FAYETTEVILLE, Ark. -The $250 billion Transportation Reauthorization legislation currently before Congress will provide only a fraction of the costs to maintain the 4 million miles of highways in the United States. But University of Arkansas researcher Kelvin Wang has developed a system that will make this vital activity fast, cheap and safe for highway workers and motorists.

"The most widely used method to survey surface distress of highway pavements - human observation - is extremely labor-intensive, error-prone, and hazardous," explained Wang, professor of civil engineering. "An ideal automated system should find all types of cracking and any other surface distress of any size, at any collection speed and under any weather conditions."

Wang led a team that developed such a system - the Digital Highway Data Vehicle (DHDV). Fully automated, the DHDV can collect and process images and other data at resolutions down to 1 millimeter at highway speed, day or night and under most weather conditions. It debuted this month at the 2003 Highway Profilers meeting in Austin, Texas.

The key features of the new DHDV compared to our previous generation: 1. Day or night operation at highway speed 2. Custom line light that is brighter than sun light at August noon time. We designed and fabricated the lights ourselves. 3. 4,096 pixels per 4 meter, or 1-millimeter resolution 4. Pavement surface can be completely covered with images at 1-mm resolution for thousands of miles. Again, the acquisition is at highway speed. 5. Much higher accuracy in crack processing

Traditionally, distress data for highway maintenance has been acquired by hand - transportation department employees in many states walk over every mile of road annually or biannually. They either rate the pavements while walking or record the pavement images and manually rate the images in office. The system developed by Wang and his staff can automatically identify and classify pavement surface cracks while the images of pavements are acquired. Wang’s digital solution makes this information readily available to the engineers who must make critical maintenance and rehabilitation decisions.

"Traditional systems are limited by accessibility, search capability of the image library, and synchronizing video data with traditional engineering data," explained Wang. "More importantly, there are situations in which multiple users need to examine the video footage at the same time."

DHDV uses digital cameras to record highway surfaces in real time and store the information directly on computer drives. The images are processed and added to critical maintenance information such as when the highway was built, what materials were used, when maintenance was performed and traffic patterns. These data are then readily available to the engineers who must make critical maintenance decisions.

With Wang’s system, high-definition images of pavement surfaces are directly captured, archived and analyzed by computers. By applying image-processing techniques to these digital images, engineers can easily identify distresses in the roadway and determine the exact scope of the problem. Distress objects such as cracks can then be classified according to length, width, orientation or other predefined categories.

"The DHDV can look at the entire width of a lane at highway speed and at 1 millimeter resolution without sub-sampling," said Wang. He believes the new DHDV breaks the record for its data collection and processing capabilities in terms of complete coverage of pavement, resolution and automated processing for cracks.  

DHDV uses custom-fabricated lights that are brighter than sunlight and eliminate the need for strobe lighting or shuttering. They also allow data acquisition at night as well as in the day, which is not easy with other available methods.

"To the best of my knowledge, DHDV is the only system in the world today that has 1 millimeter resolution, complete lane coverage and automation in cracking survey," Wang said.