US 12,340,558 B2
Anomaly detection using augmented reality (AR) and artificial intelligence (AI)
Sudha Sreejith, Bangalore (IN)
Assigned to The Boeing Company, Arlington, VA (US)
Filed by The Boeing Company, Chicago, IL (US)
Filed on Apr. 8, 2022, as Appl. No. 17/716,863.
Claims priority of application No. 202111032984 (IN), filed on Jul. 22, 2021.
Prior Publication US 2023/0021676 A1, Jan. 26, 2023
Int. Cl. G06V 10/764 (2022.01); G06V 30/19 (2022.01)
CPC G06V 10/764 (2022.01) [G06V 30/19127 (2022.01)] 20 Claims
OG exemplary drawing
 
1. A method comprising:
taking, with a camera, an image of a physical object;
processing the image by:
converting the image into a vector data file,
inputting the vector data file to a trained machine learning model artificial intelligence (AI),
executing the AI on the vector data file to produce an output comprising a classification of an anomaly in the image, and
converting the output into a reconfigured output comprising a data format configured for use with augmented reality (AR) hardware, wherein converting the output into the reconfigured output comprises:
tracking the anomaly based on the classification to generate a first tracked object,
aligning the first tracked object with the image of the physical object, and
generating, responsive to the AI classifying the anomaly as anomalous, the reconfigured output comprising render graphics executable by the AR hardware to perform highlighting with respect to the first tracked object;
transmitting the reconfigured output to the AR hardware;
displaying, using the reconfigured output, the physical object on a display device of the AR hardware; and
highlighting, on the display device concurrently with displaying the physical object, the anomaly.