Vision based road surface detection for automotive systems

Advance information about the road surface a vehicle is going to encounter can improve the performance of automotive systems. For e.g. the initial slip cycles caused by the Antilock Braking Systems (ABS) could be avoided, if it is already known that the vehicle is on a surface having a low coefficient of friction (μ). In this paper, an algorithm is developed that detects different road surfaces using streaming video acquired from a camera mounted on the hood of the vehicle. The road surfaces detected here are asphalt road, cement road, sandy road, rough asphalt road (asphalt road which is deteriorating), grassy road and rough road. The value of coefficient of friction (μ) is also given out with the detected surfaces to obtain additional information about the road surfaces. Split μ (a road having different μ conditions on the left and right side of the vehicle) and μ jump (different μ conditions on the front and rear of the vehicle) are also pre detected. One method was not sufficient to achieve the goals of this algorithm. Here several simple techniques like the Canny edge algorithm, intensity histogram, contours, Hough transform and image segmentation were employed. To prevent misdetections, the road surface detection during high motion blur is prohibited. Open Source Computer Vision developed by Intel is used for the implementation of the above mentioned methods.