> We also want to show Utrecht’s residents and visitors how much life there is underwater in the canals. The doorbell also provides information on the species and numbers of fish travelling through Utrecht’s waterways. We can use that information to improve the quality of underwater life in Utrecht.
Of course AI could do this without human intervention. They want the public to take part in this project.
In fact, I think that's even the only true goal. Operating the gate is a manual process that takes some effort and time, so it's not like they run out to operate it every time someone pushes the button. In practice, I think they still just open it every once in a while.
“Initially, our fish finder was sold for 600,000 yen per unit, which was as expensive as a house in those days. Despite its high price, the company’s technology attracted fishermen from all over the country. In fact, I heard that a number of customers in the fishing industry visited FURUNO (which was formerly based in Nagasaki City) with backpacks full of cash to buy our fish finder,”
The Furuno Fish Finder is said to be the world's first practical fishfinder; it was introduced by the Furuno brothers for use in commercial fishing vessels in *1948* in Japan.
Considering how still and opaque the picture is (I have just spent 5 minutes staring at the stream), you probably don't even need fancy AI. Some sort of basic movement/change detector will probably do the trick just fine. But as you say the point is to involved people.
Just as I was reading your comment, a boat went by or something and there was a sudden rush of sand and debris kicked up from the bottom of the feed, and for the past 5 minutes or so, there's been a constant stream of debris flying by and swirling around. Looks more like TV static than "still and opaque". Sometimes larger pieces of debris fly by.
But that said, nothing here that proper machine learning couldn't handle.
I think it goes both ways though. One can justify more complex solutions, only to not need the added complexity. One can always ignore edge cases to justify a more simple solution. only to find out that the cut functionality was not quite as "edge" as expected.
My security cameras love to declare that they detected motion because a car drove by and swept its headlights across my porch, and my cameras use a simple change detector.
What are the consequences of false positives? For me, I get a notification, check it, and roll my eyes. For a lock, it opens too often? I have no idea what the consequences of that are. Maybe that's perfectly acceptable and a simple change detector is enough. Or maybe it changes water balance somehow and causes ecological problems, or impacts boat navigation (I mean, the lock is there for a reason in the first place right?)
Sometimes complex solutions are justified. A camera is more complex than just looking into the water with your naked eye. But that necessitates having someone physically present, and able to see through the water. Everyone agrees that the complexity is worth it.
I've worked with too many PMs who want the fastest easy solution without regard for the actual use case and then get perplexed by the side effects (that they were previously told about but refused to grasp). And generally I wind up being the person forced to be physically present watching for fish because the PM thought adding a camera was too much complexity.
Yes, you've happened upon a common trend with new tech. Everyone wants to use their shiny new hammer, some, only know the shiny new hammer, and the wise among them know to just stick to the nails.
It happened with blockchain, it's happening with AI, it'll happen with the next big thing too, just as it has with many big things in the past.
An AI could after being trained, with a higher error level than 100s of humans (or 1000s) watching. Of course, what about a trinity of AIs with quorum?
But anyhow, this sort of project.. even if announced today, likely started a decade ago in someone's mind. According to archive.org, it's been around at least 3+ years. AI wasn't realistic a decade ago, and even 5 years ago? Not really viable reliability wise.
And of course, AI isn't free... they have to be trained, security updates and software updates and hardware updates over time. I mention all of this, because everyone keeps saying "AI could do this", but could it? Because it's not just "can it do it", but "can it do it as reliably and cheaply"?
People are free. And as you say? People taking part is fun.
I'm pretty sure good enough computer vision has been around for decades. The newer stuff is object classification and facial recognition/emotional state type stuff. I think we've had good enough tech to find moving fish for quite a while.
> But anyhow, this sort of project.. even if announced today
It has indeed been on and off since March 2021. This project could've been set up in no time, put a camera underwater, start the live stream, and set up the basic website. Building an AI will take way longer, with much more technical skills required, and much more money.
>"AI could do this", but could it? Because it's not just "can it do it", but "can it do it as reliably and cheaply"?
Yes and also yes, ResNet was released in 2015, VGG in 2014, since then we have developed powerful models that can process a frame in a few ms even on a Raspberry Pi.
> We also want to show Utrecht’s residents and visitors how much life there is underwater in the canals. The doorbell also provides information on the species and numbers of fish travelling through Utrecht’s waterways. We can use that information to improve the quality of underwater life in Utrecht.
Of course AI could do this without human intervention. They want the public to take part in this project.