Page 649 - NGTU_paper_withoutVideo
P. 649
Modern Geomatics Technologies and Applications
(z + z + ⋯ z ) (1)
1
2
i
μ =
k
(2)
2
2
[(z − μ) + (z − μ) + ⋯ + (z − μ) ]
2
2
i
1
σ = √( )
k
Non − Noisy μ − α × σ ≤ z ≤ μ + α × σ (3)
NR = { }
Noisy Otherwise
Noise Removal algorithm is summarized below:
Figure 5. Noise Removal algorithm
2.1.3 Ground Removing: After removing the noise, the area along X and Y axis are divided into a M × M grid. The
minimum altitude inside each cell is identified and then all the points of the cells are moved along the Z axis, where the
minimum altitude will be zero. Finally, a predefined threshold is considered to filter low-height objects.
Figure 6. Method of removing ground points using gridding
4
