Feast Your Eyeballs On This Mechanical Eyeball

Most of us, if we have bought a single board computer with the capability  to support a camera, will have succumbed to temptation and shelled out for that peripheral in the hope that we can coax our new toy into having sight. We’ll have played with the command line tool and taken a few random images of our bench, but then what? There is so much possibility in a camera that our colleague [Steven Dufresne] wanted to explore with his Raspberry Pi, so he built a motorised eyeball mount with which to do so.

Pan & tilt mounts using RC servos are nothing especially new, but in this one he’s put some design effort that maybe some of the others lack. A lot of effort has gone in to ensuring no interference between the two axes, and in a slightly macabre twist until you remember it’s a model he’s talking about, the unit has been designed to fit inside a human head.

The servos are driven from the Pi using a servo driver board he’s discussed in another video, so once he’s described the assembly with a few design tweaks thrown in he has a quick look at the software demo he’s written emulating neurons for eye tracking. He promises that will be put up somewhere for download in due course.

If you’re in the market for a pan & tilt mount for your Pi, this one could make a lot of sense to throw at your 3D printer. It’s certainly more accomplished than this previous one we’ve shown you.

Continue reading “Feast Your Eyeballs On This Mechanical Eyeball”

Raspberry Pi On The Go Powers Car System

Most new cars have GPS, rear cameras, and all the other wonders an on-board system can bring. But what if you have an old car? [Fabrice Aneche] has a 2011 vehicle, and wanted a rearview camera. He started with a touch screen, a Raspberry Pi 3, and a camera. But you know how these projects take on a life of their own. So far, the project has two entries in his blog.

It wasn’t long before he couldn’t resist the urge to add a GPS. But that’s no fun without maps. Plus you need turn-by-turn directions. [Fabrice] did a lot of the user interface using Qt5 and QML. He started out running it with X11 but that was slow. It turns out though that Qt5 can drive the Pi’s video directly without using X11, so that’s what he wound up doing. The code that isn’t in QML — mainly dealing with the GPS location — is written in Go, while the code for MOCS (My Own Car System) is on GitHub.

Continue reading “Raspberry Pi On The Go Powers Car System”

Object Detection, With TensorFlow

Getting computers to recognize objects has been a historically difficult problem in computer science, but with the rise of machine learning it is becoming easier to solve. One of the tools that can be put to work in object recognition is an open source library called TensorFlow, which [Evan] aka [Edje Electronics] has put to work for exactly this purpose.

His object recognition software runs on a Raspberry Pi equipped with a webcam, and also makes use of Open CV. [Evan] notes that this opens up a lot of creative low-cost detection applications for the Pi, such as setting up a camera that detects when a pet is waiting at the door to be let inside or outside, counting the number of bees entering and exiting a beehive, or monitoring parking spaces at an office.

This project uses a number of other toolkits as well, including Protobuf. It also makes extensive use of Python scripts, but if you’re comfortable with that and you have an application for computer vision, [Evan]’s tutorial will get you started.

Continue reading “Object Detection, With TensorFlow”

Need A Tiny CRT? Karaoke Might Just Help

[Brett] is working on a video installation, and for the past few months, has been trying to get his hands on tiny CRTs any way he can. After initially coming up short, he happened across a karaoke machine from 2005, and got down to work.

Karaoke machines of this vintage are typically fairly low-rent affairs, built cheaply on simple PCBs. [Brett] found that the unit in question was easy to disassemble, having various modules on separate PCBs joined together with ribbon cables and headers. However, such machines rarely have video inputs, as they’re really only designed to display low-res graphics from CD-G format discs.

While investigating the machine, initial research online proved fruitless. In the end, a close look at the board revealed just what [Brett] was looking for – a pin labeled video in! After throwing in a Raspberry Pi Zero and soldering up the composite output to the karaoke machine’s input pin, the screen sprung to life first time! This initial success was followed by installing a Raspberry Pi 3 for more grunt, combined with a Screenly install – and a TRS adapter the likes of which we’ve never seen before. This allows video to be easily pushed to the device remotely over WiFi. [Brett] promises us there is more to come.

Karaoke is a sparse topic in the Hackaday archives, but we’ve seen a couple builds, like this vocal processor. If you’ve got the hacks, though? You know where to send ’em.

Hackaday Belgrade: Luka Mustafa On Exploiting IoT Niches

Ecology is a strange discipline. At its most basic, it’s the study of how living things interact with their environment. It doesn’t so much seek to explain how life works, but rather how lives work together. A guiding principle of ecology is that life finds a way to exploit niches, subregions within the larger world with a particular mix of resources and challenges. It’s actually all quite fascinating.

But what does ecology have to do with Luka Mustafa’s talk at the 2018 Hackaday Belgrade Conference? Everything, as it turns out, and not just because Luka and his colleagues put IoT tools on animals and in their environments to measure and monitor them. It’s also that Luka has found a fascinating niche of his own to exploit, one on the edge of technology and ecology. As CEO of Institute IRNAS, a non-profit technology development group in Slovenia, Luka has leveraged his MEng degree, background in ham radio, and interest in LoRaWAN and other wide-area radio networks to explore ecological niches in ways that would have been unthinkable even 10 years ago, let alone in the days when animal tracking was limited by bulky radio collars.

Continue reading “Hackaday Belgrade: Luka Mustafa On Exploiting IoT Niches”

GTA: San Andreas Radio Earns Six-Star Wanted Level

[Raphaël Yancey] wanted to be able to jam to Bounce FM and Radio:X all the time, without having to steal a car or a street sweeper in San Andreas. As people who like to put on the sad piano building music from The Sims and write Hackaday posts, we can totally relate.

But this isn’t just another one of those jam-a-Pi-into-a-vintage-radio-and-call-it-a-sandwich projects (not that there’s anything wrong with those). This thing acts like a real radio. All the stations play continuously whether you’re tuned in or not, and they bleed into each other as you go up and down the dial.

After much trial and error, [Raphaël] found a Python mixer that would work, but it was no longer maintained. He forked it, squashed a bug or two, and wrote a module for KY040 rotary encoders to make them play nice with the Pi. The snake charming doesn’t stop there: the rock star of this project is [Raphaël]’s virtual radio software, which handles the audio blending as he tunes between stations. A step-by-step tutorial is coming soon, so watch [Raphaël]’s site for updates. Tune past the break to give it a listen.

Adventures in Raspi radio-ing don’t have to be one-way. Here’s how you can turn one into an AM/FM+ transmitter using a DVB-T dongle and SDR.

Continue reading “GTA: San Andreas Radio Earns Six-Star Wanted Level”

Old Time Traffic Signal Revived With A Raspberry Pi Controller

Anyone with even a passing familiarity with the classic animated shorts of the 1940s will recognize the traffic signal in the image above. Yes, such things actually existed in the real world, not just in the Looney world of [Bugs Bunny] et al. As sturdy as such devices were, they don’t last forever, though, which is why a restoration of this classic Acme traffic signal was necessary for a California museum. Yes, that Acme.

When you see a traffic signal from the early days of the automotive age like this one, it becomes quickly apparent how good the modern equivalent has become. Back in the day, with a mix of lights distributed all over the body of the signal, arms that extend out, and bells that ring when the state changes, it’s easy to see how things could get out of hand at an intersection. That complexity made the restoration project by [am1034481] and colleagues at the Southern California Traffic Museum all the more difficult. Each signal has three lights, a motor for the flag, and an annunciator bell, each requiring a relay. What’s more, the motor needs to run in both directions, so a reversing relay is needed, and the arm has a mechanism to keep it in position when motor power is removed, which needs yet another relay. With two signals, everything was doubled, so the new controller used a 16-channel relay board and a Raspberry Pi to run through various demos. To keep induced currents from wreaking havoc, zero-crossing solid state relays were used on the big AC motors and coils in the signal. It looks like a lot of work, but the end results are worth it.

Looking for more information on traffic signal controls? We talked about that a while back.