SYSTEMS AND METHODS FOR IDENTIFICATION OF ANIMAL INDIVIDUALS

Claims

1. A method for identifying animals, comprising an animal registration phase, including steps in which capture, using the device, a video of the first known animal in the first perspective and the second perspective, determine the first data vector based on the video of the first known animal using the model, receiving, at the device, a first set of identifying information of a first known animal, storing the first data vector and the first set of identifying information for the first known animal in a database, capture, using the device, video of the second known animal in the first perspective and the second perspective, determine a second data vector based on the video of the second known animal using the model, receiving, at the device, a second set of identifying information of a second known animal, storing a second data vector and a second set of identifying information for the second known animal in a database; and the animal identification phase, which includes the stages at which: capturing, using the device, at least one image of an unidentified animal, determining a new data vector based on at least one image of an unidentified animal using the model, comparing the new data vector with the first data vector and the second data vector and identifying the unidentified animal as the first known animal or the second known animal, and display on the device the identified animal, including the corresponding set of identifying information.

2. The method according to claim 1, wherein the animal identification phase further includes a step of receiving feedback on the device regarding the accuracy of identifying the unidentified animal as an identified animal.

3. The method according to claim 1, wherein the animal is a cow, sheep, horse, pig, goat, chicken, dog or cat.

4. The method of claim 1, wherein the first set of identifying information includes an ear tag, a site identifier, a gender, a note, feed quality, lameness, illness, antibiotic status, pen movement, or any combination thereof.

5. The method of claim 1, wherein the model is a machine learning model.

6. The method according to claim 5, wherein the model is trained using a set of augmented data.

7. The method of claim 6, wherein the set of augmented data includes a rotated image, a scaled image, a flipped image, a brightness-adjusted image, a generative image, or any combination of the above.

8. The method of claim 1, wherein capturing a video of the first known animal includes the step of capturing three-dimensional orientation data of the first known animal in the video.

9. The method of claim 1, wherein capturing a video of the first known animal includes the step of capturing the depth of the first known animal in the video.

10. The method of claim 1, wherein determining a new data vector based on at least one image of an unidentified animal includes a step in which key points in the image are detected, including the eyes, tip of the nose, ears, mouth and other features of the face of the unidentified animal.

11. The method of claim 5, wherein the model is trained using a triplet loss technique in which the reference input image is compared with the matching input image and the non-matching input image in order to minimize the difference between the reference input image and the matching input image and to maximize the distance between the reference input image and the non-matching input image.

12. The method according to claim 1, wherein determining the first data vector includes a step of generating at least one transformation of the muzzle of the first known animal in the video.

13. The method according to claim 1, wherein capturing a video of the first known animal includes the step of capturing the face of the first known animal using a bounding rectangle delimited on the device.

14. The method of claim 1, wherein the video capture of the first known animal is performed while the first known animal is in the animal restraint mechanism; and wherein the device is movable relative to the animal restraint mechanism.

15. The method of claim 1, wherein determining the first data vector based on the video of the first known animal includes the step of using three-dimensional orientation or depth data from the video.