Range Error Detection and Evaluation for retroreflective Road Signs in Phase-Shift MMS Point Clouds

Abstract. In this study, we addressed the problem of range errors in point clouds of road signs acquired using a mobile mapping system (MMS) equipped with a phase-shift laser scanner. Under certain conditions, retroreflective materials used for road signs cause points to be output at locations different from their true positions. Although previous studies have developed techniques to reduce range errors of time-of-flight scanners, similar research on phase-shift scanners is scarce. Furthermore, many existing approaches for detecting road signs from point clouds assume that sufficient points on the sign surface are captured. Therefore, they are inapplicable when range errors affect numerous sign points. Against this background, first, we located road signs by converting the point clouds into images and identifying the signs in the image domain. We then developed a method to determine the presence of range errors based on the standard deviation of the relative distances from a reference emission point. To validate the developed method, we used 5 km of driving data collected from general roads. Consequently, 32 road signs were extracted, of which 26 were detected as exhibiting range errors in agreement with visual assessment. These findings show the effectiveness of the developed detection method and highlight its potential for improving the reliability of identifying range error-affected road signs on general roads.