Visual detection of haloclines

1. A method comprising:

moving a camera through different depths of water within a fish enclosure;

capturing, at the different depths, images of the fish;

determining, in each of the images of the fish that are captured at the different depths, a quantity of pixels that are out-of-focus in the image and a quantity of pixels that are in-focus in the image;

determining that, in one or more images of the fish that are associated with a particular depth, a proportion of the pixels that are out-of-focus or that are in-focus in the one or more image that are associated with the particular depth is different, by more than a threshold, than a proportion of pixels that are out-of-focus or that are in-focus, respectively, in one or more other images that are associated with a depth that is adjacent to the particular depth; and

based on determining that the proportion of pixels that is associated with the particular depth is different, by more than the threshold, than the proportion of pixels that are associated with the depth that is adjacent to the particular depth, detecting a halocline at the particular depth.

2. The method of claim 1 , wherein

the adjacent depth is above the particular depth.

3. The method of claim 2 , wherein detecting the halocline at the particular depth comprises:

determining that the halocline has a top at the particular depth.

4. The method of claim 1 , wherein

the adjacent depth is below the particular depth.

5. The method of claim 4 , wherein detecting the halocline at the particular depth comprises:

determining that the halocline has a bottom at the particular depth.

6. The method of claim 1 , comprising:

determining that a number of fish detected in a particular image satisfies a count criteria, wherein the quantity of pixels is determined based on determining that the number of fish satisfies a count criteria.

7. The method of claim 1 , comprising one or more of:

adjusting one or more of a feed depth that the fish are fed or a feed rate that the fish are fed based on detection of the halocline at the particular depth;

increasing sea lice mitigation based on detection of the halocline at the particular depth; or

disregarding one or more of sea lice detection or fish detection based on detection of the halocline at the particular depth.

8. The method of claim 1 , comprising:

providing a depth map that indicates a top of the halocline and a bottom of the halocline.

9. A system comprising:

one or more computers and one or more storage devices storing instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations comprising:

moving a camera through different depths of water within a fish enclosure;

capturing, at the different depths, images of the fish;

determining, in each of the images of the fish that are captured at the different depths, a quantity of pixels that are out-of-focus in the image and a quantity of pixels that are in-focus in the image;

determining that, in one or more images of the fish that are associated with a particular depth, a proportion of the pixels that are out-of-focus or that are in-focus in the one or more image that are associated with the particular depth is different, by more than a threshold, than a proportion of pixels that are out-of-focus or that are in-focus, respectively, in one or more other images that are associated with a depth that is adjacent to the particular depth; and

based on determining that the proportion of pixels that is associated with the particular depth is different, by more than the threshold, than the proportion of pixels that are associated with the depth that is adjacent to the particular depth, detecting a halocline at the particular depth.

10. The system of claim 9 , wherein

the adjacent depth is above the particular depth.

11. The system of claim 10 , wherein detecting the halocline at the particular depth comprises:

determining that the halocline has a top at the particular depth.

12. The system of claim 9 , wherein

the adjacent depth is below the particular depth.

13. The system of claim 12 , wherein detecting the halocline at the particular depth comprises:

determining that the halocline has a bottom at the particular depth.

14. The system of claim 9 , wherein the operations comprise:

determining that a number of fish detected in a particular image satisfies a count criteria, wherein the quantity of pixels is determined based on determining that the number of fish satisfies a count criteria.

15. The system of claim 9 , the operations comprising one or more of:

adjusting one or more of a feed depth that the fish are fed or a feed rate that the fish are fed based on detection of the halocline at the particular depth;

increasing sea lice mitigation based on detection of the halocline at the particular depth; or

disregarding one or more of sea lice detection or fish detection based on detection of the halocline at the particular depth.

16. A computer-readable storage device encoded with a computer program, the program comprising instructions that when executed by one or more computers cause the one or more computers to perform operations comprising:

moving a camera through different depths of water within a fish enclosure;

capturing, at the different depths, images of the fish;

determining, in each of the images of the fish that are captured at the different depths, a quantity of pixels that are out-of-focus in the image and a quantity of pixels that are in-focus in the image;

determining that, in one or more images of the fish that are associated with a particular depth, a proportion of the pixels that are out-of-focus or that are in-focus in the one or more image that are associated with the particular depth is different, by more than a threshold, than a proportion of pixels that are out-of-focus or that are in-focus, respectively, in one or more other images that are associated with a depth that is adjacent to the particular depth; and

based on determining that the proportion of pixels that is associated with the particular depth is different, by more than the threshold, than the proportion of pixels that are associated with the depth that is adjacent to the particular depth, detecting a halocline at the particular depth.

17. The computer-readable storage device of claim 16 , wherein the adjacent depth is above the particular depth.

18. The computer-readable storage device of claim 17 , wherein detecting the halocline at the particular depth comprises:

determining that the halocline has a top at the particular depth.

19. The computer-readable storage device of claim 16 , wherein the adjacent depth is below the particular depth.

20. The computer-readable storage device of claim 19 , wherein detecting the halocline at the particular depth comprises:

determining that the halocline has a bottom at the particular depth.

21. The computer-readable storage device of claim 16 , wherein the operations comprise:

determining that a number of fish detected in a particular image satisfies a count criteria, wherein the quantity of pixels is determined based on determining that the number of fish satisfies a count criteria.

22. The computer-readable storage device of claim 16 , comprising one or more of:

adjusting one or more of a feed depth that the fish are fed or a feed rate that the fish are fed based on detection of the halocline at the particular depth;

increasing sea lice mitigation based on detection of the halocline at the particular depth; or

disregarding one or more of sea lice detection or fish detection based on detection of the halocline at the particular depth.

23. The computer-readable storage device of claim 16 , comprising:

providing a depth map that indicates a top of the halocline and a bottom of the halocline.