Abstract
Zebra crossings provide guidance and warning to pedestrians and drivers, thereby playing an important role in traffic safety management. Most previous studies have focused on detecting zebra stripes but have not provided full information about the areas, which is critical to both driver assistance systems and guide systems for blind individuals. This paper presents a stepwise procedure for recognizing and reconstructing zebra crossings using mobile laser scanning data. First, we propose adaptive thresholding based on road surface partitioning to reduce the impact of intensity unevenness and improve the accuracy of road marking extraction. Then, dispersion degree filtering is used to reduce the noise. Finally, zebra stripes are recognized according to the rectangular feature and fixed size, which is followed by area reconstruction according to arrangement patterns. We test our method on three datasets captured by an Optech Lynx mobile mapping system. The total recognition rate of 90.91% demonstrates the effectiveness of the method.
Highlights
Road markings, as critical transportation infrastructure, provide drivers and pedestrians with information about traffic regulations, warnings, and guidance [1]
mobile laser scanning (MLS) systems, which integrate laser scanners, global positioning system (GPS), inertial navigation system (INS), and charge-coupled device (CCD) cameras [16], collect information, such as 3D geospatial, texture, and laser intensity data, from complex urban areas when a vehicle is on the move
We have proposed an effective method for recognizing and reconstructing zebra crossings using mobile laser scanning data
Summary
As critical transportation infrastructure, provide drivers and pedestrians with information about traffic regulations, warnings, and guidance [1]. Yu et al [35] distinguished zebra crossings from other rectangular-shaped markings according to the geometric perpendicularity of their distribution directions and road centrelines These studies mostly focused on detecting stripes and did not provide specific information about the areas. The contributions of this paper are as follows: (1) an adaptive thresholding method based on road surface partitioning was designed to compensate for non-uniformities in intensity data and extract all types of road markings; (2) a dispersion degree filtering method was applied to reduce the noise; and (3) zebra crossings are recognized and reconstructed according to geometrical features, so that we obtain more specific information about the area, including start positions, end positions, distribution directions of zebra crossings, and road centreline directions.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
