The condition of highway structures in the UK is determined by visual inspection. There are five main types of inspection which are undertaken at different frequencies. These inspections cover a range of detail, from a cursory check for obvious defects, through to a close examination of particular areas or defects causing concern. The quality of data collected depends on the ability of inspectors to observe and accurately record details on visible defects. This could be affected by many factors, such as the environmental conditions, and the knowledge and experience of the inspectors. Possibly for these reasons, it has been found that the data provided by such inspections can vary significantly. Improvements to the quality of the inspections are therefore desirable.
This research has been carried out with the objective of developing a more objective and repeatable inspection procedure than the current system of manual inspections. The improved inspection procedure is based on the proposal that images of structures could be collected and processed (manually or automatically), to identify defects. The research has concentrated on two key areas - image collection and display, and image analysis.
The research discussed in this report is from the fourth stage of an ongoing process, with another phase of development to come in 2009-10. The results presented herein are therefore based on the current performance of the system, and may improve following further development.

Want to know more about this project?