Volume 06,Issue 04

Mobile Mapping System-based Methodology to Perform Automated Road Safety Audits to Improve Horizontal Curve Safety on Rural Roadways

Authors

Mohammadreza Koloushani , Eren Erman Ozguven, Ali Fatemi, Masoud Tabibi

Abstract
Mobile Mapping Systems (MMS) can be employed for road safety inspection purposes to obtain Global Positioning System data and achieve accurate image data of the side features. Utilizing the MMS data, this paper develops a two-phase automated methodology to evaluate the location and numerical value of the speed limit signs before horizontal curves on two-lane rural roadways. The first phase includes the collection of the GPS data, including the horizontal curve radius, the associated speed, curve starting point, and the safe distance before the curve to install the speed limit sign and its’ value. In the second phase, the required features on speed limit signs are extracted using a Support Vector Machine (SVM)-based pattern recognition method. Findings indicate that the proposed methodology leads to 92% and 97% accuracy while identifying the speed limit signpost itself in a frame and distinguishing the speed value on the sign, respectively. In addition, horizontal curve beginning points and curve radii have been identified with a precision of 97% and 90%, respectively. The proposed methodology has the potential to help attain the needed safety standards related to the location and value on the speed limit signs before the horizontal curves in a low-cost manner.

Keyword: Global Positioning System, Image processing, Mobile Mapping System, Road safety audit, Support vector machine.

PDF [ 677.29 Kb ] | Endnote File | XML