US 12,335,627 B2
Inspection camera controller sequence
Boyun Wang, Santa Clara, CA (US); Weixiong Zheng, Santa Clara, CA (US); and Kedao Wang, Santa Clara, CA (US)
Assigned to UnitX, Inc., Santa Clara, CA (US)
Filed by UnitX, Inc., Santa Clara, CA (US)
Filed on Nov. 4, 2022, as Appl. No. 17/981,380.
Prior Publication US 2024/0155247 A1, May 9, 2024
Int. Cl. H04N 23/72 (2023.01); G06T 7/00 (2017.01)
CPC H04N 23/72 (2023.01) [G06T 7/0004 (2013.01); G06T 2200/24 (2013.01); G06T 2207/10144 (2013.01); G06T 2207/10152 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A system comprising:
a lighting apparatus including a plurality of independently controllable light sources and at least one camera;
a controller; and
a computer system comprising at least one hardware processor and a non-transitory computer-readable medium storing instructions that, when executed by the at least one hardware processor, cause the at least one hardware processor to perform operations comprising:
accessing a sequence, the sequence comprising a plurality of capture configurations, each of the plurality of capture configurations defining an exposure time and an indication of which of the independently controllable light sources to light during an image capture process, each of the plurality of capture configurations having its own dedicated artificial intelligence model used only for the corresponding capture configuration and not shared with other capture configurations;
storing the sequence in a memory accessible by the controller;
receiving, from the controller, an image captured during execution of a first capture configuration of the plurality of capture configurations of the sequence, the image captured by the at least one camera of the lighting apparatus while the lighting apparatus lights some of the plurality of independently controllable light sources in accordance with the first capture configuration;
accessing an artificial intelligence model corresponding only to the first capture configuration;
passing the image through the artificial intelligence model to make one or more inferences regarding the image, the inferences pertaining to a potential defect in a component captured in the image; and
packaging the image and the one or more inferences into a data package.