A new prototype system developed by the Georgia Tech Research Institute (GTRI) could greatly improve the efficiency of road maintenance, maintaining road integrity for longer durations without the need for total re-paving’s.
This prototype could greatly benefit transportation departments around the world, saving them money on road repair costs, and limiting the danger to which their workers are exposed to while using current crack sealing methods. The benefit to commuters is also immense, with less major road projects being required on the busiest streets in the middle of summer; a brilliant recipe for blaring horns and steaming road rage.
The trailer-mounted system is equipped with a two different-colored LED's, a stereo camera, and the sealant distribution systems. As the vehicle carrying the system slowly makes its way along the road (at speeds of about 3 miles/hour, you don't want to get stuck behind one of these), it takes pictures of the LED illuminated road, and instantly analyzes them to determine whether a sealing job is needed; in just 100 milliseconds, the system can create a 'crack map', highlighting the location of any cracks where work is needed. The system can detect cracks as small as one-eighth of an inch.
The operator of the vehicle, the only person required to run the prototype system, could then instruct the system when to fire from one of its 12 nozzles, which are spaced evenly at one foot intervals, so that an entire lane can be serviced during a single pass.
Despite being a prototype, the system already functions at a high capacity for accurately detecting cracks. In a test of the system's capabilities, researchers fed their crack detection algorithm 100,000 images, and found it could spot the cracks with 83% accuracy. One major problem to be solved was the system's difficulty in making out cracks located in areas with much darker or lighter backdrops.
"Our crack detection algorithm was limited because we used a vision-based system, which was confounded by regions of high contrast caused by features other than pavement cracks, including dark stains in the pavement, lane stripes, raised-pavement markers, crack sealant and debris," said Jonathan Holmes, GTRI research engineer and project lead. "A full-scale system may require a fusion of multiple imaging sensors, such as a 3-D laser scanning system."
