A real-time detection approach to pavement rutting based on line laser in nature condition

A real-time laser path detection method of pavement rutting line was studied in this paper. In traffic running condition of normal driving mode, the lasers were used to continuously emit line laser to the pavement. A sequence of pavement rutting laser line images were acquired dynamically by a high-resolution camera and pavement rutting were detected in moving situation in real time. Aiming at the problem of difficult to extract the rutting laser line under the strong interference of the pavement complex background, the characteristics of the rutting laser image of the pavement were analyzed. The definition of non-negative feature measure and the formula of non-negative intensity were given. A method of extracting rutting laser line based on non-negative feature and peak continuity was proposed. Rather than using conventional methods of image processing, the significant and non-significant ridges of the rutting laser line were tracked rapidly using peak continuity to achieve fast and accurate extraction of the rutting laser line with complex background. A large number of verifying tests for different pavements were carried out under conditions of driving, opening traffic, and without blocking other transportation. The experimental results verify the effectiveness and accuracy of the proposed method and solve the problem of fast and accurate extraction of the rutting laser line under natural running conditions. It provides a key technical support for the automatic on-line detection of the pavement rutting and quality evaluation of road surface.