Automated Segmentation of Pavement Measurements Based on Bayesian Ideas : Experiences from Austria
Fridtjof Thomas, Alfred Weninger‐Vycudil, P Simanek · 2004
Road surface condition measurements are in Austria periodically obtained for all the federal roads. These measurements refer to 50m long road sections and form the basis for an evaluation of the current state of the road network as well as the subsequent analysis of maintenance strategies and their funding requirements. In order to systematically utilize the information contained in these measurement series, homogeneous sections have to be identified for longitudinal and transversal unevenness as well as skid resistance. For that purpose a recently published segmentation algorithm has been implemented as a standalone computer program that is capable of processing huge amounts of data in batch mode. This software takes a measurement series as input and identifies the borders of homogeneous sections by evaluating the statistical properties of the measurement series under study. A variety of quantities representing the homogeneous sections is calculated and can be exported in widely understood file formats for further processing. (a) For the covering entry of this conference, please see ITRD abstract no. E212095.