Skip to main content
  1. Home
  2. Emerging Tech
  3. Computing
  4. Smart Home
  5. News

DIY backyard security system turns on the sprinklers when it sees a trespassing cat

Add as a preferred source on Google

What’s the best way to stop cats from pooping in your garden? Depending on who you ask, answers might include ultrasonic cat repellents, chicken wire, or the use of certain plants to protect your lawn from feline defecation.

For Nvidia Systems Software Engineer Robert Bond, however, there was only ever one possible solution: training a deep learning neural network to turn on his garden sprinklers whenever it recognized a cat in his front yard.

Recommended Videos

“I mounted a camera on the garage that looks out over the garden,” Bond tells Digital Trends. “The camera is set to send images to an [Nvidia Jetson TX1] whenever something changes. The [TX1] then uses deep learning to decide whether the thing that moved is a cat or something else. If it’s a cat then it activates the sprinkler system.”

catsprinkler1
Image used with permission by copyright holder

Bond says that he started training his neural network with a data set of around 1.2 million pictures gathered from the internet. These included plenty of pictures of cats, but most of them turned out to be cutesy close-ups portraits, rather than the kind of pictures he was looking for. “It was almost worthless for what I needed to do, because I was hunting for pictures of cats slinking around in my bushes,” he explains.

Bond notes that at this point his neural network was only successfully classifying around 30 percent of the cats that came into his garden. He decided to modify the neural network’s training examples by adding pictures he had taken himself. Recognition scores immediately jumped to around 90 percent.

catsprinkler2
Image used with permission by copyright holder

The system hasn’t been perfect, however. For one thing, during the training process shadows were regularly classified as cats because they represented dark shapes moving across Bond’s lawn. Even Bond wasn’t immune from being classified as a feline. “I went to mow my lawn and the system somehow decided that the combination of me and my lawnmower counted as a cat,” he says. “I wound up getting pretty thoroughly sprayed.”

But for the most part the deep learning network has been a resounding success — and Robert Bond’s garden is now virtually cat free.

It’s simple when you know how, right?

Luke Dormehl
I'm a UK-based tech writer covering Cool Tech at Digital Trends. I've also written for Fast Company, Wired, the Guardian…
Anti-surveillance clothing is getting cheaper, but don’t expect an invisibility cloak
Affordable shirts now claim to confuse facial recognition, although their protection depends heavily on the camera and software watching you
Chart, Plot, Adult

Anti-surveillance clothing is starting to look less like an art-school experiment and more like something you could actually wear outside. Shirts designed to confuse facial recognition systems now cost about as much as ordinary streetwear, although buying one won’t make you disappear.

The Guardian reports that designers are using face-like prints, unusual cuts and infrared lights to interfere with computer vision. These techniques target specific weaknesses, so their success depends on what happens to be watching you.

Read more
This spinning drone hides in plain sight using a visual illusion
This drone doesn't turn invisible. It tricks your brain into thinking it has.
Phantom Twist

For decades, engineers have chased the dream of an invisible drone. The usual approaches have involved transparent materials, camouflage coatings, or complex optical systems that bend light around an object. Researchers at Northwestern University decided to take a completely different route. Instead of hiding the drone itself, they chose to fool the human eye.

The result is Phantom Twist, an experimental drone that spins so rapidly it almost disappears into the background. It's not technically invisible, but to anyone watching, it looks more like a faint blur than a flying machine.

Read more
This smart knitted fabric can flip switches, count your steps, and even change shape
Grandma's knitting just entered its Iron Man era
Representative Image

For most of us, knitting brings to mind sweaters, scarves, and perhaps an ambitious grandmother determined to make winter more fashionable. Researchers at Harvard University, however, have a far more futuristic vision. They've transformed ordinary knitted fabric into a programmable material capable of changing shape, acting as an electrical switch, sensing movement, and potentially forming the foundation of tomorrow's wearable technology.

The research, published in Advanced Functional Materials by scientists at the Harvard John A. Paulson School of Engineering and Applied Sciences (SEAS), demonstrates how machine-knitted textiles can "snap" between multiple stable shapes without relying on motors or rigid mechanical parts.

Read more