基于曲率滤波和改进GMM的钢轨缺陷自动视觉检测方法
|
摘要
针对传统钢轨检测技术的效率低下、精度不足、安全隐患等问题,提出了基于曲率滤波和改进高斯混合模型(GMM)的钢轨表面缺陷检测方法。首先,提出了基于垂直投影的区域定位算法和灰度对比算法,克服现场工况复杂、轨面反射不均、信道噪声干扰的难题;考虑到图像信号受强工况噪声干扰,研究了具有隐式计算和曲面保持特性的曲率滤波法进行图像去噪;建立了基于马尔科夫随机场(MRF)的高斯混合模型完成表面缺陷的精确快速分割。最后,设计了"区域定位-灰度均衡化-滤波-分割"的实验流程,实验结果验证了算法的有效性,检测性能达到了精确度92.0%,相比其他方法更加精确、快速,具有更好的鲁棒性。
This paper proposes a visual detection method for rail surface defect based on curvature filtering and improved Gaussian mixture model( GMM) aiming atthe problems of low efficiency,lack of precision and safety hazard for traditional rail inspection technique. First of all,this paper proposes a ROI location algorithm based on vertical projection and gray contrast algorithm,which overcome the difficulties of.complex field condition and rail surface reflectance inequality and signal channel noise interference,Considering the situation that the image signal is interfered by strong noise,a curvature filteringmethod with implicit computing and surface preserving power is studied to conductthe image denoising and keep the image details. Next,an improved Gaussian mixture model based on Markov random field( MRF) is established to achievethe accurate and rapid surface defect segmentation. In the end,the experiment process of region location-gray equalization-filtersegmentation was designed. The experiment results verify the effectiveness of theproposed method,the detection accuracy of 92. 0% is reached,and the proposed method is more accurate,faster and more robust than other methods.
This paper proposes a visual detection method for rail surface defect based on curvature filtering and improved Gaussian mixture model( GMM) aiming atthe problems of low efficiency,lack of precision and safety hazard for traditional rail inspection technique. First of all,this paper proposes a ROI location algorithm based on vertical projection and gray contrast algorithm,which overcome the difficulties of.complex field condition and rail surface reflectance inequality and signal channel noise interference,Considering the situation that the image signal is interfered by strong noise,a curvature filteringmethod with implicit computing and surface preserving power is studied to conductthe image denoising and keep the image details. Next,an improved Gaussian mixture model based on Markov random field( MRF) is established to achievethe accurate and rapid surface defect segmentation. In the end,the experiment process of region location-gray equalization-filtersegmentation was designed. The experiment results verify the effectiveness of theproposed method,the detection accuracy of 92. 0% is reached,and the proposed method is more accurate,faster and more robust than other methods.
引文
[1]田贵云,高斌,高运来,等.铁路钢轨缺陷伤损巡检与监测技术综述[J].仪器仪表学报,2016,37(8):1763-1780.TIAN G Y,GAO B,GAO Y L,et al.Review of railway rail defect non-destructive testing and monitoring[J].Chinese Journal of Scientific Instrument,2016,37(8):1763-1780.
[2]周宇,张杰,王少锋,等.考虑磨耗的钢轨疲劳裂纹萌生寿命预测仿真[J].铁道学报,2016,38(7):91-97.ZHOU Y,ZHANG J,WANG SH F,et al.Simulation on rail head crack initiation life prediction considering rail wear[J].Journal of the China Railway Society,2016,38(7):91-97.
[3]郭火明,王文健,刘腾飞,等.重载铁路钢轨损伤行为分析[J].中国机械工程,2014,25(2):267-272.GUO H M,WANG W J,LIU T F,et al.Analysis of damage behavior of heavy-haul railway rails[J].China Mechanical Engineering,2014,25(2):267-272.
[4]卢超,魏运飞,徐薇.钢轨踏面斜裂纹超声表面波B扫成像检测研究[J].仪器仪表学报,2010,31(10):2272-2278.LU CH,WEI Y F,XU W.Study on B-scan imaging detection for rail tread tilted cracks using ultrasonic surface wave[J].Chinese Journal of Scientific Instrument,2010,31(10):2272-2278.
[5]王典洪,甘胜丰,张伟民,等.基于监督双限制连接Isomap算法的带钢表面缺陷图像分类方法[J].自动化学报,2014,40(5):883-891.WANG D H,GAN SH F,ZHANG W M,et al.Strip surface defect image classication based on double-limited and supervise-connect isomap algorithm[J].Acta Automatica Sinica,2014,40(5):883-891.
[6]余建波,卢笑蕾,宗卫周.基于局部与非局部线性判别分析和高斯混合模型动态集成的晶圆表面缺陷探测与识别[J].自动化学报,2016,42(1):47-59.YU J B,LU X L,ZONG W ZH.Wafer defect detection and recognition based on local and nonlocal linear discriminant analysis and dynamic ensemble of gaussian mixture models[J].Acta Automatica Sinica,2016,42(1):47-59.
[7]孙作雷,茅旭初.基于激光束匹配和图模型的移动机器人相对运动估计误差分析[J].自动化学报,2011,37(2):205-213.SUN Z L,MAO X CH.Quantification of relative movement uncertainty based on laser scanmatching and graphical model for mobile robot[J].Acta Automatica Sinica,2011,37(2):205-213.
[8]周正干,彭地,李洋,等.相控阵超声检测技术中的全聚焦成像算法及其校准研究[J].机械工程学报,2015,51(10):1-7.ZHOU ZH G,PENG D,LI Y,et al.Research on phased array ultrasonic total focusing method and its calibration[J].Journal of Mechanical Engineering,2015,51(10):1-7.
[9]周德强,潘萌,常祥,等.铁磁性构件缺陷的脉冲涡流检测模式研究[J].仪器仪表学报,2017,38(6):1498-1505.ZHOU D Q,PAN M,CHANG X,et al.Research on detection modes of ferromagnetic component defects using pulsed eddy current[J].Chinese Journal of Scientific Instrument,2017,38(6):1498-1505.
[10]张成伟,崔畅.基于脉冲电流放电法的蓄电池内阻在线检测研究[J].国外电子测量技术,2017,36(11):11-14.ZHANG CH W,CUI CH.Research on battery internal resistance online detection based on pulse current discharge method[J].Foreign Electronic Measurement Technology,2017,36(11):11-14.
[11]丁建睿,黄剑华,刘家锋,等.局部特征与多示例学习结合的超声图像分类方法[J].自动化学报,2013,39(6):861-867.DING J R,HUANG J H,LIU J F,et al.Combining local features and multi-instance learning for ultrasound image classification[J].Acta Automatica Sinica,2013,39(6):861-867.
[12]陈振华,卢超,陆铭慧,等.基于声-超声检测的薄钢板多焊点结构完整性评价技术[J].机械工程学报,2013,49(16):57-61.CHEN ZH H,LU CH,LU M H,et al.Integrity evaluation on spot welded construction of thin steel sheet based on acousto-ultrasonic technique[J].Journal of Mechanical Engineering,2013,49(16):57-61.
[13]LI Q,REN S.A real-time visual inspection system for discrete surface defects of rail heads[J].IEEE Transactions on Instrumentation and Measurement,2012,61(8):2189-2199.
[14]LI Q,REN S.A visual detection system for rail surface defects[J].IEEE Transactions on Systems,Man,and Cybernetics,Part C(Applications and Reviews),2012,42(6):1531-1542.
[15]SALVADOR P,NARANJO V,INSA R,et al.Axlebox accelerations:Their acquisition and time-frequency characterisation for railway track monitoring purposes[J].Measurement,2016(82):301-312.
[16]RESENDIZ E,HART J M,AHUJA N.Automated visual inspection of railroad tracks[J].IEEE Transactions on Intelligent Transportation Systems,2013,14(2):751-760.
[17]SUN M,LIN X,WU Z,et al.Non-destructive photoacoustic detecting method for high-speed rail surface defects[C].Instrumentation and Measurement Technology Conference(I2MTC)Proceedings,IEEE,2014:896-900.
[18]GRASSIE S L.Squats and squat-type defects in rails:the understanding to date[C].Proceedings of the Institution of Mechanical Engineers,Part F:Journal of Rail and Rapid Transit,2012,226(3):235-242.
[19]THOMAS H M,HECKEL T,HANSPACH G.Advantage of a combined ultrasonic and eddy current examination for railway inspection trains[J].Insight-Non-Destructive Testing and Condition Monitoring,2007,49(6):341-344.
[20]ALIPPI C,CASAGRANDE E,SCOTTI F,et al.Composite real-time image processing for railways track profile measurement[J].IEEE Transactions on Instrumentation and Measurement,2000,49(3):559-564.
[21]LIANG B,IWNICKI S,BALL A,et al.Adaptive noise cancelling and time-frequency techniques for rail surface defect detection[J].Mechanical Systems and Signal Processing,2015(54):41-51.
[22]MANDRIOTA C,NITTI M,ANCONA N,et al.Filterbased feature selection for rail defect detection[J].Machine Vision and Applications,2004,15(4):179-185.
[23]WANG L,HANG Y,LUO S,et al.Deblurring Gaussian-blur images:A preprocessing for rail head surface defect detection[C].IEEE International Conference on Service Operations,Logistics,and Informatics(SOLI),2011:451-456.
[24]DUBEY A K,JAFFERY Z A.Maximally stable extremal region marking-based railway track surface defect sensing[J].IEEE Sensors Journal,2016,16(24):9047-9052.
[25]HE Z,WANG Y,YIN F,et al.Surface defect detection for high-speed rails using an inverse PM diffusionmodel[J].Sensor Review,2016,36(1):86-97.
[26]贺振东,王耀南,刘洁,等.基于背景差分的高铁钢轨表面缺陷图像分割[J].仪器仪表学报,2016,37(3):640-649.HE ZH D,WANG Y N,LIU J,et al.Background differencing-based high-speed rail surface defect image segmentation[J].Chinese Journal of Scientific Instrument,2016,37(3):640-649.
[27]贺振东,王耀南,毛建旭,等.基于反向PM扩散的钢轨表面缺陷视觉检测[J].自动化学报,2014,40(8):1667-1679.HE ZH D,WANG Y N,MAO J X,et al.Research on inverse P-M diffusion-based rail surface defect detection[J].Acta Automatica Sinica,2014,40(8):1667-1679.
[28]TA爦TIMUR C,KARAKSE M,AKIN E,et al.Rail defect detection with real time image processing technique[C].14th International Conference on Industrial Informatics(INDIN),IEEE,2016:411-415.
[29]HAJIZADEH S,NNEZ A,TAX D M J.Semisupervised rail defect detection from imbalanced image data[J].IFAC-Papers on Line,2016,49(3):78-83.
[30]FAGHIH-ROOHI S,HAJIZADEH S,NEZ A,et al.Deep convolutional neural networks for detection of rail surface defects[C].International Joint Conference on Neural Networks(IJCNN),IEEE,2016:2584-2589.
[31]GIBERT X,PATEL V M,CHELLAPPA R.Deep multitask learning for railway track inspection[J].IEEE Transactions on Intelligent Transportation Systems,2017,18(1):153-164.
[32]黄金,周先春,吴婷,等.混合维纳滤波与改进型TV的图像去噪模型[J].电子测量与仪器学报,2017,31(10):1659-1666.HUANG J,ZHOU X CH,WU T,et al.Image denoising model based on mixing Wiener filtering and improved total variation[J].Journal of Electronic Measurement and Instrumentation,2017,31(10):1659-1666.
[33]GONG Y,SBALZARINI I F.Curvature filters efficiently reduce certain variational energies[J].IEEE Transactions on Image Processing,2017,26(4):1786-1798.
[34]张辉,金侠挺.基于曲率滤波和反向PM电动车充电孔检测方法[J].仪器仪表学报,2016,37(7):1626-1638.ZHANG H,JIN X T.Detection method for electric vehicle charging hole based on curvature filter and inverse P-M diffusion[J].Chinese Journal of Scientific Instrument,2016,37(7):1626-1638.
[35]SANJAY-GOPAL S,HEBERT T J.Bayesian pixel classification using spatially variant finite mixtures and the generalized EM algorithm[J].IEEE Transactions on Image Processing,1998,7(7):1014-1028.
[36]NGUYEN T M,WU Q M J.Fast and robust spatially constrained gaussian mixture model for image segmentation[J].IEEE transactions on circuits and systems for video technology,2013,23(4):621-635.
[37]NIKOU C,GALATSANOS N P,LIKAS A C.A classadaptive spatially variant mixture model for image segmentation[J].IEEE Transactions on Image Processing,2007,16(4):1121-1130.
[38]乔寅骐,肖健华,黄银和,等.基于改进RHT的SAR图像机场区域提取算法[J].电子测量技术,2016,39(2):56-60.QIAO Y Q,XIAO J H,HUANG Y H,et al.Extraction algorithm for airport region in spaceborne SAR imagery based on improved randomised hough transform[J].Journal of Electronic Measurement and Instrumentation,2016,39(2):56-60.
[2]周宇,张杰,王少锋,等.考虑磨耗的钢轨疲劳裂纹萌生寿命预测仿真[J].铁道学报,2016,38(7):91-97.ZHOU Y,ZHANG J,WANG SH F,et al.Simulation on rail head crack initiation life prediction considering rail wear[J].Journal of the China Railway Society,2016,38(7):91-97.
[3]郭火明,王文健,刘腾飞,等.重载铁路钢轨损伤行为分析[J].中国机械工程,2014,25(2):267-272.GUO H M,WANG W J,LIU T F,et al.Analysis of damage behavior of heavy-haul railway rails[J].China Mechanical Engineering,2014,25(2):267-272.
[4]卢超,魏运飞,徐薇.钢轨踏面斜裂纹超声表面波B扫成像检测研究[J].仪器仪表学报,2010,31(10):2272-2278.LU CH,WEI Y F,XU W.Study on B-scan imaging detection for rail tread tilted cracks using ultrasonic surface wave[J].Chinese Journal of Scientific Instrument,2010,31(10):2272-2278.
[5]王典洪,甘胜丰,张伟民,等.基于监督双限制连接Isomap算法的带钢表面缺陷图像分类方法[J].自动化学报,2014,40(5):883-891.WANG D H,GAN SH F,ZHANG W M,et al.Strip surface defect image classication based on double-limited and supervise-connect isomap algorithm[J].Acta Automatica Sinica,2014,40(5):883-891.
[6]余建波,卢笑蕾,宗卫周.基于局部与非局部线性判别分析和高斯混合模型动态集成的晶圆表面缺陷探测与识别[J].自动化学报,2016,42(1):47-59.YU J B,LU X L,ZONG W ZH.Wafer defect detection and recognition based on local and nonlocal linear discriminant analysis and dynamic ensemble of gaussian mixture models[J].Acta Automatica Sinica,2016,42(1):47-59.
[7]孙作雷,茅旭初.基于激光束匹配和图模型的移动机器人相对运动估计误差分析[J].自动化学报,2011,37(2):205-213.SUN Z L,MAO X CH.Quantification of relative movement uncertainty based on laser scanmatching and graphical model for mobile robot[J].Acta Automatica Sinica,2011,37(2):205-213.
[8]周正干,彭地,李洋,等.相控阵超声检测技术中的全聚焦成像算法及其校准研究[J].机械工程学报,2015,51(10):1-7.ZHOU ZH G,PENG D,LI Y,et al.Research on phased array ultrasonic total focusing method and its calibration[J].Journal of Mechanical Engineering,2015,51(10):1-7.
[9]周德强,潘萌,常祥,等.铁磁性构件缺陷的脉冲涡流检测模式研究[J].仪器仪表学报,2017,38(6):1498-1505.ZHOU D Q,PAN M,CHANG X,et al.Research on detection modes of ferromagnetic component defects using pulsed eddy current[J].Chinese Journal of Scientific Instrument,2017,38(6):1498-1505.
[10]张成伟,崔畅.基于脉冲电流放电法的蓄电池内阻在线检测研究[J].国外电子测量技术,2017,36(11):11-14.ZHANG CH W,CUI CH.Research on battery internal resistance online detection based on pulse current discharge method[J].Foreign Electronic Measurement Technology,2017,36(11):11-14.
[11]丁建睿,黄剑华,刘家锋,等.局部特征与多示例学习结合的超声图像分类方法[J].自动化学报,2013,39(6):861-867.DING J R,HUANG J H,LIU J F,et al.Combining local features and multi-instance learning for ultrasound image classification[J].Acta Automatica Sinica,2013,39(6):861-867.
[12]陈振华,卢超,陆铭慧,等.基于声-超声检测的薄钢板多焊点结构完整性评价技术[J].机械工程学报,2013,49(16):57-61.CHEN ZH H,LU CH,LU M H,et al.Integrity evaluation on spot welded construction of thin steel sheet based on acousto-ultrasonic technique[J].Journal of Mechanical Engineering,2013,49(16):57-61.
[13]LI Q,REN S.A real-time visual inspection system for discrete surface defects of rail heads[J].IEEE Transactions on Instrumentation and Measurement,2012,61(8):2189-2199.
[14]LI Q,REN S.A visual detection system for rail surface defects[J].IEEE Transactions on Systems,Man,and Cybernetics,Part C(Applications and Reviews),2012,42(6):1531-1542.
[15]SALVADOR P,NARANJO V,INSA R,et al.Axlebox accelerations:Their acquisition and time-frequency characterisation for railway track monitoring purposes[J].Measurement,2016(82):301-312.
[16]RESENDIZ E,HART J M,AHUJA N.Automated visual inspection of railroad tracks[J].IEEE Transactions on Intelligent Transportation Systems,2013,14(2):751-760.
[17]SUN M,LIN X,WU Z,et al.Non-destructive photoacoustic detecting method for high-speed rail surface defects[C].Instrumentation and Measurement Technology Conference(I2MTC)Proceedings,IEEE,2014:896-900.
[18]GRASSIE S L.Squats and squat-type defects in rails:the understanding to date[C].Proceedings of the Institution of Mechanical Engineers,Part F:Journal of Rail and Rapid Transit,2012,226(3):235-242.
[19]THOMAS H M,HECKEL T,HANSPACH G.Advantage of a combined ultrasonic and eddy current examination for railway inspection trains[J].Insight-Non-Destructive Testing and Condition Monitoring,2007,49(6):341-344.
[20]ALIPPI C,CASAGRANDE E,SCOTTI F,et al.Composite real-time image processing for railways track profile measurement[J].IEEE Transactions on Instrumentation and Measurement,2000,49(3):559-564.
[21]LIANG B,IWNICKI S,BALL A,et al.Adaptive noise cancelling and time-frequency techniques for rail surface defect detection[J].Mechanical Systems and Signal Processing,2015(54):41-51.
[22]MANDRIOTA C,NITTI M,ANCONA N,et al.Filterbased feature selection for rail defect detection[J].Machine Vision and Applications,2004,15(4):179-185.
[23]WANG L,HANG Y,LUO S,et al.Deblurring Gaussian-blur images:A preprocessing for rail head surface defect detection[C].IEEE International Conference on Service Operations,Logistics,and Informatics(SOLI),2011:451-456.
[24]DUBEY A K,JAFFERY Z A.Maximally stable extremal region marking-based railway track surface defect sensing[J].IEEE Sensors Journal,2016,16(24):9047-9052.
[25]HE Z,WANG Y,YIN F,et al.Surface defect detection for high-speed rails using an inverse PM diffusionmodel[J].Sensor Review,2016,36(1):86-97.
[26]贺振东,王耀南,刘洁,等.基于背景差分的高铁钢轨表面缺陷图像分割[J].仪器仪表学报,2016,37(3):640-649.HE ZH D,WANG Y N,LIU J,et al.Background differencing-based high-speed rail surface defect image segmentation[J].Chinese Journal of Scientific Instrument,2016,37(3):640-649.
[27]贺振东,王耀南,毛建旭,等.基于反向PM扩散的钢轨表面缺陷视觉检测[J].自动化学报,2014,40(8):1667-1679.HE ZH D,WANG Y N,MAO J X,et al.Research on inverse P-M diffusion-based rail surface defect detection[J].Acta Automatica Sinica,2014,40(8):1667-1679.
[28]TA爦TIMUR C,KARAKSE M,AKIN E,et al.Rail defect detection with real time image processing technique[C].14th International Conference on Industrial Informatics(INDIN),IEEE,2016:411-415.
[29]HAJIZADEH S,NNEZ A,TAX D M J.Semisupervised rail defect detection from imbalanced image data[J].IFAC-Papers on Line,2016,49(3):78-83.
[30]FAGHIH-ROOHI S,HAJIZADEH S,NEZ A,et al.Deep convolutional neural networks for detection of rail surface defects[C].International Joint Conference on Neural Networks(IJCNN),IEEE,2016:2584-2589.
[31]GIBERT X,PATEL V M,CHELLAPPA R.Deep multitask learning for railway track inspection[J].IEEE Transactions on Intelligent Transportation Systems,2017,18(1):153-164.
[32]黄金,周先春,吴婷,等.混合维纳滤波与改进型TV的图像去噪模型[J].电子测量与仪器学报,2017,31(10):1659-1666.HUANG J,ZHOU X CH,WU T,et al.Image denoising model based on mixing Wiener filtering and improved total variation[J].Journal of Electronic Measurement and Instrumentation,2017,31(10):1659-1666.
[33]GONG Y,SBALZARINI I F.Curvature filters efficiently reduce certain variational energies[J].IEEE Transactions on Image Processing,2017,26(4):1786-1798.
[34]张辉,金侠挺.基于曲率滤波和反向PM电动车充电孔检测方法[J].仪器仪表学报,2016,37(7):1626-1638.ZHANG H,JIN X T.Detection method for electric vehicle charging hole based on curvature filter and inverse P-M diffusion[J].Chinese Journal of Scientific Instrument,2016,37(7):1626-1638.
[35]SANJAY-GOPAL S,HEBERT T J.Bayesian pixel classification using spatially variant finite mixtures and the generalized EM algorithm[J].IEEE Transactions on Image Processing,1998,7(7):1014-1028.
[36]NGUYEN T M,WU Q M J.Fast and robust spatially constrained gaussian mixture model for image segmentation[J].IEEE transactions on circuits and systems for video technology,2013,23(4):621-635.
[37]NIKOU C,GALATSANOS N P,LIKAS A C.A classadaptive spatially variant mixture model for image segmentation[J].IEEE Transactions on Image Processing,2007,16(4):1121-1130.
[38]乔寅骐,肖健华,黄银和,等.基于改进RHT的SAR图像机场区域提取算法[J].电子测量技术,2016,39(2):56-60.QIAO Y Q,XIAO J H,HUANG Y H,et al.Extraction algorithm for airport region in spaceborne SAR imagery based on improved randomised hough transform[J].Journal of Electronic Measurement and Instrumentation,2016,39(2):56-60.