Calibration of a Terrestrial Laser Scanner - Institute of Geodesy and ...

tions of the manufacturer, it can be concluded that the specifications track definedbetween areforthen being laser been was linepoint. center.cannot met, similar^90%cf. This scanner with since (white) tonominal they inves¬ of and di¬ —-60% B.rectly. have 3.2.2 target. (grey) In /-^~NThese Section\A.20% Mode 'static (dark 3.1.5,mode'. spheres AVi4Ï/\x^1A1/\t\/\s targets were \f/x iv*^ \ll used \> \**St\Ahu\h\hyV«-1 K*1*, y? had /yVderiving 'scanningcalibrationperformedcompared horizontalwell-suitedacquired Appendix importancedistancesawell-definedScanninginvestigationintroducedproperty adefinitely3.2 Distance Measurement System 35AccuracyThe accuracy of the distance measurement system based on different reflectivity values for the targets hasalready been analyzed. The question is is there a systematic deviation of the distance measurement systemfor a specific reflectivity?The results for specific reflectivity values are shown in Figure 3.12.The curves are similar to the resultsobtained by the investigation for repeatability. The conclusion is that the reflectivitydoes not influence themean value of the distance measurement system. Furthermore, the accuracy meets the requirement of themanufacturer for covering the full measuring range of 50 m. However, the correspondingmean values donot mean that the noise of the measurements, i.e.the precision,is also similar.10distance accuracy1"E5ou-5'"0 5 10 15 20 25 30 35 40 45 50range [m]Figure 3.12: Accuracy of the distance measurement system based on the 'static mode'. The targets used have differentvalues of reflectivity, namely 90% (white), 60% (grey) and 20% (dark grey).PrecisionThe precision defines the noise of a sensor. In terms of the laser scanner, the precisionsurement system gives an estimation for the empirical standard deviation of a singleof the distance mea¬distance measurement.The precision is a reliable parameter since it defines the imprecision of a single shot that occurred during ascanning process and helps to assess each single point in the point cloud.Figure 3.13 shows the standard deviation over the measuring range of 50 m for the three targets with dif¬ferent reflectivity values. It can be seen that up to a range of 10 m, the curves are close to each other with aprecision of less than 2 mm. As the lines are diverge, a difference between the reflectivity values is clearlyvisible. Since the standard deviation represents 1 a noise, the area covering each measurement, i.e. 99.9 %,has to be multiplied by a factor of three. The increasing noise is correlated to the decreasing intensityin thereflected laser beam and defines the signal-to-noise ratio (SNR). Comparing these results to the specifica¬