1. pcl::SACSegmentation [Sample Consensus methods]

plane is found which is somehow parallel to plants

plane is found which is somehow parallel to plants

  1. StatisticalOutlierRemoval

Untitled

points[]: 74013
width: 74013
height: 1
is_dense: 1
sensor origin (xyz): [0, 0, 0] / orientation (xyzw): [0, 0, 0, 1]

Cloud after filtering: 
header: seq: 0 stamp: 0 frame_id: 

points[]: 69897
width: 69897
height: 1
is_dense: 1
sensor origin (xyz): [0, 0, 0] / orientation (xyzw): [0, 0, 0, 1]
  1. Transformation

Untitled

  1. Passthrough

Untitled

  1. Region Growing Segmentation

Untitled

  1. MomentofInertia —> Eigen vector computation

Axis of PCA is not good enough to filter x-axis

Axis of PCA is not good enough to filter x-axis

  1. x-axis filtering

Some part of plants are also filtered out.

Some part of plants are also filtered out.

  1. GreedyProjection Triangulation

Untitled

TODO: convexHull and compute the area directly. : not working...

https://pcl.readthedocs.io/projects/tutorials/en/master/hull_2d.html#hull-2d