Mobile LiDAR for highway

Listed in Datasets

By Yi Lin1, Yi-ting Cheng1, Ayman Habib1

Purdue University

Highway data collected by mobile LiDAR, including point cloud, images, and trajectory.

Additional materials available

Version 1.0 - published on 02 Dec 2019 doi:10.4231/D5QY-8C29 - cite this Content may change until committed to the archive on 02 Jan 2020

Licensed under Attribution 3.0 Unported

lane-width-1.PNG lane-width-2.PNG transit.png


Lane width evaluation is one of the crucial aspects in road safety inspection, especially in work zones where a narrow lane width can result in a reduced roadway capacity and also, increase the probability of severe accidents. Using mobile mapping systems (MMS) equipped with laser scanners is a safe and cost-effective method for rapidly collecting detailed information along road surface. This study presents an approach to derive lane width estimates using point clouds acquired from a geometrically-calibrated mobile mapping system. This dataset comprises the first and second road segments along an Interstate Highway (North-South) in the United States, with a total length of approximately 36 miles and it is located at an interstate highway work zone area. In this study, the road surface along the driving lane was extracted with the assistance of vehicle trajectory data and a height buffer determined on the basis of the standard pavement cross slope for U.S. highways. Next, the lane markings were extracted using an intensity-based thresholding followed by an identification of ambiguous lane markings. Then, an algorithm was developed for determining the lane direction without using trajectory data and the derived direction was used to compute the lane width using the equally spaced centerline points. Finally, the areas identified to consist of ambiguous lane markings, narrow lanes, or wide lanes were reported as polygons marked by the starting and ending point pair in order to visualize those on the 3D data as well as their back-projection onto RGB imagery.

Cite this work

Researchers should cite this work as follows:



The data was collected on May 24, 2018.

The Purdue University Research Repository (PURR) is a university core research facility provided by the Purdue University Libraries, the Office of the Executive Vice President for Research and Partnerships, and Information Technology at Purdue (ITaP).