基于无人机LIDAR数据多尺度特征的沥青路面病害提取方法
|
摘要
目的针对大范围公路路面病害监测需求,提出基于低空无人机激光雷达遥感数据和随机森林分类算法,构建沥青路面病害目标的遥感识别模型。方法首先,基于激光点云高程信息提取多尺度表面粗糙度和高斯曲率指数,以及利用激光反射强度影像提取路面和病害目标的几何特征,然后基于提取的48个多尺度统计特征利用随机森林算法建立了沥青路面坑槽与塌陷两类主要病害的识别模型。采用搭载于ScoutB1-100低空无人直升机平台的RIEGL-VUX100激光雷达扫描仪,获取了新疆石河子市与沙湾县交界处的一段县级沥青道路的激光点云数据,对所提出方法和模型进行了验证。结果本文所提出的模型可较好识别路面的塌陷与坑槽病害目标,以地面调查和目视解译结果为参照的验证精度为92.3%,Kappa系数为0.902,优于其他两种常用的机器学习分类模型,可为公路养护部门提供一种新的快速路面病害监测方法。
Objective To meet the requirements of timely monitoring highway pavement distresses,this study presented a new approach for the retrieval of asphalt pavement distresses from the point cloud data acquired by a low-altitude unmanned aerial vehicle(UAV) LIDAR system based on the random forest classifier(RFC).Methods The multi-scale surface roughness and Gaussian curvature indices were first extracted from the digital surface model(DSM) data that were generated from the LIDAR point cloud,and the geometric features of image objects were derived from the segmented laser intensity image.After that,the resultant 48 multi-scale statistical features were together adopted to build up the RFC model for identifying the asphalt pavement distresses such as potholes and subsidence.To validate our proposed classification model,a RIEGL-VUX100 laser scanner mounted on a Scout B1-100 unmanned helicopter were utilized to acquire the point cloud data over an area a county-level asphalt highway road goes through.Results show that the proposed method can identify and retrieve the distresses of potholes and subsidence in the study area.Results The RF algorithm achieved the identification of pavement distresses with an overall accuracy of 92.3% and Kappa coefficient of 0.902,which outperformed the maximum likelihood classifier(MLC) and support vector machine(SVM) classifier.Thus,the proposed approach can offer a valuable tool for the department of highway maintenance to timely acquire road pavement distresses imformation.
Objective To meet the requirements of timely monitoring highway pavement distresses,this study presented a new approach for the retrieval of asphalt pavement distresses from the point cloud data acquired by a low-altitude unmanned aerial vehicle(UAV) LIDAR system based on the random forest classifier(RFC).Methods The multi-scale surface roughness and Gaussian curvature indices were first extracted from the digital surface model(DSM) data that were generated from the LIDAR point cloud,and the geometric features of image objects were derived from the segmented laser intensity image.After that,the resultant 48 multi-scale statistical features were together adopted to build up the RFC model for identifying the asphalt pavement distresses such as potholes and subsidence.To validate our proposed classification model,a RIEGL-VUX100 laser scanner mounted on a Scout B1-100 unmanned helicopter were utilized to acquire the point cloud data over an area a county-level asphalt highway road goes through.Results show that the proposed method can identify and retrieve the distresses of potholes and subsidence in the study area.Results The RF algorithm achieved the identification of pavement distresses with an overall accuracy of 92.3% and Kappa coefficient of 0.902,which outperformed the maximum likelihood classifier(MLC) and support vector machine(SVM) classifier.Thus,the proposed approach can offer a valuable tool for the department of highway maintenance to timely acquire road pavement distresses imformation.
引文
[1]中国交通运输部.2013年交通运输行业发展统计公报[EB].2013.http://www.China highway.com/news/2017/1101039.php.
[2]Pan Y,Zhang X,Tian J,et al.Mapping asphalt pavement aging and condition using multiple endmember spectral mixture analysis in Beijing,China[J].Journal of Applied Remote Sensing,2017,11(1):016003-1-18.
[3]Shahin M.Y.Pavement management for airports,roads,and parking lots(2eds)[M].Series in Lecture Notes in Mathematics,2008,Springer.
[4]潘一凡.基于光谱分析和卷积神经网络的沥青路面老化与病害遥感监测方法[D].北京:北京大学博士学位论文,2018.
[5]Sun N,Zhang J,Huang G,Zhao Z,Lu J.Review of Road Extraction Methods from SAR Image[A/OL].IOP Conference Series:Earth and Environmental Science,2014,17:012245.10.1088/1755-1315/17/1/012245..
[6]Shahi K,Shafri H Z M,Taherzadeh E,et al.A novel spectral index to automatically extract road networks from World View-2 satellite imagery[J].The Egyptian Journal of Remote Sensing and Space Science,2015,18(1):27-33.
[7]Moranduzzo T,Melgani F.Automatic car counting method for unmanned aerial vehicle images[J].IEEE Transactions on Geoscience and Remote Sensing,2014,52(3):1635-1647.
[8]Mei A,Salvatori R,Fiore N,et al.Integration of field and laboratory spectral data with multi-resolution remotely sensed imagery for asphalt surface differentiation[J].Remote Sensing,2014,6(4):2765-2781.
[9]Themistocleous K,Neocleous K,Pilakoutas K,et al.Damage assessment using advanced non-intrusive inspection methods:integration of space,UAV,GPR,and field spectroscopy[A].Proceedings of Second International Conference on Remote Sensing and Geoinformation of the Environment(RS-Cy2014),Vol.9229,Paphos,Cyprus,2014.
[10]Paska E P.State-of-the-art remote sensing geospatial technologies in support of transportation monitoring and management[D].USA:The Ohio State University,2009.
[11]Pulugurtha S S,Ogunro V,Pando M A,et al.Preliminary results towards developing thresholds for pavement condition maintenance:safety perspective[J].Procedia-Social and Behavioral Sciences,2013,104:302-311.
[12]郭姣.基于车载激光的道路平整度检测系统研究[D].北京:首都师范大学,2013.
[13]Giacinto G,Roli F.An approach to the automatic design of multiple classifier systems[J].Pattern Recognition Letters,2001,22(1):25-33.
[14]Cutler D R,Edwards Jr T C,Beard K H,et al.Random forests for classification in ecology[J].Ecology,2007,88(11):2783-2792.
[15]Guo L,Chehata N,Mallet C,et al.Relevance of airborne lidar and multispectral image data for urban scene classification using Random Forests[J].ISPRS Journal of Photogrammetry and Remote Sensing,2011,66(1):56-66.
[16]高仁强,张显峰,孙权,等.基于无人机Li DAR数据的公路路面监测和平整度评价方法研究[J].应用基础与工程科学学报,2018,26(4):681-696.Gao R Q,Zhang X F,Sun Q,et al.Road pawement monitoring and roughness assessment based on UAV Li DAR data[J].Journal of Basic Science and Engineering,2018,26(4):681-696.
[17]黄先锋,李娜,张帆,等.利用Li DAR点云强度的十字剖分线法道路提取[J].武汉大学学报(信息科学版),2015,40(12):1563-1569.Huang X F,Li N,Zhang F,et al.Automatic power lines extraction method from airborne Li DAR point cloud[J].Geomatics and Information Science of Wuhan University,2015,40(12):1563-1569.
[18]苏理宏,李小文,黄裕霞.遥感尺度问题研究进展[J].地球科学进展,2001,16(4):544-548.Su L H,Li X W,Huang Y X.An review on scale in remote sensing[J].Advance in Earth Sciences,2001,16(4):544-548.
[19]Breiman L.Bagging predictors[J].Machine learning,1996,24(2):123-140.
[20]Ho T K.The random subspace method for constructing decision forests[J].IEEE Transactions on Pattern Analysis and&Machine Intelligence,1998,20:832-844.
[21]Breiman,L.Random forests[J].Machine Learning,2001,45:5-32.
[22]Chan J,Paelinckx D.Evaluation of random forest and adaboost tree-based ensemble classification and spectral band selection for ecotype mapping using airborne hyperspectral imagery[J].Remote Sensing of Environment,2008,112(6):2999-3011.
[23]Tian S,Zhang X,Tian J,Sun Q.Random forest classification of wetland land covers from multi-sensor data in the arid region of Xinjiang,China[J].Remote Sensing,2016,8(11):954.
[2]Pan Y,Zhang X,Tian J,et al.Mapping asphalt pavement aging and condition using multiple endmember spectral mixture analysis in Beijing,China[J].Journal of Applied Remote Sensing,2017,11(1):016003-1-18.
[3]Shahin M.Y.Pavement management for airports,roads,and parking lots(2eds)[M].Series in Lecture Notes in Mathematics,2008,Springer.
[4]潘一凡.基于光谱分析和卷积神经网络的沥青路面老化与病害遥感监测方法[D].北京:北京大学博士学位论文,2018.
[5]Sun N,Zhang J,Huang G,Zhao Z,Lu J.Review of Road Extraction Methods from SAR Image[A/OL].IOP Conference Series:Earth and Environmental Science,2014,17:012245.10.1088/1755-1315/17/1/012245..
[6]Shahi K,Shafri H Z M,Taherzadeh E,et al.A novel spectral index to automatically extract road networks from World View-2 satellite imagery[J].The Egyptian Journal of Remote Sensing and Space Science,2015,18(1):27-33.
[7]Moranduzzo T,Melgani F.Automatic car counting method for unmanned aerial vehicle images[J].IEEE Transactions on Geoscience and Remote Sensing,2014,52(3):1635-1647.
[8]Mei A,Salvatori R,Fiore N,et al.Integration of field and laboratory spectral data with multi-resolution remotely sensed imagery for asphalt surface differentiation[J].Remote Sensing,2014,6(4):2765-2781.
[9]Themistocleous K,Neocleous K,Pilakoutas K,et al.Damage assessment using advanced non-intrusive inspection methods:integration of space,UAV,GPR,and field spectroscopy[A].Proceedings of Second International Conference on Remote Sensing and Geoinformation of the Environment(RS-Cy2014),Vol.9229,Paphos,Cyprus,2014.
[10]Paska E P.State-of-the-art remote sensing geospatial technologies in support of transportation monitoring and management[D].USA:The Ohio State University,2009.
[11]Pulugurtha S S,Ogunro V,Pando M A,et al.Preliminary results towards developing thresholds for pavement condition maintenance:safety perspective[J].Procedia-Social and Behavioral Sciences,2013,104:302-311.
[12]郭姣.基于车载激光的道路平整度检测系统研究[D].北京:首都师范大学,2013.
[13]Giacinto G,Roli F.An approach to the automatic design of multiple classifier systems[J].Pattern Recognition Letters,2001,22(1):25-33.
[14]Cutler D R,Edwards Jr T C,Beard K H,et al.Random forests for classification in ecology[J].Ecology,2007,88(11):2783-2792.
[15]Guo L,Chehata N,Mallet C,et al.Relevance of airborne lidar and multispectral image data for urban scene classification using Random Forests[J].ISPRS Journal of Photogrammetry and Remote Sensing,2011,66(1):56-66.
[16]高仁强,张显峰,孙权,等.基于无人机Li DAR数据的公路路面监测和平整度评价方法研究[J].应用基础与工程科学学报,2018,26(4):681-696.Gao R Q,Zhang X F,Sun Q,et al.Road pawement monitoring and roughness assessment based on UAV Li DAR data[J].Journal of Basic Science and Engineering,2018,26(4):681-696.
[17]黄先锋,李娜,张帆,等.利用Li DAR点云强度的十字剖分线法道路提取[J].武汉大学学报(信息科学版),2015,40(12):1563-1569.Huang X F,Li N,Zhang F,et al.Automatic power lines extraction method from airborne Li DAR point cloud[J].Geomatics and Information Science of Wuhan University,2015,40(12):1563-1569.
[18]苏理宏,李小文,黄裕霞.遥感尺度问题研究进展[J].地球科学进展,2001,16(4):544-548.Su L H,Li X W,Huang Y X.An review on scale in remote sensing[J].Advance in Earth Sciences,2001,16(4):544-548.
[19]Breiman L.Bagging predictors[J].Machine learning,1996,24(2):123-140.
[20]Ho T K.The random subspace method for constructing decision forests[J].IEEE Transactions on Pattern Analysis and&Machine Intelligence,1998,20:832-844.
[21]Breiman,L.Random forests[J].Machine Learning,2001,45:5-32.
[22]Chan J,Paelinckx D.Evaluation of random forest and adaboost tree-based ensemble classification and spectral band selection for ecotype mapping using airborne hyperspectral imagery[J].Remote Sensing of Environment,2008,112(6):2999-3011.
[23]Tian S,Zhang X,Tian J,Sun Q.Random forest classification of wetland land covers from multi-sensor data in the arid region of Xinjiang,China[J].Remote Sensing,2016,8(11):954.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |