AN AUTOMATED METHOD FOR STREET FLOOR DETECTION USING TERRESTRIAL LIDAR POINT CLOUD

Abstract

In recent years, the rapid increase in the demand for street floor information has drawn remarkable interest to the use of laser scanner
data. It provides highly dense point cloud data with three dimensional (3D) position, intensity and range from the sensor to target. The street
floor detection in urban areas is difficult task due to the complicated patterns and many contextual objects. In present study an automated method
for detection of street floor using the Terrestrial Laser Scanner (TLS) point cloud dataset has been proposed. Proposed method includes ground
point filtering, rough street floor classification, edge detection and point in polygon test, in order to detect the street floor. Proposed method has
been tested at a captured TLS point cloud. Completeness and correctness of the proposed method are 95.14% and 97.42% respectively.
Automatically detecting a highly detailed street floor helps in maintaining the pavement by estimating the road surface conditions.