Application of Artificial Intelligence for Speed-Limit-Signs Recognition and Location before Horizontal Curves

Traffic signs are essential for road-safety and must be inspected and audited regularly in order to extend their service life and to maintain sustainable safety. In many countries, such audits are carried out traditionally by using manpower. This may result in inaccuracies due to human error in addition to being costly and time-consuming. Therefore, aiming to save time and resources and to enhance the accuracy, it was attempted to present a novel method for detecting speed-limit signs, recognizing the numbers on them, and calculating the distance until the points of curves of radii less than ۳۳۰ m.First, the horizontal curve is characterized using GPS and the speed-limit is then calculated. Moreover, all traffic signs are detected, then, speed limit signs are identified and classified using image processing, and the speed limit indicated by the sign is recognized. The safety of the horizontal curve is evaluated by comparing the numbers on the speed-limit signs with the speed calculated based on the curve properties. Next, the location of the speed-limit signs and their distance from the horizontal curve is calculated based on the difference in the speed on the straight and curved parts of the path and then, compared with regulatory measures.

کلیدواژه ها:

Artificial Intelligence ، Intelligent Transport Systems (ITS) ، Image Processing ، Speed Limit Signs ، Safety of Horizontal Curves.