US 12,462,404 B2
System and method for transforming video data into directional object count
Terrance Edward Boult, Colorado Springs, CO (US)
Assigned to PUSHPIN TECHNOLOGY, L.L.C., Evergreen, CO (US)
Filed by PUSHPIN TECHNOLOGY, L.L.C., Evergreen, CO (US)
Filed on Apr. 24, 2024, as Appl. No. 18/644,858.
Application 18/644,858 is a continuation of application No. 17/650,744, filed on Feb. 11, 2022, granted, now 12,002,225.
Application 17/650,744 is a continuation of application No. 16/435,008, filed on Jun. 7, 2019, granted, now 11,288,820, issued on Mar. 29, 2022.
Claims priority of provisional application 62/682,906, filed on Jun. 9, 2018.
Prior Publication US 2024/0273737 A1, Aug. 15, 2024
This patent is subject to a terminal disclaimer.
Int. Cl. G06T 7/254 (2017.01); G06N 20/00 (2019.01); G06T 5/50 (2006.01); G06T 7/215 (2017.01)
CPC G06T 7/254 (2017.01) [G06N 20/00 (2019.01); G06T 5/50 (2013.01); G06T 7/215 (2017.01)] 6 Claims
OG exemplary drawing
 
1. A system for transforming a sequence of image data into moving object counts comprising:
a video camera including a video sensor, a video processor, and input/output for sequentially entering input images consisting of a single column containing a moving object region to be counted; and
a computer system receiving the input images consisting of a single column containing the moving object region to be counted from the video camera, creating a sweep image, and determining object counts by counting objects in the sweep image, the computer system further providing for machine learning and directional computation; and
a display that communicates the object count to an external system;
wherein the computer system applies sweep images for machine learning to search for the parameters that optimize sweep image-based transformation.