US 12,276,617 B2
Method for defect detection for rolling elements
Ray Amerson, Rock Hill, NC (US); and Balaji Chandrasekaran, Charlotte, NC (US)
Assigned to Schaeffler Technologies AG & Co. KG, Herzogenaurach (DE)
Filed by Schaeffler Technologies AG & Co. KG, Herzogenaurach (DE)
Filed on Jun. 16, 2022, as Appl. No. 17/841,993.
Prior Publication US 2023/0408421 A1, Dec. 21, 2023
Int. Cl. G01N 21/89 (2006.01); G01N 21/892 (2006.01)
CPC G01N 21/8901 (2013.01) [G01N 21/8922 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method of detecting defects in a rolling element, the method comprising:
(i) scanning an outer surface of a rolling element via a microscope assembly to obtain a plurality of surface images of the rolling element;
(ii) associating each of the plurality of surface images with a specific region of the rolling element, wherein the plurality of surface images are stored in a database;
(iii) identifying defects based on the plurality of surface images via a processing module using an initial set of analysis parameters;
(iv) generating a plurality of interactive inspection icons associated with the plurality of surface images that each correspond to a unique geographic region of the rolling element, and displaying the plurality of interactive inspection icons via a user interface, wherein the user interface includes an input interface including an approve option and a rejection option;
(v) storing decisional information regarding selection of the approve option and the rejection option from the input interface in the database; and
(vi) updating the initial set of analysis parameters based on the decisional information;
wherein step (iii) further comprises identifying the defects via an AI vision solution engine that is configured to identify defects based on analysis parameters.