Semi-automatic inspection tool of pavement condition from three-dimensional profile scans

Abstract

In the present work the preliminary study on qualitative and quantitative assessment of the road surface condition is presented. The work is motivated by a very actual and important problem of visual inspection of the road network. In general the road managers are obliged to perform a regular, periodic maintenance road testing, also the road constructors and civil engineers in the recent design framework – ‘design-build-maintain’ are interested in seasonal inspections of their constructions. Therefore the semi- or full-automatic visual testing methods of the road surface condition are under a constant development and in the center of interest of many research groups. Herein the main attention is focus on the automatic quantification of a road surface damage type from the three-dimensional cloud points. The stereo vision system, its intrinsic and extrinsic parameters as well as the correspondence-reconstruction problems are here assumed theoretically, since the practical design of the full methodology is still under the development in the ongoing national project.