This question can be answered well by the co-founders of the Dunedin-based Iris Data Science, Mr. Benoit Auvray & Greg Peyroux.
They have been working on the recognition of faces of shapes in order to re-identify the sheep cheaply, possibly eliminating the need for the ear-tags meanwhile resolving the other farm manager and way 6broader issues.
While facial recognition has been established for the pigs in China as well as the cattle in the U.S., there wasn’t any known study or work being done on the sheep.
The images of the faces of the sheep were collected and then into a machine-learning model. Gradually, it learned itself how to identify the sheep by finding recognizable features.
The project was able to gather attention when it was featured at MobileTECH 2019 in Rotorua in April which basically a yearly event which shows various digital technologies for the forestry, agriculture as well as horticultural sectors & at the TEXpo.
Mister Peyroux was cautiously very confident about the potential of the project. He passed the following statement: “In this game, the success is not completely defined by how much good the idea is. There are a lot of other things that should happen but we are getting too much amused.”
They developed an initial prototype and were hoping to get done with the field trials by the end of the year. While it could be wild in the industry, they would have to wait and see how it would go.
Those farmers who would want to use it would need to first install a system of the camera, in fact on a race, for the collection of images of their sheep.
Some giant players in the industry were discussing with them about their project, while they were also interested in a profound learning pasture quality system which they had established.