Exercise 1

Import file and change Scalar Field display setting. Read point clouds number. Change point size.

Import file and change Scalar Field display setting. Read point clouds number. Change point size.

Exercise 2

Point cloud colored by reflection intensity.

Point cloud colored by reflection intensity.

Filter out the high intensity part. Lots of the remaining part is low-reflective plants.

Filter out the high intensity part. Lots of the remaining part is low-reflective plants.

Point cloud colored by elevation.

Point cloud colored by elevation.

image.png

Ambient occlusion.

Ambient occlusion.

image.png

Ambient occlusion is a common rendering technique, which visualizes a 3D scene based on the level of  illumination the scene received from an ambient light source (often modeled by a hemisphere representing an open sky vault). Ambient occlusion has been found as an effective method for rendering  point cloud. Geometric details in ambient occlusion images appear clearer compared to elevation-based  visualization. More details can be found in “Visualisation of urban airborne laser scanning data with  occlusion images” by Hinks et al. (2015).

close up view point for ambient occlusion.

close up view point for ambient occlusion.

Exercise 3

image.png

curb height = 0.045 ft = 1.3 cm

curb height = 0.045 ft = 1.3 cm

Light pole height = 14 ft = 4.26 m

Light pole height = 14 ft = 4.26 m

<aside> <img src="/icons/light-bulb_gray.svg" alt="/icons/light-bulb_gray.svg" width="40px" />

Question: what exactly is the unit of these datasets?

</aside>

Exercise 4

The histogram of the left image.

The histogram of the left image.

Selected area with relative homogeneous intensity.

Selected area with relative homogeneous intensity.

Original tls

Original tls

tls part filtering for intensity between 10-20

tls part filtering for intensity between 10-20

the remaining

the remaining

Exercise 5

Approach 1 – 2D raster discretization