US 12,354,351 B2
Automatic and semi-automatic traffic configuration system
Justin Alexander Eichel, Kitchener (CA); Chu Qing Hu, Newmarket (CA); Fatemeh Mohammadi, Waterloo (CA); and David Martin Swart, Waterloo (CA)
Assigned to Miovision Technologies Incorporated, Kitchener (CA)
Filed by Miovision Technologies Incorporated, Kitchener (CA)
Filed on Nov. 18, 2021, as Appl. No. 17/529,455.
Claims priority of provisional application 63/198,907, filed on Nov. 20, 2020.
Prior Publication US 2022/0165060 A1, May 26, 2022
Int. Cl. G06V 20/40 (2022.01); G06V 10/762 (2022.01); G06V 20/54 (2022.01); G06V 20/64 (2022.01)
CPC G06V 20/41 (2022.01) [G06V 10/762 (2022.01); G06V 20/54 (2022.01); G06V 20/64 (2022.01); G06V 2201/08 (2022.01)] 20 Claims
OG exemplary drawing
 
1. A method of refining a configuration used by a video analysis system for analyzing videos, comprising:
deploying the configuration to at least one video capturing device positioned to capture a video of a scene, the configuration including a spatial arrangement of a first set of one or more elements within the scene, a region of interest including the first set of one or more elements defined within the scene for analyzing the video, and one or more of settings and parameters associated with the video and the video capturing device, wherein the region of interest comprises at least one roadway or traffic intersection;
receiving video data from the at least one video capturing device, the video data including data associated with one or more of objects detected within the scene;
assigning a label to each of the one or more objects detected within the scene, the label being indicative of a type of the object;
determining a movement pattern for each type of object based on the data associated with the one or more objects detected within the scene;
refining the spatial arrangement of the first set of one or more of the elements within the scene based on the determined movement patterns corresponding to each type of objects;
identifying one or more areas within the scene indicating changes to the scene over time based on the received data associated with the objects, wherein the identified one or more areas include elements not included in the first set of one or more elements from the region of interest as defined within the deployed configuration;
refining the region of interest to include a second set of one or more elements that include the elements not included in the first set of one or more elements based on the changes to the scene in the identified one or more areas to generate a refined region of interest;
adjusting the one or more of the settings and parameters associated with the video and the video capturing device based on the refined spatial arrangement and the refined region of interest to generate a refined configuration for use in analyzing videos; and
deploying the refined configuration to the at least one video capturing device.