Estimating Pavement Roughness by Fusing Color and Depth Data Obtained from an Inexpensive RGB-D Sensor.

摘要: Measuring pavement roughness and detecting surface defects are two of the most important tasks in management. While existing measurement approaches expensive, primary aim this paper is to use a cost-effective sufficiently accurate RGB-D sensor estimate outdoor environment. An algorithm proposed process data autonomously quantify road roughness. To end, calibrated for estimating collected. The collected depth frames RGB images registered create 3D surfaces. We found that there significant correlation between estimated International Roughness Index (IRI) using manual measured IRI rod level. By considering Power Spectral Density (PSD) analysis repeatability measurement, results show solution can accurately different