US 11,899,711 B2
Merchant logo detection artificial intelligence (AI) for injecting user control to ISO back-end transaction approvals between acquirer processors and issuer processors over data communication networks
Zhiqiang Zhang, San Ramon, CA (US); Vaduvur Bharghavan, Morgan Hill, CA (US); Qi Chen, San Jose, CA (US); Zhiling Liu, Sunnyvale, CA (US); and Kun Qian, Stanford, CA (US)
Assigned to ONDOT SYSTEMS INC., San Jose, CA (US)
Filed by ONDOT SYSTEMS INC., San Jose, CA (US)
Filed on Dec. 14, 2020, as Appl. No. 17/121,544.
Application 17/121,544 is a continuation in part of application No. 16/551,166, filed on Aug. 26, 2019, abandoned.
Application 16/551,166 is a continuation in part of application No. 16/227,560, filed on Dec. 20, 2018, abandoned.
Application 16/227,560 is a continuation in part of application No. 14/058,229, filed on Oct. 19, 2013, abandoned.
Application 14/058,229 is a continuation in part of application No. 13/527,544, filed on Jun. 19, 2012, abandoned.
Prior Publication US 2021/0165823 A1, Jun. 3, 2021
Int. Cl. G06F 16/58 (2019.01); G06N 3/04 (2023.01); G06Q 20/38 (2012.01); G06F 16/583 (2019.01); G06F 16/532 (2019.01)
CPC G06F 16/5866 (2019.01) [G06F 16/532 (2019.01); G06F 16/583 (2019.01); G06N 3/04 (2013.01); G06Q 20/388 (2013.01)] 10 Claims
OG exemplary drawing
 
1. A computer-implemented method in a user control server, implemented at least partially in hardware, for merchant logo detection with artificial intelligence (AI) for improving user control to ISO back-end transaction approvals between acquirer processors and issuer processors over a data communication network, the method comprising:
receiving a transmission, at a network communication interface coupled to the data communication network, ISO data packets with merchant name;
transforming raw merchant data from the ISO data packets to enriched merchant data;
identifying logo candidates for a specific ISO data from external resources based on the enriched merchant data;
filtering out low quality images of the logo candidates with image analysis including entropy ratio evaluations of the logo candidates;
processing the logo candidates with high quality filtering including classification of the logo candidates with a deep learning classifier for distinguishing logos from non-logos;
selecting a logo from the logo candidates to associate with the ISO data packets; and
generating a display having the selected logo associated with a transaction of the ISO data packets.