US 12,288,399 B2
Secure edge platform using image classification machine learning models
Shaun Sandoval, Milton, FL (US); Chad Carpenter, Montrose, SD (US); Joshua Smolenak, Racho Mission Viejo, CA (US); James Straatmeyer, Hartford, SD (US); and Tony Schoborg, Parker, SD (US)
Assigned to Sandoval Custom Creations, Inc., Larkspur, CO (US)
Filed by Sandoval Custom Creations, Inc, Larkspur, CO (US)
Filed on Sep. 20, 2022, as Appl. No. 17/933,529.
Claims priority of provisional application 63/246,687, filed on Sep. 21, 2021.
Prior Publication US 2023/0093698 A1, Mar. 23, 2023
Int. Cl. G06V 20/52 (2022.01); G06F 21/62 (2013.01); G06V 10/70 (2022.01); G06V 10/764 (2022.01); G06V 20/62 (2022.01); G06V 40/16 (2022.01)
CPC G06V 20/52 (2022.01) [G06F 21/6245 (2013.01); G06V 10/764 (2022.01); G06V 10/87 (2022.01); G06V 20/625 (2022.01); G06V 40/172 (2022.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method comprising:
identifying, by an edge platform that comprises at least one camera, one or more particular image classification models from among multiple image classification models that are associated with the edge platform, each image classification model being configured to generate classification output data based upon processing image data that is generated by the at least one camera included on the edge platform;
receiving, by the edge platform, particular image data that was generated by the at least one camera included on the edge platform;
providing, by the edge platform, the particular image data to the one or more particular image classification models;
in response to providing the particular image data to the one or more particular image classification models, receiving, by the edge platform, particular classification output data from the one or more particular image classification models; and
in response to receiving the particular classification output data from the one or more particular image classification models, generating, by the edge platform, augmentation data associated with the particular image data, combining the augmentation data and the particular classification output data according to a structured format to generate detection data, then transmitting, by the edge platform, the detection data to a central server platform.