- The use of Mobile Laser Scanning for collecting highway conditions such as pavement markings is increasing.
- Since pavement markings are made intentionally reflective their return intensity can be used to auto-extract them from the point cloud.
- Extracting them can be a challenge given the accuracies of current MLS systems (cm-level), relative to the thickness of the markings (mm-level).
- A recently published paper, “Efficient and Robust Lane Marking Extraction From Mobile Lidar Point Clouds,” by Jung, et al provides an important methodology for automating the process.
Surveys of roadways with Mobile Laser Scanning (MLS) are now being conducted on a regular basis by many transportation agencies to provide detailed geometric information to support a wide range of applications, including asset management.
Most MLS systems provide intensity (return signal strength) data as a point attribute in georeferenced point clouds, which may be used to estimate retro-reflectivity of pavement markings for effective maintenance. Nevertheless, the extraction of pavement markings from mobile lidar data remains an open challenge, due to variable noise, degree of wear on the markings, and road conditions.
Specifically, the primary objectives of this paper are to:
(1) Develop an efficient framework to automatically extract lane markings from 3D mobile lidar data that can handle a wide range of road geometries,
(2) Propose a constrained RANSAC segmentation that extracts the road surface in the absence of curb structure and can account for road curvature and grade changes,
(3) Propose a line association process to consider poorly-worn lane markings such that the data can help agencies assess the condition of each individual stripe,
(4) Present a filtering method for noise using the Dip test statistics of the intensity distribution, and
(5) Evaluate the proposed approach under a variety of road conditions, including urban, rural, expressway, and a specifically-designed test site.
This paper addresses these challenges, presenting a novel approach for efficient, reliable extraction of lane markings, including those that have been significantly worn.
For the complete paper click here. Note – it is free for a short time only.
