As part of the development and implementation programme to support the introduction of SCANNER surveys on all local roads in England, this project investigated the capabilities of automated systems to detect visible defects other than cracking on all types and classes of local road, from principal roads to minor unclassified roads.
A number of visible defects are considered important in the evaluation of maintenance need by local highway engineers (for example, fatting up, fretting, ravelling, crazing and potholes). Cracking is the only visible defect that is monitored by the current SCANNER surveys. Visual Inspection (VI) surveys include other forms of surface deterioration, described as 'wearing course deterioration', 'surface deterioration' and 'defective surface dressing'. This research investigated whether it is possible to measure these other defects directly and, if not, whether the current systems could provide information that could act as proxies for the presence of these defects.
This research was carried out by Scott Wilson Pavement Engineering and the University of Nottingham. The final project report is here: Other visible defects
The report concludes that, despite the difficulty of developing an image analysis system which can replicate the complex visual processing of human surveying, this project demonstrated the feasibility of building systems to record and report the presence of other visible defects, using reasonably inexpensive technologies "off the shelf".
Working with experienced road engineers, the project developed a detailed classification system for the defects visible on the road surface. Following a review of available technologies, image analysis was identified as the most promising method for further development, possibly linked to surface texture and metal detection (to eliminate road ironwork). The project identified the need for very carefully controlled illumination levels and developed outline "rules" for the detection of individual defects.
However, at the current stage of development, it was only possible to recommend a simple condition number " visible defect intensity" to report the percentage area of the carriageway affected by visible defects, and a specification was developed.
Further research would be required, with controlled illumination, to investigate the possibilities for discriminating between different types of visible defects. Further research would also be required to investigate the possibility of combining information about the three dimensional surface shape and the localised variation of surface texture with information from image analysis, to be able to identify specific types of surface deterioration on a variety of road surfaces.