The volume measurement tool in PIX4Dcatch measures the volume between a base surface (defined by a boundary polygon) and the acquired point cloud.
Input data
Using the point cloud as input, the volume measurement tool works by reconstructing object surfaces from the point cloud in true 3D. This allows it to fully capture the intricate details of any object’s shape, such as cavities on the walls of trenches, overhangs and hollow areas, and similar complex geometrical features. The tool only considers point cloud points within the horizontal footprint of the base surface. The base surface separates the regions of space named cut (above the surface) and fill (below the surface) volumes. For a base surface to be considered valid, its horizontal footprint cannot self-intersect (the surface has to be 2.5 dimensional).
Results
After drawing a volume polygon and clicking compute, the results are displayed.
The measured volumes are shown together with their uncertainty, more specifically:
-
Cut volume [units³ ]: Volume above the volume base. The volume of solid objects (anything but air) above the base. For instance, the volume of material in a stockpile;
-
Fill volume [units³ ]: Volume below the volume base. The volume of empty space (air) below the base. For instance, the volume of sand it would take to fill a trench;
-
Cut volume uncertainty [units³ ]. See the uncertainty section for more details;
-
Fill volume uncertainty [units³ ]. See the uncertainty section for more details.
Computed surfaces are shown together with the input point cloud, allowing for an inspection of the results to ensure that the correct object shape has been reconstructed.
Volume uncertainty
Volume measurements are reported with an uncertainty that can be read as computed volume +/- uncertainty. This means that the volume is estimated to lie in the [volume-uncertainty, volume+uncertainty] interval with a confidence of about 68%. For confidences of 95% and 99%, the estimated intervals are respectively [volume-2*uncertainty, volume+2*uncertainty] and [volume-3*uncertainty, volume+3*uncertainty].
The uncertainty estimation is based on a theoretical model, which accounts for different sources of uncertainty, such as the point cloud noise, completeness and density. The model does not take scale errors (which can be prevented using an RTK receiver) or errors in the base surface into account. An extensive validation of the theoretical model has been carried out by running experiments on various types of objects.