US 12,243,220 B2
Systems and methods for equipment inspection
Milan Karunaratne, Orange, CA (US); and Nathan Thomas North, Seattle, WA (US)
Assigned to Transportation IP Holdings, LLC, Norwalk, CT (US)
Filed by Transportation IP Holdings, LLC, Norwalk, CT (US)
Filed on Nov. 10, 2021, as Appl. No. 17/523,343.
Claims priority of provisional application 63/130,085, filed on Dec. 23, 2020.
Prior Publication US 2022/0198632 A1, Jun. 23, 2022
Int. Cl. G06T 7/00 (2017.01)
CPC G06T 7/001 (2013.01) [G06T 2207/10016 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/30164 (2013.01)] 17 Claims
OG exemplary drawing
 
1. A method comprising:
receiving image data representative of an appearance of an equipment part, the image data obtained prior to performance of an inspection process on the equipment part;
evaluating the image data with a neural network formed from multiple layers that transform the image data using a machine learning model that defines baseline image data to determine whether the image data indicates that the equipment part has been prepared for the inspection process; and
preventing the performance of the inspection process on the equipment part responsive to the evaluation of the image data by the neural network with respect to the baseline image data indicating that the equipment part has not been prepared for the inspection process, wherein the image data is evaluated with respect to the baseline image data to determine whether the equipment part has been one or more of cleaned, marked with inspection marks, or marked with identifying marks.