How are computers used in pavement management?

Question

What will be an ideal response?

Answer 1

Computer vision is an area of artificial intelligence (AI) research that
involves the use of sensors and computers to emulate human vision. As applied to
pavement management, the objective is to develop an automated approach capable
of identifying the type, severity, and extent of pavement distresses. The process
generally involves the following four basic steps: image acquisition, image
digitization, image processing, and pattern recognition.
For pavement management, the image is usually captured using a video
camera. Digitization then creates a numerical representation of the image (digitized
image) suitable for computer processing in the form of an array of numbers where
the integer value of each element in the array represents the color or the gray tone
of the corresponding area in the original image. Values range from 0, representing
black, to 255, representing white. The cracked region will have low values due to
crack shadows being much darker than the surrounding background.
The image processing step seeks to extract the different distresses from the
background. The image is first transformed into what is known as a “binary
image”, where the distresses are indicated by black pixels and the background by
white. Algorithms are also applied to clean up the image and remove the “noise.”
Counting the number of black pixels will thus yield an idea of the distressed area.
Finally, the pattern recognition step classifies the distress in a pavement surface
image as a particular type of distress.