CPC G06T 7/70 (2017.01) [G06F 18/2431 (2023.01); G06T 3/40 (2013.01); G06T 7/11 (2017.01); G06T 9/00 (2013.01); G06V 10/255 (2022.01); G06V 20/20 (2022.01); G06V 20/41 (2022.01); G06V 20/52 (2022.01); G08B 13/19604 (2013.01); G08B 13/19606 (2013.01); G08B 13/19673 (2013.01); G08B 13/19689 (2013.01); G08B 29/185 (2013.01); H04N 7/185 (2013.01); H04N 7/56 (2013.01); H04N 23/67 (2023.01); H04N 23/69 (2023.01); H04N 23/695 (2023.01); G06T 2207/20132 (2013.01); G08B 21/18 (2013.01)] | 4 Claims |
1. A method, comprising:
receiving, from a first image capture device, a first series of images of a scene, the first series of images having a first set of resolution attributes;
receiving, from the first image capture device, a second series of images of the scene, the second series of images having a second set of resolution attributes that are higher than the first set of resolution attributes;
detecting an object of interest in the first series of images;
locating the object of interest in the second series of images;
selecting an area of the second series of images, the area including the object of interest;
analyzing the area; and
when it is determined additional processing is needed to classify the object of interest:
receiving, from a second image capture device, a third series of images of the area, the third series of images comprising an optically zoomed view of the area;
performing a first crop operation within the optically zoomed view of the area, the first crop operation including the object of interest;
generating a cropped image based on performing the first crop operation; and
classifying the object of interest using the cropped image.
|