Share this post on:

Is can also be within the expected error of hypsometer-based height measurements someter-based height measurements [66].[66].Figure 12. Left shows a scatter plot with the automatically measured tree height GNE-371 Cell Cycle/DNA Damage compared together with the the reference tree height. 12. Left shows a scatter plot from the automatically measured tree height compared with reference tree height. The Figure 12. Left shows a scatter plot on the automatically measured tree height compared using the reference tree height. The grey shows what a perfect outcome would be (ybe (y = x). Ideal shows distribution on the height errors. The line line shows what a perfect outcome would = (y Correct shows the the distribution from the height errors. greygrey line shows what an ideal result could be x). = x). Proper shows the distribution with the height errors.Remote Sens. 2021, 13, 4677 Remote Sens. 2021, 13, x FOR PEER REVIEW19 of 31 18 of3.three. All Stem Diameter Measurements 3.three. All Stem Diameter Measurements Out of a total of 7022 reference diameter measurements, 5141 were matched to auOut of a total of 7022 reference diameter measurements, 5141 have been matched to autotomatically measured TLS point cloud-based measurements. The imply, median and matically measured TLS point cloud-based measurements. The mean, median and rootroot-mean-squared (RMS/RMSE) diameter errors had been 0.032 m, 0.02 m, and 0.103 m, mean-squared (RMS/RMSE) diameter errors respectively. This can be visualised in Figure 13. were 0.032 m, 0.02 m, and 0.103 m, respectively. That is visualised in Figure 13.Figure 13. Left shows a scatter plot of reference diameters against the remotely sensed and automatically IQP-0528 custom synthesis extracted diamFigure 13. Left shows a scatter plot of reference diameters against the remotely sensed and automatically extracted diameter eter measurements. This does include the measurements which had been not successfully matched involving reference and measurements. This does notnot involve the measurements which had been not effectively matchedbetween reference and automatic datasets. Right offers a histogram to visualise the distribution of the measurement errors. automatic datasets. Appropriate delivers a histogram to visualise the distribution with the measurement errors.three.four. Tree and Measurement Detection Completeness 3.4. Tree and Measurement Detection Completeness The measurement completeness (the fractions of measurements detected out on the total The measurement completeness (the fractions of in Figure 14. On the left,out in the number of reference measurements) are shown measurements detected the comtotal quantity of reference measurements)shown, and around the ideal, the completeness of all pleteness for reference tree detection is are shown in Figure 14. Around the left, the completeness for reference tree detection is shown,tree detection completeness for 32of all diameter diameter measurements is shown. The and on the proper, the completeness plots was bemeasurements is shown. 1 plot (Plot 38) failing to detectfor 32trees. Out of 49 plots, 30 tween 0.95 and 1.0, with the tree detection completeness any plots was amongst 0.95 and 1.0,trees detected. (Plotmean and median completeness values for tree30 had all trees had all with a single plot The 38) failing to detect any trees. Out of 49 plots, detection were detected.1. imply and median completeness values for tree detection have been 0.91 and 1. 0.91 and the The measurement detection completeness was anticipated toto be much less than the tree demeasurement detection completeness was anticipated be much less than the tre.

Share this post on:

Author: haoyuan2014