US 12,217,127 B2
Protecting against malicious attacks in images
Gad Karmi, Davis, CA (US); and Mark Bapst, South Barrington, IL (US)
Assigned to QUALCOMM Incorporated, San Diego, CA (US)
Filed by QUALCOMM Incorporated, San Diego, CA (US)
Filed on Feb. 10, 2023, as Appl. No. 18/167,787.
Prior Publication US 2024/0273322 A1, Aug. 15, 2024
Int. Cl. G06K 7/14 (2006.01); G06F 16/955 (2019.01); G06T 19/00 (2011.01); H04L 9/32 (2006.01)
CPC G06K 7/1443 (2013.01) [G06F 16/9566 (2019.01); G06K 7/1417 (2013.01); G06T 19/006 (2013.01); H04L 9/3247 (2013.01)] 30 Claims
OG exemplary drawing
 
1. A method for image processing, comprising:
obtaining, from one or more image sensors, an image of an environment;
detecting a first visual code in the image;
decoding the first visual code;
verifying safety of contents of the first visual code, wherein verifying the safety of the contents of the first visual code comprises:
determining that the first visual code is signed;
determining that the first visual code includes an indication of a signing authority; and
verifying the safety of the contents of the first visual code based on the indication of the signing authority;
determining that the contents of the first visual code contain a uniform resource locator (URL); and
prefetching data linked by the URL based on verifying the safety of the contents of the first visual code.