US 12,169,921 B2
Visual inspection system for automated detection of particulate matter in flexible medical containers
Armin Kappeler, Evanston, IL (US); Aggelos K. Katsaggelos, Chicago, IL (US); Georgios Andrea Bertos, Chicago, IL (US); Kirk Andrew Ashline, Third Lake, IL (US); and Neal Anthony Zupec, Spring Grove, IL (US)
Assigned to BAXTER INTERNATIONAL INC., Deerfield, IL (US); BAXTER HEALTHCARE SA, Glattpark (CH); and NORTHWESTERN UNIVERSITY, Evanston, IL (US)
Filed by Baxter International Inc., Deerfield, IL (US); Baxter Healthcare SA, Glattpark (CH); and Northwestern University, Evanston, IL (US)
Filed on Dec. 23, 2020, as Appl. No. 17/132,812.
Application 17/132,812 is a continuation of application No. 14/606,582, filed on Jan. 27, 2015, granted, now 10,878,551.
Claims priority of provisional application 61/932,065, filed on Jan. 27, 2014.
Prior Publication US 2021/0150699 A1, May 20, 2021
Int. Cl. G06T 7/00 (2017.01); G01N 15/1429 (2024.01); G01N 15/1433 (2024.01); G01N 15/1434 (2024.01); G06T 7/215 (2017.01); G01N 15/10 (2006.01); G01N 15/14 (2006.01); G01N 35/00 (2006.01)
CPC G06T 7/0004 (2013.01) [G01N 15/1429 (2013.01); G01N 15/1433 (2024.01); G01N 15/1434 (2013.01); G06T 7/215 (2017.01); G01N 2015/1027 (2024.01); G01N 2015/1488 (2013.01); G01N 2015/1497 (2013.01); G01N 2035/009 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method comprising:
detecting a first object in a fluid contained by a soft-sided flexible receptacle that appears in a plurality of sequential frames of digital image data of the receptacle;
determining at least one motion parameter corresponding to movement of the first object through the fluid;
classifying the first object using a classifier to determine if the first object is a big bubble corresponding to a predetermined big bubble object class; and
if the first object is not classified in the big bubble object class, classifying the first object into one of a small bubble object class or a particle object class based on the at least one motion parameter.