Skip to main content
. 2018 Nov 15;18(11):3960. doi: 10.3390/s18113960

Figure 11.

Figure 11

Comparison between region-specific and combined training datasets. Results labeled “combined dataset” are trained on images from both Witten and Manhattan. Validation set accuracy on the vertical axis is specific to the region indicated on the horizontal axis. (a) RGB (satellite) image input and (b) LiDAR image input.