US 12,249,057 B2
Systems and methods for creating and/or analyzing three-dimensional models of infrastructure assets
Benjamin T. Schmidt, Oakmont, PA (US); Miguel Dickson, Pittsburgh, PA (US); Christoph Mertz, Pittsburgh, PA (US); and Anshul Goyal, Pittsburgh, PA (US)
Assigned to RoadBotics, Inc., Pittsburgh, PA (US)
Filed by RoadBotics, Inc., Pittsburgh, PA (US)
Filed on Aug. 28, 2023, as Appl. No. 18/457,213.
Application 18/457,213 is a continuation of application No. 17/925,958, granted, now 11,769,238, issued on Sep. 26, 2023, previously published as PCT/US2021/032820, filed on May 17, 2021.
Claims priority of provisional application 63/026,602, filed on May 18, 2020.
Prior Publication US 2024/0127416 A1, Apr. 18, 2024
Int. Cl. G06K 9/00 (2022.01); G06T 7/00 (2017.01); G06T 7/20 (2017.01); G06T 7/90 (2017.01); G06T 17/00 (2006.01); G06V 20/56 (2022.01)
CPC G06T 7/0002 (2013.01) [G06T 7/20 (2013.01); G06T 7/90 (2017.01); G06T 17/00 (2013.01); G06V 20/588 (2022.01); G06T 2207/10016 (2013.01); G06T 2207/10024 (2013.01); G06T 2207/10028 (2013.01); G06T 2207/30184 (2013.01); G06T 2207/30256 (2013.01); G06T 2210/56 (2013.01)] 24 Claims
OG exemplary drawing
 
1. A computer-implemented method comprising acts of:
analyzing sensor data to detect at least one infrastructure asset, wherein:
the sensor data comprises a plurality of images captured by a moving camera, the plurality of images comprising a first image and a second image;
the at least one infrastructure asset comprises at least one sign; and
analyzing sensor data to detect at least one infrastructure asset comprises:
identifying, from the first image, a first plurality of pixels corresponding to a first instance of a selected signage type; and
identifying, from the second image, a second plurality of pixels corresponding to a second instance of the selected signage type; and
using the plurality of images to generate a point cloud, wherein the point cloud is georeferenced based on geospatial metadata associated with the plurality of images;
identifying, from the point cloud, one or more points corresponding to the at least one sign, comprising:
mapping the first plurality of pixels to a first plurality of points in the point cloud;
mapping the second plurality of pixels to a second plurality of points in the point cloud;
determining an extent of overlap between the first and second pluralities of points; and
in response to determining that the extent of overlap exceeds a selected threshold, merging the first and second pluralities of points; and
using coordinates of the one or more points in the point cloud to associate geospatial information with the at least one sign.