Main Content

Segment organized 3-D range data into clusters

segments organized 3-D range data `labels`

= segmentLidarData(`ptCloud`

,`distThreshold`

)`ptCloud`

into clusters. The
function assigns an integer cluster label to each point in the point cloud and
returns the cluster label of all points in `labels`

.

The function groups two neighboring points into the same cluster if their
Euclidean distance is less than `distThreshold`

or if the angle
between the sensor and two neighboring points is at least 5 degrees.

sets the angle constraint for grouping points into the same cluster to
`labels`

= segmentLidarData(`ptCloud`

,`distThreshold`

,`angleThreshold`

)`angleThreshold`

.

`[`

also returns the number of clusters.`labels`

,`numClusters`

] = segmentLidarData(___)

`[___] = segmentLidarData(___,`

also sets the minimum and maximum number of points in each cluster, specified as a
2-element vector or as a scalar value. When you specify
`NumClusterPoints=[1,Inf]`

)`NumClusterPoints`

as a scalar, the maximum number of points in
the cluster is unrestricted. The function sets the `labels`

to
`0`

when clusters are outside of the specified range.

The `segmentLidarData`

function uses distance and angle thresholds to
cluster neighboring points. The function groups two neighboring points into the same
cluster if their Euclidean distance is less than the input
`distThreshold`

or if the angle between the sensor and
neighboring points is greater than or equal to the input
`angleThreshold`

. If you do not specify
`angleThreshold`

, the function sets this angle to
`5`

degrees.

For example, suppose `angleThreshold`

is set to
`90`

. Because angles α and β in the figure are both greater than
the specified threshold of 90 degrees, the function groups points A, B, and C into the
same cluster. Because angle σ is less than the 90-degree threshold, the function groups
point D into a separate cluster. Each angle the function uses for clustering is formed
by the line from a point to the sensor and the line from that same point to the
neighboring point

[1] Bogoslavskyi, I. “Efficient Online Segmentation for Sparse 3D Laser
Scans.” *Journal of Photogrammetry, Remote Sensing and Geoinformation
Science*. Vol. 85, Issue 1, 2017, pp. 41–52.

`pointCloud`

| `pcsegdist`

| `pcfitplane`

| `velodyneFileReader`

| `segmentGroundFromLidarData`