The influence of the incident angles variation on the accuracy of TLS point cloud based on surface reflectivity and roughness

Abstract

The accuracy of measuring TLS point clouds depends on the measuring travel time, range detector, and surface properties. TLS conducted interesting experimental tests to examine the effect of different scanned materials in terms of reflectivity and surface roughness on the quality of the measured TLS data at different incident angles. Different types of materials specifically (wood, glass, steel, Ekoplast, and adhesive total station target) were selected for this purpose. During the experiments, 24 scans were measured as the selected four materials scanned at six incident angles (0˚, 15˚, 30˚, 45˚, 60˚, and 75˚) while the range was fixed to about 5m. The experiment's results reveal that smooth surfaces are more highly affected by the accuracy of the measured 3D point clouds than rough surfaces at different incident angles. At 0˚ incident angle, the total station target reflects about 20cm closer to the scanner than the other materials; this difference decreases with increasing incident angle. At 75˚ incidence angle, the difference decreases to become approximately 2mm in comparison to the other materials. The maximum RMSE of rough materials is less than 1cm except for wood material at 30˚, while the Maximum RMSE for smooth materials reaches 4cm in 45˚ glass material. Furthermore, different materials have different intensities, both smooth materials, glass and steel, have different levels of accuracy due to different properties. The minimum RMSE in glass is 1.47cm, and the maximum RMSE for steel reaches 1.17cm.