| CPC G06V 10/42 (2022.01) [A01K 61/95 (2017.01); G06F 18/2185 (2023.01); G06F 18/23213 (2023.01); G06F 18/2433 (2023.01); G06T 7/0012 (2013.01); G06T 7/62 (2017.01); G06T 7/70 (2017.01); G06V 20/05 (2022.01); G06V 40/10 (2022.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01)] | 18 Claims |

|
1. A computer-implemented method comprising:
generating images of farmed fish within an enclosure using an underwater camera;
applying a machine learning-trained model to features of the images to generate respective estimates for the farmed fish in the images, wherein each of the estimates represent a range of weights;
generating a distribution that indicates, for each of multiple values or ranges, a count of the farmed fish whose respective generated estimate is associated with the value or range; and
adjusting an automated process associated with the enclosure based at least on the distribution.
|