AI recognition of inventories - PIX4Dinspect


The Artificial Intelligence AI inventory recognition algorithm is able to detect inventory patterns in the reconstructed 3D Model and compute their position, orientation, and dimension.

Important: Good quality acquisition, focus, exposure, resolution, large enough inventories to be clearly identifiable are strong requirements for the success of the algorithm. Under these conditions, a 3D Model with enough density is expected to be available.
Note: Panel antennas, microwave antennas, and ancillary boxes (TMAs, RRUs, RRHs, splitter, etc)  are recognized by the AI algorithm.

The outputs of the algorithm are the position, orientation, and dimension of the inventory.

In particular, the inventory's properties that are automatically detected are the following:

  • Type: Type of inventory.
  • Downtilt: Downward angle with respect to the gravity field lines. Values range from -90 to +90 degrees.
  • Azimuth: Yaw angle with respect to the true North. Values range from 0 to 360 degrees.
  • Plumb: Roll angle (side tilt) with respect to the gravity field lines. Values range from -180 to +180 degrees.
  • Width: Width of the antenna.
  • Height: Height of the antenna.
  • Depth: Depth of the antenna.
  • Altitude (CL): Altitude of the antenna center point with respect to the altitude of the Center Line of the antenna.
  • Coordinates: Coordinates of the detected antenna.

Under the hypothesis that the centerline of the tower is perfectly aligned with gravity, the orientation angles and dimensions of the antenna would be the ones shown in the images.

Side_view-downtilt__1_.png Downtilt_plumb_0__1_.png
Figure 1. Side view. Figure 2. Zero-degree downtilt and plumb.
Positive_downtilt_plumb__1_.png Negative_downtilt_plumb.png
Figure 3. Positive downtilt and plumb. Figure 4. Negative downtilt and plumb.
Top_view-azimuth__1_.png Top_view_az_330.png
Figure 5. Top view. 45 degrees of azimuth. Figure 6. Top view. 330 degrees of azimuth.
Was this article helpful?
4 out of 5 found this helpful

Article feedback (for troubleshooting, post here instead)


Article is closed for comments.