Rehabilitation and maintenance of pavement are well established and costly operations. One of the major challenges of these operations is determining the optimum maintenance and repair schedule. In any given city, there are hundreds of miles of pavement constructed at different times and are in different states of deterioration. It is the job of municipal engineers to establish a cost effective schedule that prioritizes the repair or even the reconstruction of different segments of the roads. The schedule is usually based on an overall Pavement Condition Index (PCI). Typically, inspectors are sent out to observe the pavement conditions and conduct accurate measurements to be used in computing the PCI; however the cost associated with such inspection missions is often high.
This thesis proposes a novel approach to estimate a cost effective Pavement Indicator (PI) for the entire city (or any area of interest). The proposed approach exploits newly available miniature cameras, GPS, wireless networking and Digital Signal Processing to automatically and continually collect visual information about different segments of the road, and combines these images to establish a live map of the city roads where different colours correspond to an approximate estimate of a pavement indicator (PI). The proposed technique is not a replacement of the traditional inspection, but rather it is a tool to identify the sections that are in greater need of repair. The technique involves taking pictures of various sections of the road network using cameras mounted on public vehicles and transmitting these images to a processing centre. Each image is processed using image filtering techniques to produce an initial estimate of PI. The cumulative effect of these estimates produces regional estimates that become more statistically accurate as time goes by, and the overall PI map is continually updated to maintain a global visual map of PI. The proposed pavement indicator (PI) map can then be linked to an optimization software package to determine the most cost effective road rehabilitation schedule.