架空输电通道图像监测中大场景双目测距的方法及校正算法
|
摘要
针对双目测距技术无法适应大场景架空线路监测通道中应用需求的问题,提出一种双目测距的标定方法和数值校正算法。通过用标定杆代替传统的黑白棋格标定板对双目相机进行标定,并对标定结果进行比较;通过缩放图像对焦距的标定结果进行改进;通过拟牛顿法对旋转角度寻优,以地面为基准旋转重建的世界坐标系。现场实测证明:提出的标定方法解决了传统的双目标定方法在大场景中无法标定的问题;焦距校正算法解决了在变焦相机在双目算法中的不适应性,进一步提高了测量精度;角度校正算法实现了直接高度的测量,解决了由于架空线路通道中现场地面环境复杂,物体底部被遮挡,而无法进行高度测量的问题。
In view of the fact that binocular distance measurement technology could not meet the application requirements of overhead lines monitoring in large scenes, we proposed a calibration method and a regulation algorithm in binocular stereo vision. Binocular cameras were calibrated with a calibration pole instead of the traditional checkerboard, and the calibration results were compared. Calibration results of the focal length were improved through zooming images. Reconstructed world coordinate system was rotated with the ground being reference, and angles of the rotation were optimized by the quasi Newton method. The field experiment results show that using the proposed calibration method can solve the problem that the conventional calibration method cannot be utilized in large scene. Regulation algorithm of focal lengths can solve the problem of inadaptability of zoom cameras for binocular stereo vision and further improve the accuracy of measurement. Angles regulation algorithm can realize direct height measurement and solve the problem that the height cannot be measured because the bottom of obstacles is veiled due to the complex environment in transmission line corridor.
In view of the fact that binocular distance measurement technology could not meet the application requirements of overhead lines monitoring in large scenes, we proposed a calibration method and a regulation algorithm in binocular stereo vision. Binocular cameras were calibrated with a calibration pole instead of the traditional checkerboard, and the calibration results were compared. Calibration results of the focal length were improved through zooming images. Reconstructed world coordinate system was rotated with the ground being reference, and angles of the rotation were optimized by the quasi Newton method. The field experiment results show that using the proposed calibration method can solve the problem that the conventional calibration method cannot be utilized in large scene. Regulation algorithm of focal lengths can solve the problem of inadaptability of zoom cameras for binocular stereo vision and further improve the accuracy of measurement. Angles regulation algorithm can realize direct height measurement and solve the problem that the height cannot be measured because the bottom of obstacles is veiled due to the complex environment in transmission line corridor.
引文
[1]胡毅.电线路运行故障的分析与防治[J].高电压技术,2007,33(3):1-8.HU Yi.Analysis on operation faults of transmission line and countermeasures[J].High Voltage Engineering,2007,33(3):1-8.
[2]胡毅.影响送电网安全运行的有关问题及对策[J].高电压技术,2005,31(4):77-79.HU Yi.Analysis and research of the faults occurred on transmission line[J].High Voltage Engineering,2005,31(4):77-79.
[3]孙凤杰,赵孟丹,刘威,等.输电线路在线监测技术研究[J].南方电网技术,2012,36(4):17-22.SUN Fengjie,ZHAO Mengdan,LIU Wei,et al.Study on the on-line monitoring technology of overhead power transmission lines[J].Southern Power System Technology,2012,36(4):17-22.
[4]冈萨雷斯,伍兹.数字图像处理[M].3版.阮秋琦,阮宇智,译.北京:电子工业出版社,2011.GONZALEZ R C,WOODS R E.Digital image processing[M].3rd ed.RUAN Qiuqi,RUAN Yuzhi,translated.Beijing,China:Publishing House of Electronics Industry,2011.
[5]KATRASNIK J,PERNUS F,LIKAR B.A survey of mobile robots for distribution power line inspection[J].IEEE Transactions on Power Delivery,2009,25(1):485-493.
[6]GOLIGHTLY I,JONES D.Corner detection and matching for visual tracking during power line inspection[J].Image&Vision Computing,2003,21(9):827-840.
[7]PEUNGSUNGWAL S,PUNGSIRI B,CHAMNONGTHAI K,et al.Autonomous robot for a power transmission line inspection[C]∥IEEEInternational Symposium on Circuits and Systems.Sydney,Australia:IEEE,2001:121-124.
[8]詹朝铖,李正波.基于Adaboost的架空输电线路巡线机器人障碍识别[J].计算机与数字工程,2009,37(11):130-133.ZHAN Chaocheng,LI Zhengbo.Obstacle recognition for transmission line inspection robot based on Adaboost[J].Computer&Digital Engineering,2009,37(11):130-133.
[9]郝艳捧,刘国特,薛艺为,等.输电线路覆冰厚度的小波分析图像识别[J].高电压技术,2014,40(2):368-373.HAO Yanpeng,LIU Guote,XUE Yiwei,et al.Wavelet image recognition of ice thickness on transmission lines[J].High Voltage Engineering,2014,40(2):368-373.
[10]胡伟,吴发献,刘家齐,等.基于统计学习的电力设施现场危险目标识别方法[J].华东电力,2012,40(10):1815-1819.HU Wei,WU Faxian,LIU Jiaqi,et al.On-site threatening object recognition method for power equipment based on statistical learning[J].East China Electric Power,2012,40(10):1815-1819.
[11]杨浩,吴畏.基于三维重建的输电线路覆冰在线监测方法[J].电力系统自动化,2012,36(23):103-108.YANG Hao,WU Wei.On-line monitoring method of icing transmission lines based on 3D reconstruction[J].Automation of Electric Power Systems,2012,36(23):103-108.
[12]金立军,姚春羽,闫书佳,等.基于航拍图像的输电线路异物识别[J].同济大学学报(自然科学版),2013,41(2):277-281.JIN Lijun,YAO Chunyu,YAN Shujia,et al.Recognition of extra matters on transmission lines based on aerial images[J].Journal of Tongji University(Natural Science),2013,41(2):277-281.
[13]陈斯雅,王滨海,盛戈皞,等.采用图像摄影的输电线路弧垂测量方法[J].高电压技术,2011,37(4):904-909.CHEN Siya,WANG Binhai,SHENG Gehao,et al.Application of digital image processing and photogrammetric Technology to sag measuring method[J].High Voltage Engineering,2011,37(4):904-909.
[14]朱炜,何冰,刘新平,等.输电线路监控系统中运动目标的智能检测与识别[J].华东电力,2011,39(10):1699-1704.ZHU Wei,HE Bing,LIU Xinping,et al.Intelligent measurement and identification technique for moving objects in transmission line monitoring system[J].East China Electric Power,2011,39(10):1699-1704.
[15]FAN L,YAO X,QI H,et al.An automatic control system for end robot based on binocular vision position[C]∥IEEE International Conference on Robotics and Biomimetics.Bangkok,Thailan:IEEE,2008:914-919.
[16]韩志文.基于立体视觉的地铁异物侵限分析[D].北京:北京交通大学,2012.HAN Zhiwen.Analysis on metro gauge intrusion based on stereovision[D].Beijing,China:Beijing Jiaotong University,2012.
[17]乐振春,刘文韬,赵文彬.基于图像识别的嵌入式架空线路监测系统[J].华东电力,2014,42(1):61-66.LE Zhenchun,LIU Wentao,ZHAO Wenbin.Embedded system based invader detection for high-voltage transmission line[J].East China Electric Power,2014,42(1):61-66.
[18]SZELISKI R.Computer vision:algorithms and applications[M].Bern,Switzerland:Springer Nature,2010.
[19]ZHANG Z.A flexible new technique for camera calibration[J].IEEETransactions on Pattern Analysis&Machine Intelligence,2000,22(11):1330-1334.
[20]马颂德,张正友.计算机视觉-计算理与算法基础[M].北京:科学出版社,1998.MA Songde,ZHANG Zhengyou.Computer vision theory and algorithm[M].Beijing,China:Science Press,1998.
[21]于洁.输电线路防外力破坏预警系统研究[D].北京:华北电力大学,2015.YU Jie.Research on the warning system of preventing from external damaging for transmission line[J].Beijing,China:North China Electric Power University,2015.
[22]DAVIDON W C.Variable metric method for minimization[J].Siam Journal on Optimization,2006,1(1):1-17.
[23]HUANG G S,CHENG C E.3D coordinate identification of object using binocular vision system for mobile robot[C]∥2013 CACS International Automatic Control Conference(CACS).Nantou,China:IEEE,2013:91-96.
[2]胡毅.影响送电网安全运行的有关问题及对策[J].高电压技术,2005,31(4):77-79.HU Yi.Analysis and research of the faults occurred on transmission line[J].High Voltage Engineering,2005,31(4):77-79.
[3]孙凤杰,赵孟丹,刘威,等.输电线路在线监测技术研究[J].南方电网技术,2012,36(4):17-22.SUN Fengjie,ZHAO Mengdan,LIU Wei,et al.Study on the on-line monitoring technology of overhead power transmission lines[J].Southern Power System Technology,2012,36(4):17-22.
[4]冈萨雷斯,伍兹.数字图像处理[M].3版.阮秋琦,阮宇智,译.北京:电子工业出版社,2011.GONZALEZ R C,WOODS R E.Digital image processing[M].3rd ed.RUAN Qiuqi,RUAN Yuzhi,translated.Beijing,China:Publishing House of Electronics Industry,2011.
[5]KATRASNIK J,PERNUS F,LIKAR B.A survey of mobile robots for distribution power line inspection[J].IEEE Transactions on Power Delivery,2009,25(1):485-493.
[6]GOLIGHTLY I,JONES D.Corner detection and matching for visual tracking during power line inspection[J].Image&Vision Computing,2003,21(9):827-840.
[7]PEUNGSUNGWAL S,PUNGSIRI B,CHAMNONGTHAI K,et al.Autonomous robot for a power transmission line inspection[C]∥IEEEInternational Symposium on Circuits and Systems.Sydney,Australia:IEEE,2001:121-124.
[8]詹朝铖,李正波.基于Adaboost的架空输电线路巡线机器人障碍识别[J].计算机与数字工程,2009,37(11):130-133.ZHAN Chaocheng,LI Zhengbo.Obstacle recognition for transmission line inspection robot based on Adaboost[J].Computer&Digital Engineering,2009,37(11):130-133.
[9]郝艳捧,刘国特,薛艺为,等.输电线路覆冰厚度的小波分析图像识别[J].高电压技术,2014,40(2):368-373.HAO Yanpeng,LIU Guote,XUE Yiwei,et al.Wavelet image recognition of ice thickness on transmission lines[J].High Voltage Engineering,2014,40(2):368-373.
[10]胡伟,吴发献,刘家齐,等.基于统计学习的电力设施现场危险目标识别方法[J].华东电力,2012,40(10):1815-1819.HU Wei,WU Faxian,LIU Jiaqi,et al.On-site threatening object recognition method for power equipment based on statistical learning[J].East China Electric Power,2012,40(10):1815-1819.
[11]杨浩,吴畏.基于三维重建的输电线路覆冰在线监测方法[J].电力系统自动化,2012,36(23):103-108.YANG Hao,WU Wei.On-line monitoring method of icing transmission lines based on 3D reconstruction[J].Automation of Electric Power Systems,2012,36(23):103-108.
[12]金立军,姚春羽,闫书佳,等.基于航拍图像的输电线路异物识别[J].同济大学学报(自然科学版),2013,41(2):277-281.JIN Lijun,YAO Chunyu,YAN Shujia,et al.Recognition of extra matters on transmission lines based on aerial images[J].Journal of Tongji University(Natural Science),2013,41(2):277-281.
[13]陈斯雅,王滨海,盛戈皞,等.采用图像摄影的输电线路弧垂测量方法[J].高电压技术,2011,37(4):904-909.CHEN Siya,WANG Binhai,SHENG Gehao,et al.Application of digital image processing and photogrammetric Technology to sag measuring method[J].High Voltage Engineering,2011,37(4):904-909.
[14]朱炜,何冰,刘新平,等.输电线路监控系统中运动目标的智能检测与识别[J].华东电力,2011,39(10):1699-1704.ZHU Wei,HE Bing,LIU Xinping,et al.Intelligent measurement and identification technique for moving objects in transmission line monitoring system[J].East China Electric Power,2011,39(10):1699-1704.
[15]FAN L,YAO X,QI H,et al.An automatic control system for end robot based on binocular vision position[C]∥IEEE International Conference on Robotics and Biomimetics.Bangkok,Thailan:IEEE,2008:914-919.
[16]韩志文.基于立体视觉的地铁异物侵限分析[D].北京:北京交通大学,2012.HAN Zhiwen.Analysis on metro gauge intrusion based on stereovision[D].Beijing,China:Beijing Jiaotong University,2012.
[17]乐振春,刘文韬,赵文彬.基于图像识别的嵌入式架空线路监测系统[J].华东电力,2014,42(1):61-66.LE Zhenchun,LIU Wentao,ZHAO Wenbin.Embedded system based invader detection for high-voltage transmission line[J].East China Electric Power,2014,42(1):61-66.
[18]SZELISKI R.Computer vision:algorithms and applications[M].Bern,Switzerland:Springer Nature,2010.
[19]ZHANG Z.A flexible new technique for camera calibration[J].IEEETransactions on Pattern Analysis&Machine Intelligence,2000,22(11):1330-1334.
[20]马颂德,张正友.计算机视觉-计算理与算法基础[M].北京:科学出版社,1998.MA Songde,ZHANG Zhengyou.Computer vision theory and algorithm[M].Beijing,China:Science Press,1998.
[21]于洁.输电线路防外力破坏预警系统研究[D].北京:华北电力大学,2015.YU Jie.Research on the warning system of preventing from external damaging for transmission line[J].Beijing,China:North China Electric Power University,2015.
[22]DAVIDON W C.Variable metric method for minimization[J].Siam Journal on Optimization,2006,1(1):1-17.
[23]HUANG G S,CHENG C E.3D coordinate identification of object using binocular vision system for mobile robot[C]∥2013 CACS International Automatic Control Conference(CACS).Nantou,China:IEEE,2013:91-96.