信息系统的开发与应用[J];铁道机车车辆工人;2011年11期
3 谢拴勤;郭振安;;基于热棒工作状态检测问题的分析[J];计算机测量与控制;2008年07期
相关硕士学位论文 前2条
1 惠美玲;基于四元数小波变换的结构状态检测研究[D];大连理工大学;2016年
2 田安刚;列车司控运停状态检测分析系统研究[D];青岛科技大学;2012年
，本文编号：2310015
本文链接：https://www.wllwen.com/kejilunwen/ruanjiangongchenglunwen/2310015.html
