Mohamed Saleh tests a system that measures the texture of pavemen.
The texture of the roads we drive is a factor in the noise, comfort, and safety travelers experience. Measuring the pavement texture, however, entails interrupting traffic and the need for traffic control measures — while providing only a limited sample of the road.
Ph.D. student Mohamed Saleh is developing a stereo vision system that can provide continual road texture measurements at highway speeds over an entire area and will not interrupt traffic.
Saleh’s system uses two cameras placed side by side capturing the same image, yielding a slightly different views of a 3-D scene. The images are processed, aligning the epipolar lines and performing edge detection. A matching algorithm identifies the candidate correspondence and produces a 3-D image, allowing depth measures of the area to be examined.
Computer representation of the cracked texture
“The benefit of an area measurement, instead of a thin longitudinal line, will give us better characterization of the road surface,” says Saleh, who is working with advisor Lynn Abbott. “This is needed to better determine most tire-pavement interactions, such as friction, noise, splash and spray, rolling resistance, and tire wear.”