Underwater camera as light sensor

1. A computer-implemented method comprising:

obtaining, by a camera system that includes an upward-facing camera and one or more side-facing cameras, images of a surface of water captured by the upward-facing camera facing towards the surface of the water from a depth within an enclosure that encloses aquatic livestock;

determining that the depth of the upward-facing camera satisfies one or more depth criteria;

in response to determining that the depth of the upward-facing camera satisfies the one or more depth criteria, determining from the images of the surface of the water that were captured by the upward-facing camera, an ambient light metric at the depth;

determining that one or more action criteria are satisfied based on the ambient light metric at the depth; and

based on determining that the one or more action criteria are satisfied based on the ambient light metric at the depth, initiating performance of an action to be performed for the aquatic livestock.

2. The computer-implemented method of claim 1 , wherein determining that the one or more action criteria are satisfied based on the ambient light metric at the depth comprises:

determining an amount of ambient light based on the ambient light metric; and

determining that the amount of ambient light at the depth exceeds a light threshold.

3. The computer-implemented method of claim 1 , wherein determining that one or more action criteria are satisfied comprises:

determining an amount of ambient light based on the ambient light metric;

generating a data set that represents the amount of ambient light determined at the depth and previous amounts of ambient light determined at the depth;

providing the data set to a machine-learning trained model that is trained to output actions for input data sets that represent amounts of ambient light determined at depths; and

receiving an indication of the action from the machine-learning trained model.

4. The computer-implemented method of claim 1 , wherein the ambient light metric is determined using an intensity value of one or more pixels of the image.

5. The computer-implemented method of claim 1 , wherein determining that the depth of the camera satisfies one or more depth criteria comprises:

comparing the depth to a depth threshold;

determining that the depth does not exceed the depth threshold; and

based on determining that the depth does not exceed the depth threshold, determining that the depth of the camera satisfies the one or more depth criteria.

6. The computer-implemented method of claim 1 , wherein determining from the images of the surface of the water, an ambient light metric at the depth comprises:

determining a first amount of ambient light at the depth from a first image of the images;

determining a second amount of ambient light at the depth from a second image of the images; and

determining the ambient light metric at the depth based on an aggregate of the first amount of ambient light at the depth from the first image of the images and the second amount of ambient light at the depth from the second image of the images.

7. The computer-implemented method of claim 1 , wherein initiating performance of an action to be performed for the aquatic livestock comprises:

transmitting an instruction to a feeder subsystem, wherein the instruction indicates to provide feed to the aquatic livestock.

8. The computer-implemented method of claim 1 , wherein initiating performance of an action to be performed for the aquatic livestock comprises:

transmitting an instruction to a feeder subsystem, wherein the instruction indicates to stop providing feed to the aquatic livestock.

9. The computer-implemented method of claim 1 , wherein determining that one or more action criteria are satisfied based on the ambient light metric at the depth comprises:

determining a predicted amount of future ambient light at the depth based on the current ambient light metric and environmental data;

determining that the predicted amount of ambient light at the depth exceeds a threshold value; and

based on determining that the predicted amount of ambient light at the depth exceeds the threshold value, determining that the one or more action criteria are satisfied.

10. The computer-implemented method of claim 1 , wherein determining that one or more action criteria are satisfied based on the amount of ambient light at the depth comprises:

determining a feeding metric based on the ambient light metric and a measurement of feed delivered over a period;

determining that the feeding metric satisfies one or more feeding metric criteria; and

based on determining that the feeding metric satisfies the one or more feeding metric criteria, determining that the one or more action criteria are satisfied.

11. A system comprising:

one or more processors; and

one or more non-transitory computer-readable media storing instructions that, when executed, cause the one or more processors to perform operations comprising:

obtaining, by a camera system that includes an upward-facing camera and one or more side-facing cameras, images of a surface of water captured by the upward-facing camera facing towards the surface of the water from a depth within an enclosure that encloses aquatic livestock;

determining that the depth of the upward-facing camera satisfies one or more depth criteria;

in response to determining that the depth of the upward-facing camera satisfies the one or more depth criteria, determining from the images of the surface of the water that were captured by the upward-facing camera, an ambient light metric at the depth;

determining that one or more action criteria are satisfied based on the ambient light metric at the depth; and

based on determining that the one or more action criteria are satisfied based on the ambient light metric at the depth, initiating performance of an action to be performed for the aquatic livestock.

12. The system of claim 11 , wherein determining that one or more action criteria are satisfied comprises:

determining an amount of ambient light based on the ambient light metric; and

determining that the amount of ambient light at the depth exceeds a light threshold.

13. The system of claim 11 , wherein determining that one or more action criteria are satisfied comprises:

determining an amount of ambient light based on the ambient light metric;

generating a data set that represents the amount of ambient light determined at the depth and previous amounts of ambient light determined at the depth;

providing the data set to a machine-learning trained model that is trained to output actions for input data sets that represent amounts of ambient light determined at depths; and

receiving an indication of the action from the machine-learning trained model.

14. The system of claim 11 , wherein determining that the depth of the camera satisfies one or more depth criteria comprises:

comparing the depth to a depth threshold;

determining that the depth does not exceed the depth threshold; and

based on determining that the depth does not exceed the depth threshold, determining that the depth of the camera satisfies the one or more depth criteria.

15. The system of claim 11 , wherein determining from the images of the surface of the water, an ambient light metric at the depth comprises:

determining a first amount of ambient light at the depth from a first image of the images;

determining a second amount of ambient light at the depth from a second image of the images; and

determining the ambient light metric at the depth based on an aggregate of the first amount of ambient light at the depth from the first image of the images and the second amount of ambient light at the depth from the second image of the images.

16. The system of claim 11 , wherein initiating performance of an action to be performed for the aquatic livestock comprises:

transmitting an instruction to a feeder subsystem, wherein the instruction indicates to provide feed to the aquatic livestock.

17. The system of claim 11 , wherein initiating performance of an action to be performed for the aquatic livestock comprises:

transmitting an instruction to a feeder subsystem, wherein the instruction indicates to stop providing feed to the aquatic livestock.

18. The system of claim 11 wherein determining that one or more action criteria are satisfied based on the ambient light metric at the depth comprises:

determining a predicted amount of future ambient light at the depth based on the current ambient light metric and environmental data;

determining that the predicted amount of ambient light at the depth exceeds a threshold value; and

based on determining that the predicted amount of ambient light at the depth exceeds the threshold value, determining that the one or more action criteria are satisfied.

19. The system of claim 11 , wherein determining that one or more action criteria are satisfied based on the amount of ambient light at the depth comprises:

determining a feeding metric based on the ambient light metric and a measurement of feed delivered over a period;

determining that the feeding metric satisfies one or more feeding metric criteria; and

based on determining that the feeding metric satisfies the one or more feeding metric criteria, determining that the one or more action criteria are satisfied.

20. A non-transitory, computer-readable medium storing one or more instructions executable by a computer system to perform operations comprising:

obtaining, by a camera system that includes an upward-facing camera and one or more side-facing cameras, images of a surface of water captured by the upward-facing camera facing towards the surface of the water from a depth within an enclosure that encloses aquatic livestock;

determining that the depth of the upward-facing camera satisfies one or more depth criteria;

in response to determining that the depth of the upward-facing camera satisfies the one or more depth criteria, determining from the images of the surface of the water that were captured by the upward-facing camera, an ambient light metric at the depth;

determining that one or more action criteria are satisfied based on the ambient light metric at the depth; and

based on determining that the one or more action criteria are satisfied based on the ambient light metric at the depth, initiating performance of an action to be performed for the aquatic livestock.