基于SVM的高速公路路面浅层病害的自动检测算法研究
|
摘要
随着高速公路通车里程的不断增加和路网规模的不断扩大以及使用年限的不断增长,高速公路的养护工作变得日趋繁重和重要。探地雷达(groundpenetrating radar,GPR)作为一种快速、连续、安全、高精度、高分辨率的无损实时探测工具,越来越广泛地应用于高速公路路面浅层病害检测。但是,GPR跟光学成像设备不同,它不能直接反映目标的特征。因此,使用GPR勘查高速公路路面浅层质量时,如何由获取的GPR数据解释高速公路路面浅层质量状况成为问题的关键。
本课题结合现代数字处理、信号检测技术及模式识别算法,实现了高速公路路面浅层病害的自动检测,其主要技术手段及具体研究内容为:
1、探地雷达原始数据预处理方法的研究,主要内容包括运用滑动平均法对GPR原始数据进行噪声抑制、使用包络检波器和阈值检测技术自动检测路面浅层层界面、使用Savitzky-Golay滤波器进行层界面平滑,以及ROI(region ofinterest)提取。
2、基于时域和小波域回波信号的特征提取。总共在时域与小波域内提取了6个特征,即时域的三个特征:信号的最大幅值MAX_s、信号幅值的平均绝对偏差MAD_s、原始信号幅值互相关XCORR_s;小波域的三个特征:合成信号在各级小波上的互相关之和XCORR_(d123)、第三级小波近似系数a3的最大幅值M-AX_(a3)、第三级小波近似系数a3的平均绝对偏差MAD_(a3)。
3、基于支持向量机(support vector machine,SVM)的路面浅层病害检测。利用专家经验,从层界面反射信号中提取出好路面和坏路面(有病害路面)样本信号,划分成训练样本和测试样本。提取其6个时域小波域特征,用训练样本训练SVM,得到相应的支持向量网络,并把测试样本作为输入,由SVM进行特征分类,从而检测出路面浅层是否存在病害。
The maintenance of highway has been getting more and more important and labor-intensive because roadway network system is extending, and most of built roads are aging. As a fast, continuous, secure, nondestructive, real-time detection tool with high precision, GPR (ground penetrating radar) has been used in highway pavement distress detection. However, GPR is different from optical imaging equipment as it is not able to reflect the feature of object directly. Thus, during GPR is used to survey the quality of highway pavement, how to interpret the acquired GPR data as the quality status of highway pavement becomes the key issue.
Using modern digital processing, signal detection and pattern recognition algorithm, highway pavement distress has been detected automatically in this project. The main techniques and detailed research work are list out as follows:
1. Research on GPR original data preprocessing algorithm for clutter suppression, layer interface detection, layer interface smoothing, and ROI extraction.
2. Research on feature extraction algorithm in time domain and wavelet domain. Extract three features from time domain: maximum amplitudes (MAX_S), mean absolute deviation (MAD) of amplitudes (MAD_S), and cross-correlation of original signals amplitude (XCORR_S). And extract another three features from wavelet: summation of cross-correlation of synthesized signals at all level of wavelet (XCORR_(d123)), magnitude of wavelet approximate coefficient a_3 (MAX_(a3)), and MAD of approximate coefficient a_3 (MAD_(a3)).
3. Research on pavement distress detection algorithm based on SVM. Basing on expert experiences training samples and testing samples are selected from GPR reflected signals in good and deteriorative pavement. Train SVM with six extracted features of training samples obtaining support vector network correspondingly, then input testing samples into SVM and implement feature classification determining whether there is distress in pavement or not.
本课题结合现代数字处理、信号检测技术及模式识别算法,实现了高速公路路面浅层病害的自动检测,其主要技术手段及具体研究内容为:
1、探地雷达原始数据预处理方法的研究,主要内容包括运用滑动平均法对GPR原始数据进行噪声抑制、使用包络检波器和阈值检测技术自动检测路面浅层层界面、使用Savitzky-Golay滤波器进行层界面平滑,以及ROI(region ofinterest)提取。
2、基于时域和小波域回波信号的特征提取。总共在时域与小波域内提取了6个特征,即时域的三个特征:信号的最大幅值MAX_s、信号幅值的平均绝对偏差MAD_s、原始信号幅值互相关XCORR_s;小波域的三个特征:合成信号在各级小波上的互相关之和XCORR_(d123)、第三级小波近似系数a3的最大幅值M-AX_(a3)、第三级小波近似系数a3的平均绝对偏差MAD_(a3)。
3、基于支持向量机(support vector machine,SVM)的路面浅层病害检测。利用专家经验,从层界面反射信号中提取出好路面和坏路面(有病害路面)样本信号,划分成训练样本和测试样本。提取其6个时域小波域特征,用训练样本训练SVM,得到相应的支持向量网络,并把测试样本作为输入,由SVM进行特征分类,从而检测出路面浅层是否存在病害。
The maintenance of highway has been getting more and more important and labor-intensive because roadway network system is extending, and most of built roads are aging. As a fast, continuous, secure, nondestructive, real-time detection tool with high precision, GPR (ground penetrating radar) has been used in highway pavement distress detection. However, GPR is different from optical imaging equipment as it is not able to reflect the feature of object directly. Thus, during GPR is used to survey the quality of highway pavement, how to interpret the acquired GPR data as the quality status of highway pavement becomes the key issue.
Using modern digital processing, signal detection and pattern recognition algorithm, highway pavement distress has been detected automatically in this project. The main techniques and detailed research work are list out as follows:
1. Research on GPR original data preprocessing algorithm for clutter suppression, layer interface detection, layer interface smoothing, and ROI extraction.
2. Research on feature extraction algorithm in time domain and wavelet domain. Extract three features from time domain: maximum amplitudes (MAX_S), mean absolute deviation (MAD) of amplitudes (MAD_S), and cross-correlation of original signals amplitude (XCORR_S). And extract another three features from wavelet: summation of cross-correlation of synthesized signals at all level of wavelet (XCORR_(d123)), magnitude of wavelet approximate coefficient a_3 (MAX_(a3)), and MAD of approximate coefficient a_3 (MAD_(a3)).
3. Research on pavement distress detection algorithm based on SVM. Basing on expert experiences training samples and testing samples are selected from GPR reflected signals in good and deteriorative pavement. Train SVM with six extracted features of training samples obtaining support vector network correspondingly, then input testing samples into SVM and implement feature classification determining whether there is distress in pavement or not.
引文
[1]林红梅.辉煌60年:中国公路通车总里程60年增长45倍[EB/01].http://www.gov.cn/jrzg /2009-08-16/content_1393479.htm.2009-08-16
[2]焦彦龙.2007年交通发展亮点[EB/01].http://www.moc.gov.cn/zhuzhan/gonzuohuiyi/quanguojiaotong_GZHY/2008jiaotonggong_HY/Meitibaodao/200801/t20080107_457360.ht ml.2008-01-07.
[3]国家高速公路网规划[R].交通部规划研究院,2004.
[4]张彦杰.探地雷达在道路检测中的应用研究[D].[硕士论文].吉林大学,2007.
[5]沙庆林.高速公路沥青路面早期损坏与对策[J].长沙理工大学学报(自然科学版),2006,3(3):1-6.
[6]沙庆林.进一步提高高等级公路沥青路面的使用性能和耐久性[J].中国公路学报,1995,1(1):7-14.
[7]潘冬子,章光,刘世奇,李颖.混凝土梁无损检测新技术及其进展[J].公路,2004,2(4):18-23
[8]R.Evans,M.Frost,M.Stonecliffe-Jones and N.Dixon.A review of pavement assessment using ground penetrating radar(GPR)[C].12th International Conference on Ground Penetrating Radar,Birmingham,UK,June,2008.
[9]R.B.Wu,X.Li,J.Li.Continuous pavement profiling with ground penetrating radar[C].IEE Proceedings of Radar,Sonar and Navigation,2003,149(4):183-193.
[10]A.Benedetto,A.Pensa.Indirect diagnosis of pavement structural damages using surface GPR reflection techniques[J].Journal of Applied Geophysics,2007(62):107-123.
[11]张全升.高速公路路面和路基病害检测理论与方法研究[D].[博士论文].中国地质大学,2007.
[12]K.R.Maser,T.Scullion.Automated detection of pavement layer thickness and subsurface moisture using ground penetrating radar[J].Journal of Transportation Board,1996,115(2):86-89.
[13]I.L.Al-Qadi,S.Lahouar,K.Jiang.Analysis tool for determining flexible pavement layer thickness at highway speed[C].Transportation Research Board of the National Academics 85~(th) Annual Meeting,Washington,D.C.,U.S.A,March,2006.
[14]T.Saarenketo,T.Scullion.Road evaluation with ground penetrating radar[J].Journal of Applied Geophysics,2000,43(2-4):119-138.
[15]D.R.Sonyok,J.Zhang.Ground penetration radar for highway infrastructure condition diagnostics:overview of current applications and future development[C].Transportation Research Board of the National Academics 87~(th) Annual Meeting,Washington,D.C.,U.S.A,February,2008.
[16]I.L.Al-Quadi,S.Lahouar,K.Jiang.Validation of ground penetrating radar accuracy for estimating pavement layer thickness[C].Transportation Research Board of the National Academics 84~(th) Annual Meeting,Washington,D.C.,U.S.A,August,2005.
[17]H.Stoffregen,U.Yaramanci,T.Zenker,G.Wessolek.Accuracy of soil water content measurements using ground penetrating radar:comparison of ground penetrating radar and lysimeter data[J].Journal of Hydrology,2002,267(3-4):201-206.
[18]S.Laurens,J.P.Balayssac,J.Rhazi,G.Klysz,and G.Arliguie.Non-destructive evaluation of concrete moisture by GPR:Experimental study and direct modeling[J].Materials and Structures,2005,38(9):827-832.
[19]S.Lahouar,I.L.Al-Qadi,Automatic detection of multiple pavement layers from GPR data[J].NDT&E,2008,41(2):69-81.
[20]U.Spagnolini,Permittivity measurements of multilayered media with monostatic pulse radar [J].IEEE Transaction on Geoscience and Sensing,1997,35(2):454-463.
[21]A.Benedetto,F.Benedetto,M.R.D.Blasiis,G.Giunta,Reliability of signal processing technique for pavement damages detection and classification using ground penetrating radar [J].IEEE Sensor Journal,2005,5(3):471-480.
[22]S.Lahouar.Development of Data Analysis Algorithms for Interpretation of Ground Penetrating Radar Data[D].Virginia Polytechnic Institute and State University,Blacksburg,Virginia,2003.
[23]Brawijaya.A new methodology to diagnose pavement subsurface condition using ground penetrating radar[D].Rensselaer Polytechnic Institute,Troy,New York,2005.
[24]江玉乐,黄鑫,张楠.探地雷达在公路隧道衬砌检测中的应用[J].煤田地质与勘探,2008,36(2):76-78.
[25]朱希安,苑守成.探地雷达在公路质量无损检测中的应用研究[J].煤田地质与勘探,2002,30(5):47-51.
[26]D.J.Daniels.Surface-Penetrating Radar[J].Electronics & Communication Engineering Journal,1996,8(4):165-182.
[27]周辉林.探地雷达回波信号特征选择与分类[D].[博士论文].武汉大学,2006.
[28]李大心.探地雷达方法与应用[M].北京:地质出版社,1994.
[29]杨建国.小波分析及其工程应用[M].北京:机械工业出版社,2005.
[30]胡广书.现代信号处理教程[M].北京:清华大学出版社,2004.
[31]王明详,宁宇蓉,王晋国.基于Mallat算法的一维离散小波变换的实现[J].西北大学学报(自然科学版),2006,3(7):364-368.
[32]J.Li,R.Wu.An efficient algorithm for time delay estimation[J].IEEE Trans Signal Process,1998,46(8):2231-2236.
[33]樊勇,周正欧,徐嘉莉.利用非线性拟合抑制前视探地雷达杂波[J].现代雷达,2008,30(12):43-48.
[34]A.V.D.Merwe and I.J.Gupta.A novel signal processing technique for clutter reduction in GPR measurements of small,shallow land mines[J].IEEE Transactions on Geoscience and Remote Sensing,2000,38(6):2627-2637.
[35]廖立坚,杨新安,黄凯,叶培旭.小波域KL变换用于去除探地雷达直耦波[J].工程勘察,2008,6(24):72-75.
[36]裴益轩,郭名.滑动平均法的基本原理及应用[J].火炮发射与控制学报,2001,1(6):21-23.
[37]孙建龙.滑动平均“窗口”选择初探[J].物探与化探,1985,9(3):235-238.
[38]C.A.Glasbey,R.Jones.Fast computation of moving average and related filters in octagonal windows[J].Pattern Recognition Letters,1997,18:555-565.
[39]D.Donoho,I.Johnstone.Ideal spatial adaptation via wavelet shrinkage[J].Biometrika.1994,81(3):425-455.
[40]M.Lang et al.Noise reduction using an undecimated discrete wavelet transform[J].IEEE Signal Processing Letters,1995,3:10-12.
[41]W.Gander,J.Hrebicek and S.Barton.Solving Problems in Scientific Computing Using Maple and Matlab(Fourth Edition)[M].New York:Springer-Verlag Berlin Heidelberg,1993.
[42]刘华林,杨万麟,梅元媛,赵建宏.修正最近特征分类器及其在雷达目标识别中的应用[J].计算机应用,2007,4(38):894-896.
[43]谷秧波,武妍,王守觉,朱君波.原点无关最近特征分类器及其在人脸识别的应用[J].同济大学学报(自然科学版),2006,10(24):1398-1402.
[44]毛勇.基于支持向量机的特征选择方法的研究与应用[D].[博士论文].浙江大学,2006.
[45]边肇祺,张学工.模式识别[M].北京:清华大学,1999.
[46]V.N.Vapnik.The Nature of Statistical Learning Theory[M].New York:Springer-Verlag Berlin Heidelberg,1999.
[47]祁亨年.支持向量机及其应用研究综述[J].计算机工程,2004,30(10):6-9.
[48]蒋琳.基于支持向量机的特征提取方法研究与应用[D].[硕士论文].湖南大学,2006.
[49]安金龙,王正欧,马振平.一种新的支持向量机多类分类方法[J].信息与控制,2004,33(3):262-267.
[50]V.N.Vapnik.An overview of statistical learning theory[J].IEEE Transaction on Neural Networks,1999,10(5):988-999.
[2]焦彦龙.2007年交通发展亮点[EB/01].http://www.moc.gov.cn/zhuzhan/gonzuohuiyi/quanguojiaotong_GZHY/2008jiaotonggong_HY/Meitibaodao/200801/t20080107_457360.ht ml.2008-01-07.
[3]国家高速公路网规划[R].交通部规划研究院,2004.
[4]张彦杰.探地雷达在道路检测中的应用研究[D].[硕士论文].吉林大学,2007.
[5]沙庆林.高速公路沥青路面早期损坏与对策[J].长沙理工大学学报(自然科学版),2006,3(3):1-6.
[6]沙庆林.进一步提高高等级公路沥青路面的使用性能和耐久性[J].中国公路学报,1995,1(1):7-14.
[7]潘冬子,章光,刘世奇,李颖.混凝土梁无损检测新技术及其进展[J].公路,2004,2(4):18-23
[8]R.Evans,M.Frost,M.Stonecliffe-Jones and N.Dixon.A review of pavement assessment using ground penetrating radar(GPR)[C].12th International Conference on Ground Penetrating Radar,Birmingham,UK,June,2008.
[9]R.B.Wu,X.Li,J.Li.Continuous pavement profiling with ground penetrating radar[C].IEE Proceedings of Radar,Sonar and Navigation,2003,149(4):183-193.
[10]A.Benedetto,A.Pensa.Indirect diagnosis of pavement structural damages using surface GPR reflection techniques[J].Journal of Applied Geophysics,2007(62):107-123.
[11]张全升.高速公路路面和路基病害检测理论与方法研究[D].[博士论文].中国地质大学,2007.
[12]K.R.Maser,T.Scullion.Automated detection of pavement layer thickness and subsurface moisture using ground penetrating radar[J].Journal of Transportation Board,1996,115(2):86-89.
[13]I.L.Al-Qadi,S.Lahouar,K.Jiang.Analysis tool for determining flexible pavement layer thickness at highway speed[C].Transportation Research Board of the National Academics 85~(th) Annual Meeting,Washington,D.C.,U.S.A,March,2006.
[14]T.Saarenketo,T.Scullion.Road evaluation with ground penetrating radar[J].Journal of Applied Geophysics,2000,43(2-4):119-138.
[15]D.R.Sonyok,J.Zhang.Ground penetration radar for highway infrastructure condition diagnostics:overview of current applications and future development[C].Transportation Research Board of the National Academics 87~(th) Annual Meeting,Washington,D.C.,U.S.A,February,2008.
[16]I.L.Al-Quadi,S.Lahouar,K.Jiang.Validation of ground penetrating radar accuracy for estimating pavement layer thickness[C].Transportation Research Board of the National Academics 84~(th) Annual Meeting,Washington,D.C.,U.S.A,August,2005.
[17]H.Stoffregen,U.Yaramanci,T.Zenker,G.Wessolek.Accuracy of soil water content measurements using ground penetrating radar:comparison of ground penetrating radar and lysimeter data[J].Journal of Hydrology,2002,267(3-4):201-206.
[18]S.Laurens,J.P.Balayssac,J.Rhazi,G.Klysz,and G.Arliguie.Non-destructive evaluation of concrete moisture by GPR:Experimental study and direct modeling[J].Materials and Structures,2005,38(9):827-832.
[19]S.Lahouar,I.L.Al-Qadi,Automatic detection of multiple pavement layers from GPR data[J].NDT&E,2008,41(2):69-81.
[20]U.Spagnolini,Permittivity measurements of multilayered media with monostatic pulse radar [J].IEEE Transaction on Geoscience and Sensing,1997,35(2):454-463.
[21]A.Benedetto,F.Benedetto,M.R.D.Blasiis,G.Giunta,Reliability of signal processing technique for pavement damages detection and classification using ground penetrating radar [J].IEEE Sensor Journal,2005,5(3):471-480.
[22]S.Lahouar.Development of Data Analysis Algorithms for Interpretation of Ground Penetrating Radar Data[D].Virginia Polytechnic Institute and State University,Blacksburg,Virginia,2003.
[23]Brawijaya.A new methodology to diagnose pavement subsurface condition using ground penetrating radar[D].Rensselaer Polytechnic Institute,Troy,New York,2005.
[24]江玉乐,黄鑫,张楠.探地雷达在公路隧道衬砌检测中的应用[J].煤田地质与勘探,2008,36(2):76-78.
[25]朱希安,苑守成.探地雷达在公路质量无损检测中的应用研究[J].煤田地质与勘探,2002,30(5):47-51.
[26]D.J.Daniels.Surface-Penetrating Radar[J].Electronics & Communication Engineering Journal,1996,8(4):165-182.
[27]周辉林.探地雷达回波信号特征选择与分类[D].[博士论文].武汉大学,2006.
[28]李大心.探地雷达方法与应用[M].北京:地质出版社,1994.
[29]杨建国.小波分析及其工程应用[M].北京:机械工业出版社,2005.
[30]胡广书.现代信号处理教程[M].北京:清华大学出版社,2004.
[31]王明详,宁宇蓉,王晋国.基于Mallat算法的一维离散小波变换的实现[J].西北大学学报(自然科学版),2006,3(7):364-368.
[32]J.Li,R.Wu.An efficient algorithm for time delay estimation[J].IEEE Trans Signal Process,1998,46(8):2231-2236.
[33]樊勇,周正欧,徐嘉莉.利用非线性拟合抑制前视探地雷达杂波[J].现代雷达,2008,30(12):43-48.
[34]A.V.D.Merwe and I.J.Gupta.A novel signal processing technique for clutter reduction in GPR measurements of small,shallow land mines[J].IEEE Transactions on Geoscience and Remote Sensing,2000,38(6):2627-2637.
[35]廖立坚,杨新安,黄凯,叶培旭.小波域KL变换用于去除探地雷达直耦波[J].工程勘察,2008,6(24):72-75.
[36]裴益轩,郭名.滑动平均法的基本原理及应用[J].火炮发射与控制学报,2001,1(6):21-23.
[37]孙建龙.滑动平均“窗口”选择初探[J].物探与化探,1985,9(3):235-238.
[38]C.A.Glasbey,R.Jones.Fast computation of moving average and related filters in octagonal windows[J].Pattern Recognition Letters,1997,18:555-565.
[39]D.Donoho,I.Johnstone.Ideal spatial adaptation via wavelet shrinkage[J].Biometrika.1994,81(3):425-455.
[40]M.Lang et al.Noise reduction using an undecimated discrete wavelet transform[J].IEEE Signal Processing Letters,1995,3:10-12.
[41]W.Gander,J.Hrebicek and S.Barton.Solving Problems in Scientific Computing Using Maple and Matlab(Fourth Edition)[M].New York:Springer-Verlag Berlin Heidelberg,1993.
[42]刘华林,杨万麟,梅元媛,赵建宏.修正最近特征分类器及其在雷达目标识别中的应用[J].计算机应用,2007,4(38):894-896.
[43]谷秧波,武妍,王守觉,朱君波.原点无关最近特征分类器及其在人脸识别的应用[J].同济大学学报(自然科学版),2006,10(24):1398-1402.
[44]毛勇.基于支持向量机的特征选择方法的研究与应用[D].[博士论文].浙江大学,2006.
[45]边肇祺,张学工.模式识别[M].北京:清华大学,1999.
[46]V.N.Vapnik.The Nature of Statistical Learning Theory[M].New York:Springer-Verlag Berlin Heidelberg,1999.
[47]祁亨年.支持向量机及其应用研究综述[J].计算机工程,2004,30(10):6-9.
[48]蒋琳.基于支持向量机的特征提取方法研究与应用[D].[硕士论文].湖南大学,2006.
[49]安金龙,王正欧,马振平.一种新的支持向量机多类分类方法[J].信息与控制,2004,33(3):262-267.
[50]V.N.Vapnik.An overview of statistical learning theory[J].IEEE Transaction on Neural Networks,1999,10(5):988-999.