Point Cloud Fetch

In this guide we establish a secure connection to a Qb2 device. Afterwards, the point cloud data is requested and one or multiple point cloud frames are received. Finally, we get data from the received point cloud and plot a single point cloud frame.

Please follow the API Guide first to install the Python package.

Fetch point cloud from Qb2

We use the PointCloud service from blickfeld_qb2.core_processing.services namespace over the gRPC channel and issue a single get (retrieve one point cloud frame) or streaming request (continuously receive point cloud frames) to the device. The code snippets below show examples of connecting to a Qb2 with hostname qb2-xxxxxxxxx and fetching point cloud frames from it.

Get single frame

The following example shows how to fetch a single point cloud frame using the Get method.

import blickfeld_qb2
from blickfeld_qb2.base.grpc.channel import Channel

# open secure connection to Qb2
with Channel(fqdn_or_ip="qb2-xxxxxxxxx") as channel:
    service = blickfeld_qb2.core_processing.services.PointCloud(channel)
    frame = service.get().frame
    # print frame id
    print("received frame with id", frame.id)

Continuously stream point cloud frames

The following example shows how to continuously stream point cloud frames from a Qb2 device:

import blickfeld_qb2
from blickfeld_qb2.base.grpc.channel import Channel

# open secure connection to Qb2
with Channel(fqdn_or_ip="qb2-xxxxxxxxx") as channel:
    service = blickfeld_qb2.core_processing.services.PointCloud(channel)
    # request point cloud stream from Qb2
    for i, response in enumerate(service.stream()):
        # extract point cloud frame from response
        frame = response.frame
        # print frame id
        print("received frame with id", frame.id)
        # get 3 frames and stop
        if i == 2:
            break

Get data from point cloud frame

In the code snippet below we read out the cartesian coordinates xyz for each point in the received point cloud frame. We further use these values to calculate azimuth and elevation vectors.

import numpy as np

# get x,y,z coordinates from the frame points
xyz = frame.binary.cartesian

# calculate azimuth and elevation
distance = np.sqrt(xyz[:,0] ** 2 + xyz[:,1] ** 2 + xyz[:,2] ** 2)
azimuth = np.arctan(xyz[:,0] / xyz[:,1])
elevation = np.arcsin(xyz[:,2] / distance)

Plot point cloud frame

Finally, azimuth, elevation and photon_count vectors are used to plot the desired field of view from the point cloud frame as shown in the code snippet below.

import matplotlib.pyplot as plt

index = np.argwhere((azimuth > np.deg2rad(20)) & (azimuth < np.deg2rad(30)) & (elevation > np.deg2rad(5)))
plt.scatter(azimuth[index], elevation[index], c=frame.binary.photon_count[index], s=0.1)
plt.xlabel("Azimuth [rad]")
plt.ylabel("Elevation [rad]")
plt.show()

The plot result is shown in the Figure below:

Figure 1. Point Cloud Frame