Lane Crack Detection Based on Saliency

Lane cracks are one of the biggest threats to pavement conditions. The automatic detection lane can not only assist evaluation road quality and quantity but also be used develop best crack repair plan, so as keep level ensure driving safety. Although extracted from images because gray intensity pixels is lower than background intensity, it still a challenge extract continuous complete three-lane with complex texture, high noise, uneven illumination. Different threshold segmentation edge detection, this study designed algorithm dual positioning. An image-enhancement method based on saliency proposed for first time. Based Bayesian probability, each pixel judged calculated. Then, Fréchet distance improvement triangle relationship introduced determine whether key point fracture endpoint fast-moving should terminated. In addition, remote-sensing process was developed calculate length width by inverting squint collected mobile phones. A large number different types, illumination, interference conditions were tested. average extraction accuracy 89.3%, recall rate 87.1%, F1 value 88.2% showed that could detect in well.