KEN BIRMAN: Here they were working on real problems, and no one had ever done this before. How do you build a health app for a cow?
COLBY CASTLE: So this is the feed alley. And then these are just the different pens the cows are in. So they had an idea to mount cameras throughout the barn, and those cameras would be able to use the photo-identification technology. You'd enter the cows ID number, and it would be able to find her wherever she is in the barn.
JULIO GIORDANO: The critical thing is that if you can recognize the cow first, and then you can identify a change in behavior of the animal, then there would be an alert system that would indicate to the farmer that there's this specific animal that needs attention.
KEN BIRMAN: We also had the rough equivalent of Fitbits for the cows. We were getting data about their movements and activity patterns of other kinds. There's information on how they're eating, and what they're choosing eating, how much liquid they're drinking during the day.
COLBY CASTLE: If you're looking for a certain cow, instead of having to walk the entire pen and come back looking for one cow, then you be able to say, oh, even just generally, where she is, you can just head that direction and find her. That's a very useful application, especially on larger dairies, where they have larger pen sizes.
JULIO GIORDANO: Facial recognition, or identification of animals, is very important because farmers need to know who their cows are. And currently, it's done through ear tags or RFID technologies, which are not always great. I mean, they have issues. One of the potential benefits would be to not have to place any sort of device for identification of the animal, but just do it noninvasively.
I think it's important that this is not seen like our students are creating all these new technologies that didn't exist before. These things are already there. The algorithms that are being used or are going to be used are improving all the time. And that's where I think that there's a lot of potential for computer scientists to continue help us develop better algorithms, more efficient ways of using the data.
KEN BIRMAN: The technology is not a rigid thing. It's a tool to solve real problems.
COLBY CASTLE: This technology is not that far off, and it's up and coming. And it's cool being a part of it too.
We've received your request
You will be notified by email when the transcript and captions are available. The process may take up to 5 business days. Please contact firstname.lastname@example.org if you have any questions about this request.
This spring Cornell dairy herd management collaborated with cloud computing students to design digital farming solutions. One result: "A health app for cows."