ARTIFICIAL INTELLIGENCE, THE HEART OF OUR INNOVATIONS
Our camera has been entirely thought out and designed in France.
Thanks to our partnership with Nvidia, we were able to put a suitable component within our solution: the Nvidia Jetson Nano. Jetson Nano is a small, powerful computer for embedded AI systems and IoT that delivers the power of modern AI in a low-power platform. This computer is then powerful enough to ensure the proper functioning of our algorithm!
First of all, the basis of intelligent information obviously comes from the horse's stall. From there, the camera observes the horse 24/7 and learns its behaviour, habits and movements.
Therefore our AI model is unique for each horse and can alert in case of issues like start of colics, but also the overall welfare of your horse.
FROM THE HORSE TO YOU
In order to be able to alert you in the event of abnormal behavior detected in your horse's box, the information circulates along a well-defined path.
The artificial intelligence engine located in the camera ....
Then, to be stored and retrieved, all this information is sent to the cloud on our AWS (Amazon web services) servers with which we have a partnership. This not only stores the data but also secures it. Once in the cloud, the different data can go to another part of the artificial intelligence algorithm.
Finally, the data is returned to the mobile application. It will then allow you not only to visualize your horse but also to be alerted in the event of abnormal behavior detected thanks to the passage of data in the algorithms of artificial intelligence.
AN EFFICIENT NIGHT VISION
Day and night, keep an eye on your horse. The night vision integrated in our Smart Cam One, allows you to constantly check the condition of your horse. On our side, that allows us not to miss any event that might occur at night when there is no one at the stable. So as soon as unusual behavior on the part of your horse is detected, you are instantly alerted.
LANA, THE UNIQUE SELF LEARNING EQUINE AI
To make our artificial intelligence engine work, we use an algorithm that performs image and video recognition through a process called computer vision. In order to make this algorithm work, it is necessary to categorize each of the images recorded by the camera. Very tedious work, it is nevertheless necessary.
This categorization, also called "labeling" represents a set of image detection and analysis methods to allow the automation of a specific task. On average, at least 1,500 and 2,000 tagged images are required to enable this automation of specific tasks. Thus, each artificial intelligence engine located in each camera is unique and made for your horse.
If you want to know more about artificial intelligence, to understand how it works in a more precise way, we invite you to read our article "all about artificial intelligence".