WIT Press

Image-based identification and GIS-integration of vehicle restraint systems and evaluation of safety effects


Free (open access)


Volume 9 (2019), Issue 4



Page Range

344 - 355

Paper DOI



WIT Press


Christian Stefan, Csaba Beleznai, Isabela Erdelean & Matthias Hahn


Collisions between vehicles leaving the road and unforgiving roadside objects such as trees, poles, road signs, etc. constitute a major road safety issue. On the Austrian road network, approximately 7.500 injury crashes occur every year due to run-off-road (ROR) manoeuvres (i.e. 20% of all injury crashes on public roads), contributing 35% to fatalities and 25% to serious injuries. Vehicle restraint systems (VRS)  such as guardrails, concrete barriers, terminals or crash attenuators play a decisive role in mitigating the consequences of ROR crashes. unfortunately, most national road administrations (NRA) do not have a centralized data management, while geo-referenced information on VRS and their safety-related attributes are also not available as digitized data. researchers from the AIT have developed a novel approach to investigate, classify and evaluate VRS by means of image data processing, towards providing a comprehensive VRS inventory. The information obtained can be used for benefit-cost-analyses, road safety inspections and the evaluation of the effectiveness of different vehicle restraint systems.


Asset management, database, decision tree, geographical information system, inventory control, run-off-road crashes, traffic safety, vehicle restraint system.