详情

当前位置：首页> 国外交通科技报告数据库 >详情

原文传递 Development of an Asphalt Pavement Raveling Detection Algorithm Using Emerging 3D Laser Technology and Macrotexture Analysis

题名：	Development of an Asphalt Pavement Raveling Detection Algorithm Using Emerging 3D Laser Technology and Macrotexture Analysis
作者：	Tsai, Y.; Wang, Z.
关键词：	Asphalt pavement materials##Raveling##Emerging 3D Laser Technology##Development##Detection algorithms##Macrotexture##Aggregation behavior##Automatic rutting##Pavement markings##Georgia (state)##Georgia Department of Transportation (GDOT)##
摘要：	The product of this IDEA concept exploration research project includes the algorithms developed to automatically detect, classify, and measure asphalt pavement raveling using three-dimensional (3D) pavement data obtained from 3D line laser imaging technology (for brevity, 3D laser technology). The proposed method includes five major components: (1) data pre-processing to remove data outliers, detect pavement markings and edge drop-off, and extract the candidate pavement portion for raveling detection; (2) computation of each subsection with feature set for raveling analysis; each 3D pavement data file covers a 5-m pavement section that is divided into six subsections; (3) raveling classification using Random Forest models, a supervised leaning technique with the known raveling classification as the learning samples; (4) post-processing to aggregate the six subsection-based raveling classification outcomes for determining the raveling severity level; and (5) aggregation of the detection outcomes to measure and report the raveling condition, including percentage and severity level, at the segment level based on highway agencies’ survey practices; normally, a segment level is 1mile long. The proposed algorithms were developed and validated using the raveling severity levels defined by the pavement condition survey protocol used by the Georgia Department of Transportation (GDOT) and can be extended to other state DOTs’ protocols by re-training the classification components using corresponding ground truth data. The proposed algorithms were built based on the commonly used 3D pavement data that had already been collected (in this case, for automatic rutting and cracking data collection). Using common, already-collected data for raveling detection could save great amounts of time and money by eliminating or reducing the need for separate field data collections. The research outcome of this study has a significant impact on state transportation agencies and industry in that it can reduce time and money spent on collecting asphalt pavement raveling data.
总页数：	50
报告类型：	科技报告