Automated identification of fish filets

1 . A method, comprising:

acquiring, using a digital camera, a first image of a first fish filet;

encoding the first image into a first feature vector consumable by at least one predictive model, wherein at least one predictive model includes a machine learning model that is trained, using previously acquired images of second fish filets and corresponding labels indicative of types of fish from which the second fish filets were cut, to classify images of fish filets into two or more categories corresponding to the types of fish;

processing, with the at least one predictive model, the first feature vector to identify a first type of fish from which the first fish filet was cut; and

causing at least one device to output an indication of the first type of fish identified by the at least one predictive model.

2 . The method of claim 1 , further comprising:

associating a second image, included among the previously acquired images, with a third image of a whole fish from which one of the second fish filets was cut;

determining, based on the third image, that the whole fish is of the first type;

determining, based on the second image being associated with the third image, that the one of the second fish filets is of the first type;

associating the second image with a first tag indicating that the second image is of a fish filet cut from a fish of the first type;

encoding the second image into a second feature vector consumable by the at least one predictive model; and

using the second feature vector and the first tag to train the machine learning model.

3 . The method of claim 1 , further comprising:

disposing an indicator of the first type of fish at a location where one of the second fish filets is to be cut;

acquiring a second image of the one of the second fish filets that includes a representation of the indicator;

determining, based on the indicator represented in the second image, that the one of the second fish filets was cut from a fish of the first type;

associating the second image with a first tag indicating that the second image is of a fish filet cut from a fish of the first type;

encoding the second image into a second feature vector consumable by the at least one predictive model; and

using the second feature vector and the first tag to train the machine learning model.

4 . The method of claim 1 , wherein:

the at least one device includes a mobile device operated by a user;

the mobile device includes the digital camera used to acquire the first image; and

the at least one predictive model resides on the mobile device.

5 . The method of claim 1 , wherein the at least one device includes a mobile device operated by a user, the mobile device includes the digital camera used to acquire the first image, and the at least one predictive model resides on a remote computing system, and the method further comprises:

receiving, by the remote computing system and from the mobile device, the first image; and

sending, from the remote computing system to the mobile device, the indication of the first type of fish.

6 . A method, comprising:

acquiring, with a digital camera of a mobile device, a first image of a first fish filet;

causing the first image to be processed by at least one predictive model to identify a first type of fish from which the first fish filet was cut, wherein at least one predictive model includes a machine learning model that is trained, using previously acquired images of second fish filets and corresponding labels indicative of types of fish from which the second fish filets were cut, to classify images of fish filets into two or more categories corresponding to the types of fish; and

outputting, with the mobile device, an indication of the first type of fish.

7 . The method of claim 6 , further comprising:

associating a second image, included among the previously acquired images, with a third image of a whole fish from which one of the second fish filets was cut;

determining, based on the third image, that the whole fish is of the first type;

determining, based on the second image being associated with the third image, that the one of the second fish filets was cut from a fish of the first type;

associating the second image with a first tag indicating that the second image is of a fish filet cut from a fish of the first type;

encoding the second image into a second feature vector consumable by the at least one predictive model; and

using the second feature vector and the first tag to train the machine learning model.

8 . The method of claim 6 , further comprising:

disposing an indicator of the first type of fish at a location where one of the second fish filets is to be cut;

acquiring a second image of the one of the second fish filets that includes a representation of the indicator;

determining, based on the indicator represented in the second image, that the one of the second fish filets was cut from a fish of the first type;

associating the second image with a first tag indicating that the second image is of a fish filet cut from a fish of the first type;

encoding the second image into a second feature vector consumable by the at least one predictive model; and

using the second feature vector and the first tag to train the machine learning model.

9 . The method of claim 6 , wherein the at least one predictive model resides on the mobile device.

10 . The method of claim 6 , wherein the at least one predictive model resides on a remote computing system, and the method further comprises:

sending, from the mobile device to the remote computing system, the first image of the first fish filet; and

receiving, by the mobile device and from the remote computing system, the indication of the first type of fish.

11 . A system, comprising:

at least one processor; and

at least one computer-readable medium encoded with instructions which, when executed by the at least one processor, cause the system to:

acquire, using a digital camera, a first image of a first fish filet,

encode the first image into a first feature vector consumable by at least one predictive model, wherein at least one predictive model includes a machine learning model that is trained, using previously acquired images of second fish filets and corresponding labels indicative of types of fish from which the second fish filets were cut, to classify images of fish filets into two or more categories corresponding to the types of fish;

process, with the at least one predictive model, the first feature vector to identify a first type of fish from which the first fish filet was cut, and

cause at least one device to output an indication of the first type of fish identified by the at least one predictive model.

12 . The system of claim 11 , wherein the at least one computer-readable medium is further encoded with additional instructions which, when executed by the at least one processor, further cause the system to:

associate a second image, included among the previously acquired images, with a third image of a whole fish from which one of the second fish filets was cut;

determine, based on the third image, that the whole fish is of the first type;

determine, based on the second image being associated with the third image, that the one of the second fish filets was cut from a fish of the first type;

associate the second image with a first tag indicating that the second image is of a fish filet that was cut from a fish of the first type;

encode the second image into a second feature vector consumable by the at least one predictive model; and

use the second feature vector and the first tag to train the machine learning model.

13 . The system of claim 11 , wherein the at least one computer-readable medium is further encoded with additional instructions which, when executed by the at least one processor, further cause the system to:

acquire a second image of one of the second fish filets that includes a representation of an indicator of the first type of fish that is disposed at a location where the one of the second fish filets is being cut;

determine, based on the indicator represented in the second image, that the one of the second fish filets was cut from a fish of the first type;

associate the second image with a first tag indicating that the second image is of a fish filet that was cut from a fish of the first type;

encode the first image into a second feature vector consumable by the at least one predictive model; and

use the second feature vector and the first tag to train the machine learning model.

14 . The system of claim 11 , wherein:

the at least one device includes a mobile device operated by a user; and

the at least one predictive model resides on the mobile device.

15 . The system of claim 11 , wherein the at least one device includes a mobile device operated by a user and the at least one predictive model resides on a remote computing system, and the at least one computer-readable medium is further encoded with additional instructions which, when executed by the at least one processor, further cause the system to:

receive, by the remote computing system and from the mobile device, the first image; and

send, from the remote computing system to the mobile device, the indication of the first type of fish.

16 . A mobile device, comprising:

a digital camera;

a display;

at least one processor; and

at least one computer-readable medium encoded with instructions which, when executed by the at least one processor, cause the mobile device to:

cause the digital camera to acquire a first image of a first fish filet,

cause the first image to be processed by at least one predictive model to identify a first type of fish from which the first fish filet was cut, wherein at least one predictive model includes a machine learning model that is trained, using previously acquired images of second fish filets and corresponding labels indicative of types of fish from which the second fish filets were cut, to classify images of fish filets into two or more categories corresponding to the types of fish, and

cause the display to output an indication of the first type of fish.

17 . The mobile device of claim 16 , wherein the at least one predictive model resides on the mobile device.

18 . The mobile device of claim 16 , wherein the at least one predictive model resides on a remote computing system, and the at least one computer-readable medium is further encoded with additional instructions which, when executed by the at least one processor, further cause the mobile device to:

send, to the remote computing system, the first image of the first fish filet; and

receive, from the remote computing system, the indication of the first type of fish.