By Tamar Avraham, Ph.D.
AI lifeguard : Before we dive in…
If you unlocked your smartphone or authenticated a purchase just by looking at your screen or avoided a hazard while driving your car, it’s likely you were benefiting from visual-based AI!
These days, artificial intelligence is something we all use every single day. It’s proving useful, not just in discovering new shows on Netflix or getting product recommendations on Amazon, but also by adding extra layers of security and safety in our everyday lives.
Most standard pool alarms today feature “yesterday’s” technology – motion sensors and mechanical floats.
In this article, we’re going to explore how AI is being used in next-generation pool alarms like MYLO, making our home pools a whole lot safer.
MYLO is “AI-powered”. What does that even mean?
Good idea, let’s start with the basics. Artificial intelligence is the simulation of human intelligence by computers.
We call MYLO the “AI-powered lifeguard” because it uses artificial intelligence to mimic human behaviors (particularly that of a human lifeguard), analyze large amounts of data, and make decisions about whether someone inside a pool may be at risk of drowning.
My pool alarm seems smart, why do I need an AI Lifeguard?
Let’s start with what makes your pool alarm “smart” (but not so intelligent!)
Most pool alarms, possibly the ones you own, use motion sensors that are designed to detect excess water movement and issue an alarm if someone or something unexpectedly falls into the water.
It’s smart enough to sense, but it’s not intelligent enough to see what happening in the pool. As a blind system, pool alarms based on sensors are limited to alerting you to entry into an empty pool only. And, they are plagued by false alarms triggered by wind or pool cleaning robots.
By contrast, vision-based AI like MYLO uses a combination of cameras and algorithms to see what’s happening in a pool that’s being used by people.
Mmmh, sounds like MYLO is similar the AI safety system in my car…
Great observation! The AI in your MYLO AI-powered Lifeguard works in somewhat similar fashion to vision-based car safety systems, such as MobileEye, that also use cameras and algorithms to scan the road landscape, detecting potential collisions and accidents.
In a not-too-dissimilar way, MYLO uses its dual cameras and drowning detection algorithms to visually scan the pool landscape, above and below the water, to detect any behavior that indicates a drowning risk.
And, raising an audible and visual alarm if needed to alert you so you can act.
Can AI distinguish between someone potentially at risk of drowning and someone just messing around?
Another spot-on question. The MYLO AI has learned (or more accurately, it’s been trained) to do just that. The powerful patented algorithms used by MYLO, and developed by its founders, have been trained using millions of video and static images to identify specific drowning behaviors. This training has also taught MYLO to be able to identify between humans and objects (that’s why MYLO, unlike most swimming pool alarms, won’t trigger a false alarm when it sees your pool cleaner robot on your pool floor). This is extremely sophisticated and powerful technology. What’s more MYLO has been trained to monitor multiple swimmers or pool users at the same time, and detect which behaviours are normal and which are suspicious.
Closing Thoughts
Pool safety technology has made amazing leaps in the last few years.
Next-generation pool alarms like MYLO are using vision-based artificial intelligence to give pool owners never-before-possible layers of protection, such as continuous in-pool monitoring and detection of suspicious activity that may indicate a silent drowning or swimmer in distress.
Most of us don’t have the luxury of employing a lifeguard at home. In the meantime, MYLO’s AI-powered lifeguard can certainly add important extra layers of protection. That’s great news for families with young kids and home pool owners.
About the author
Tamar Avraham, Ph.D., is an expert in computer vision and AI, with more than 20 years of experience in both academic research and development of machine learning algorithms and software. She is the co-founder of Coral Smart Pool and co-inventor of MYLO, the AI-powered lifeguard.