基于引导滤波和积分投影算法的轨道扣件定位
|
摘要
精确定位是实现轨道扣件缺陷计算机自动检测的基础,为此提出了一种改进引导滤波去噪和灰度积分投影结合模板匹配算法的轨道扣件定位方法.首先,通过具有良好边缘保持能力的改进引导滤波算法对轨道扣件图像进行去噪;其次,利用改进Canny算法在Opencv平台对扣件图像进行边缘检测,实现轨道扣件图像边缘检测的自适应性;再次,采用灰度积分投影算法结合先验知识粗定位扣件区域;最后,通过模板匹配算法精确定位轨道扣件.仿真实验表明:所采用的算法具有较好的定位能力,可以准确地定位轨道扣件区域,为进一步的扣件识别提供了可靠的基础.
Precise positioning is the basis for automatic computer detection of track fastener defects.This study proposed a method of track fastener positioningbased on improved guided filter de-noising and gray integral projection combined with template matching algorithm.Firstly,the track fastener images were de-noised by an improved guided filtering algorithm with good edge retention;Then,using the advanced Canny algorithm on the Opencv platform to quickly achieved the adaptive edge detection of the track fastener image;Thirdly,gray integral projection algorithm combined with prior knowledge is used to locate the coarse fastener area;Finally,the track fastener were precisely located by the template matching algorithm.This simulation results show that the proposed algorithm has good positioning ability and can accurately locate the track fastener region,which provides a reliable basis for further fastener identification.
Precise positioning is the basis for automatic computer detection of track fastener defects.This study proposed a method of track fastener positioningbased on improved guided filter de-noising and gray integral projection combined with template matching algorithm.Firstly,the track fastener images were de-noised by an improved guided filtering algorithm with good edge retention;Then,using the advanced Canny algorithm on the Opencv platform to quickly achieved the adaptive edge detection of the track fastener image;Thirdly,gray integral projection algorithm combined with prior knowledge is used to locate the coarse fastener area;Finally,the track fastener were precisely located by the template matching algorithm.This simulation results show that the proposed algorithm has good positioning ability and can accurately locate the track fastener region,which provides a reliable basis for further fastener identification.
引文
[1] 翁长根,郝丹,王阳,等.扣件失效对城市轨道交通列车-浮置板式轨道系统竖向振动响应的影响[J].铁道科学与工程学报,2008,5(2):29-33.
[2] 刘馨,穆颖,张斌.基于计算机视觉的钢轨扣件检测算法研究[J].华东交通大学学报,2017,34(2):72-77.
[3] MARINO F,DISTANTE A,MAZZEO P L,et al.A real-time visual inspection system for railway maintenance:automatic hexagonal-head bolts detection[J].IEEE Transactions on System,Man and Cybernetics,Part C:Applications and Reviews,2007,37(3):418-428.
[4] 钱广春.基于计算机视觉的铁路扣件缺失快速探测方法研究[D].上海:上海交通大学,2011:1-2.
[5] 刘甲甲,熊鹰,李柏林,等.基于计算机视觉的轨道扣件缺陷自动检测算法研究[J].铁道学报,2016,38(8):73-80.
[6] 马姗.铁路扣件识别的研究[D].北京:北京交通大学,2012:8-12.
[7] XIA Y,XIE F,JIANG Z.Broken railway fastener detection based on adaboost algorithm[C]//International Conference optoelectronics and Image Procession.Washington,DC:IEEE Press,2011:313-316.
[8] 张丛.机器视觉轨道缺陷检测成像系统的研究[D].南昌:南昌大学,2017:20-25.
[9] HE K M,SUN J,TANG X O.Guided image filter[C]//European Conference on Computer Vision.Washington,DC:IEEE Computer Society,2010:1-14.
[10] 蒋卓承.采用自适应高斯引导滤波对条纹图像去噪和三维重建[D].深圳:深圳大学,2015:27-31.
[11] 钱小燕.引导滤波的红外图像预处理算法[J].科学技术与工程,2015,15(21):32-37.
[12] 芦碧波,王建龙,郑艳梅.全变分引导图像去噪[J].计算机工程与应用,2016,52(8):207-210.
[13] 曾俊.图像边缘检测技术及其应用研究[D].武汉:华中科技大学,2011:17-19.
[14] 邓动伟.图像边缘检测算法的改进及其应用研究[D].长沙:湖南师范大学,2016:29-39.
[15] 冯建强,刘文波,于盛林.基于灰度积分投影的人眼定位[J].计算机仿真,2005,4(22):75-77.
[16] MOHAMMED G J,HONG B R,JARJES A A.Accurate pupil features extraction based on new projection function[J].Computing and Informatics,2010,4(29):663-680.
[17] 吴福培,张宪民,邝泳聪,等.无铅焊点鲁棒定位的灰度积分投影算法[J].华南理工大学学报(自然科学版),2009,37(9):98-103.
[18] 王凯,李柏林.基于视觉追踪的轨道扣件状态实时自动检测算法研究[C]//第十一届全国博士生学术年会:信息技术与安全专题.成都:中国通信学会,2013:449-458.
[19] 张伟.基于机器视觉的驾驶人疲劳状态识别关键问题研究[D].北京:清华大学,2011:104-107.
[20] 吴禄慎,万超,陈华伟,等.一种改进的十字交叉轨道扣件定位方法[J].铁道标准设计,2016,60(12):49-53.
[21] 刘易鑫.铁路扣件状态的识别研究[D].石家庄:河北科技大学,2018:22-31.
[2] 刘馨,穆颖,张斌.基于计算机视觉的钢轨扣件检测算法研究[J].华东交通大学学报,2017,34(2):72-77.
[3] MARINO F,DISTANTE A,MAZZEO P L,et al.A real-time visual inspection system for railway maintenance:automatic hexagonal-head bolts detection[J].IEEE Transactions on System,Man and Cybernetics,Part C:Applications and Reviews,2007,37(3):418-428.
[4] 钱广春.基于计算机视觉的铁路扣件缺失快速探测方法研究[D].上海:上海交通大学,2011:1-2.
[5] 刘甲甲,熊鹰,李柏林,等.基于计算机视觉的轨道扣件缺陷自动检测算法研究[J].铁道学报,2016,38(8):73-80.
[6] 马姗.铁路扣件识别的研究[D].北京:北京交通大学,2012:8-12.
[7] XIA Y,XIE F,JIANG Z.Broken railway fastener detection based on adaboost algorithm[C]//International Conference optoelectronics and Image Procession.Washington,DC:IEEE Press,2011:313-316.
[8] 张丛.机器视觉轨道缺陷检测成像系统的研究[D].南昌:南昌大学,2017:20-25.
[9] HE K M,SUN J,TANG X O.Guided image filter[C]//European Conference on Computer Vision.Washington,DC:IEEE Computer Society,2010:1-14.
[10] 蒋卓承.采用自适应高斯引导滤波对条纹图像去噪和三维重建[D].深圳:深圳大学,2015:27-31.
[11] 钱小燕.引导滤波的红外图像预处理算法[J].科学技术与工程,2015,15(21):32-37.
[12] 芦碧波,王建龙,郑艳梅.全变分引导图像去噪[J].计算机工程与应用,2016,52(8):207-210.
[13] 曾俊.图像边缘检测技术及其应用研究[D].武汉:华中科技大学,2011:17-19.
[14] 邓动伟.图像边缘检测算法的改进及其应用研究[D].长沙:湖南师范大学,2016:29-39.
[15] 冯建强,刘文波,于盛林.基于灰度积分投影的人眼定位[J].计算机仿真,2005,4(22):75-77.
[16] MOHAMMED G J,HONG B R,JARJES A A.Accurate pupil features extraction based on new projection function[J].Computing and Informatics,2010,4(29):663-680.
[17] 吴福培,张宪民,邝泳聪,等.无铅焊点鲁棒定位的灰度积分投影算法[J].华南理工大学学报(自然科学版),2009,37(9):98-103.
[18] 王凯,李柏林.基于视觉追踪的轨道扣件状态实时自动检测算法研究[C]//第十一届全国博士生学术年会:信息技术与安全专题.成都:中国通信学会,2013:449-458.
[19] 张伟.基于机器视觉的驾驶人疲劳状态识别关键问题研究[D].北京:清华大学,2011:104-107.
[20] 吴禄慎,万超,陈华伟,等.一种改进的十字交叉轨道扣件定位方法[J].铁道标准设计,2016,60(12):49-53.
[21] 刘易鑫.铁路扣件状态的识别研究[D].石家庄:河北科技大学,2018:22-31.