论文部分内容阅读
针对当前探地雷达(GPR)数据解释主要依赖专家经验存在的解释结果主观性强和数据解释周期长等问题,利用高速公路路基病害将导致其厚度和层界面反射信号的幅度发生改变等客观信息,结合探地雷达杂波抑制、层界面检测和平滑、感兴趣区域(ROI)提取、特征提取和模式识别技术,提出了一种新颖的高速公路路基病害自动检测算法,并利用该算法对江西省昌九高速公路南昌段采集的GPR数据进行了分析。研究结果表明:该算法的检测结果与结合专家经验和钻孔取芯样本构建的Ground Truth数据库的吻合度高达92.7%,且具有自动、快速等优越性,可为指导制定合适的养护策略及合理分配养护资金提供科学依据。
In view of the fact that GPR data interpretation mainly relies on expert experience, such as the subjectivity of interpretation and the long data interpretation period, the use of expressway subgrade diseases will lead to the objective information such as the change of the thickness and the reflection of the layer interface , A novel algorithm for automatic detection of subgrade defects of expressway is presented in combination with clutter suppression, layer interface detection and smoothing, ROI (region of interest) extraction, feature extraction and pattern recognition. GPR data collected from Nanchang Changjiu Expressway was analyzed. The results show that the proposed algorithm is 92.7% identical to the Ground Truth database based on expert experience and drilling coring samples, and has the advantages of being automatic and rapid, which can guide the formulation of appropriate conservation strategies and reasonable Allocate conservation funds to provide a scientific basis.