US 11,657,528 B2
System and method for mobile 3D scanning and measurement
Mehmet Afiny Affan Akdemir, Toronto (CA); Christian Garcia Salguero, Toronto (CA); and Victoria Sophie Howe, Toronto (CA)
Assigned to Xesto Inc., Toronto (CA)
Filed by Xesto Inc., Toronto (CA)
Filed on Jul. 21, 2020, as Appl. No. 16/934,007.
Claims priority of provisional application 62/941,779, filed on Nov. 28, 2019.
Prior Publication US 2021/0166411 A1, Jun. 3, 2021
Int. Cl. G06T 7/593 (2017.01); G06T 7/33 (2017.01); G06T 19/00 (2011.01)
CPC G06T 7/593 (2017.01) [G06T 7/33 (2017.01); G06T 19/00 (2013.01); G06T 2207/20036 (2013.01); G06T 2207/20084 (2013.01)] 24 Claims
OG exemplary drawing
 
1. A system for three-dimensional scanning and measurement, comprising:
a mobile device to scan a plurality of images of an object, the plurality of images providing views of the object from at least two angles;
at least one processor coupled to the mobile device to process the plurality of images of the object, the plurality of images providing views of the object from at least two angles;
a data store coupled to the at least one processor, the data store comprising a non-transient computer-readable storage medium having stored thereon computer-executable instructions for execution by the processor;
wherein the at least one processor executes computer-readable instructions stored on the computer-readable medium, comprising:
first instructions to receive the plurality of images of the object, the plurality of images providing views of the object from at least two angles;
second instructions to preprocess the plurality of images employing morphological refinement;
third instructions to create a source point cloud based at least in part on the plurality of images;
fourth instructions to remove outliers from the source point cloud;
fifth instructions to globally register the source point cloud, whereby a globally registered source point cloud is generated;
sixth instructions to generate a transformed source point cloud based at least in part on the globally registered source point cloud;
seventh instructions to compare the transformed source point cloud with a target point cloud stored on the non-transient computer-readable storage medium, whereby a point cloud comparison is generated;
eighth instructions to generate a stitched point cloud based at least in part on the point cloud comparison, whereby a stitched 3D model is produced;
ninth instructions to measure the stitched 3D model; and
tenth instructions executed by the at least one processor to input the source point cloud into a neural network coupled to the processor configured for parameter optimization based at least in part on one of an amount of overlap between the stitched point cloud and labeled correspondences.