Automated assessment of delineation and signage systems
Research has indicated that about 4% of road accidents can be attributed to poorly maintained road markings and signs. The current inspection method for road signs and lines is based on a manual inspection by road engineers. This can be very subjective, time-consuming and expensive, and has associated safety risks. This research investigated an assessment method that is inherently more objective and carried out more safely from the inspector point of view.
The objective of this project is to develop vision-based systems that can be used to evaluate road signing and lining that has become defective due to damage and/or wear and tear. Such a system could enable the automatic detection of defects and could provide objective performance parameters for these important road safety features. The system has the potential for reducing risk to contractors working on the road as well as increasing the safety of road users. It would also reduce the costs associated with maintaining the highway by eliminating time-consuming manual assessment.
The outcomes from the research supported by this funding have been significant. In conjunction with the NRA, a vehicular mobile mapping system has been developed to address specific problems in this area. The system is composed of a number of navigation sensors for geo-referencing (positioning) and a multi-camera (stereo) system for automated feature extraction from imagery. As the research agenda of the team developed, it became clear that there was a commercial opportunity in this area. This was pursued in 2010 with support from an Enterprise Ireland Commercialisation Fund award. The commercialisation of the research is expected to develop through licensing the patents and relevant intellectual property to a start-up company. Innovation Partnership projects will also be scoped out with this start-up venture. The research group is now investigating other industry sectors in which the technology can be applied.