For intelligent vehicles, road lane detection is a challenging task during nighttime due to low visibility of markers lanes on road. Autonomous vehicles must survive in several road environments such as sunny, foggy, rainy, shadowy, inside tunnel and during night. The major problem in road lanes is detection during nighttime. This paper presents a robust and new detection process based on the road studs tracking at nighttime. Instead of painted lanes on road, we detected road studs (Cat Eyes) in nighttime. The studs have higher intensity and more visibility during night. The proposed methods for stud’s detection are Canny Edge Detector and Hough Transform. First, we took the input image, then extracted the region of interest (ROI) from the original image and converted the image from RGB to gray scale. In the next step, the canny edge detector applied. Finally, the roads studs detected by Hough transform. Our experiment results shows that the proposed method for road studs detection is robust and accurate in night scenes.
Key Words: Road Studs, Cat Eyes, Region of Interest, Canny Edge Detection, Hough Transform, Computer Vision, Image Processing.