Man Buys Two Metric Tons of LEGO Bricks; Sorts Them Via Machine Learning

iStock // Ekaterina Minaeva
iStock // Ekaterina Minaeva

Jacques Mattheij made a small, but awesome, mistake. He went on eBay one evening and bid on a bunch of bulk LEGO brick auctions, then went to sleep. Upon waking, he discovered that he was the high bidder on many, and was now the proud owner of two tons of LEGO bricks. (This is about 4400 pounds.) He wrote, "[L]esson 1: if you win almost all bids you are bidding too high."

Mattheij had noticed that bulk, unsorted bricks sell for something like €10/kilogram, whereas sets are roughly €40/kg and rare parts go for up to €100/kg. Much of the value of the bricks is in their sorting. If he could reduce the entropy of these bins of unsorted bricks, he could make a tidy profit. While many people do this work by hand, the problem is enormous—just the kind of challenge for a computer. Mattheij writes:

There are 38000+ shapes and there are 100+ possible shades of color (you can roughly tell how old someone is by asking them what lego colors they remember from their youth).

In the following months, Mattheij built a proof-of-concept sorting system using, of course, LEGO. He broke the problem down into a series of sub-problems (including "feeding LEGO reliably from a hopper is surprisingly hard," one of those facts of nature that will stymie even the best system design). After tinkering with the prototype at length, he expanded the system to a surprisingly complex system of conveyer belts (powered by a home treadmill), various pieces of cabinetry, and "copious quantities of crazy glue."

Here's a video showing the current system running at low speed:

The key part of the system was running the bricks past a camera paired with a computer running a neural net-based image classifier. That allows the computer (when sufficiently trained on brick images) to recognize bricks and thus categorize them by color, shape, or other parameters. Remember that as bricks pass by, they can be in any orientation, can be dirty, can even be stuck to other pieces. So having a flexible software system is key to recognizing—in a fraction of a second—what a given brick is, in order to sort it out. When a match is found, a jet of compressed air pops the piece off the conveyer belt and into a waiting bin.

After much experimentation, Mattheij rewrote the software (several times in fact) to accomplish a variety of basic tasks. At its core, the system takes images from a webcam and feeds them to a neural network to do the classification. Of course, the neural net needs to be "trained" by showing it lots of images, and telling it what those images represent. Mattheij's breakthrough was allowing the machine to effectively train itself, with guidance: Running pieces through allows the system to take its own photos, make a guess, and build on that guess. As long as Mattheij corrects the incorrect guesses, he ends up with a decent (and self-reinforcing) corpus of training data. As the machine continues running, it can rack up more training, allowing it to recognize a broad variety of pieces on the fly.

Here's another video, focusing on how the pieces move on conveyer belts (running at slow speed so puny humans can follow). You can also see the air jets in action:

In an email interview, Mattheij told Mental Floss that the system currently sorts LEGO bricks into more than 50 categories. It can also be run in a color-sorting mode to bin the parts across 12 color groups. (Thus at present you'd likely do a two-pass sort on the bricks: once for shape, then a separate pass for color.) He continues to refine the system, with a focus on making its recognition abilities faster. At some point down the line, he plans to make the software portion open source. You're on your own as far as building conveyer belts, bins, and so forth.

Check out Mattheij's writeup in two parts for more information. It starts with an overview of the story, followed up with a deep dive on the software. He's also tweeting about the project (among other things). And if you look around a bit, you'll find bulk LEGO brick auctions online—it's definitely a thing!

Mapping Technology Reveals 'Lost Cities' on National Geographic

Lin uses his iPad to visualize scanning data of a crusaders' fortress at the lagoon in Acre, Israel.
Lin uses his iPad to visualize scanning data of a crusaders' fortress at the lagoon in Acre, Israel.
Blakeway Productions/National Geographic

Imagine what Pompeii looked like before the lava hit, or Mayan pyramids before the jungle took over. In the past decade, scientists have been able to explore human settlements long since abandoned by using a new wave of accessible technology. Instead of needing an expensive plane and crew to fly aerial sensors, for example, explorers can mount them on cheaper drones and pilot them into previously unreachable areas. The resulting data can tell us more about the past, and the future, than ever before.

That’s the premise of Lost Cities with Albert Lin, a new TV series premiering on National Geographic on Sunday, October 20.

Lin, an engineer and National Geographic Explorer, uses cutting-edge tools to shed light on centuries-old cities in the most beautiful places on Earth. Ground-penetrating radar reveals buried structures without disturbing the landscape. A drone-mounted remote sensing method called LIDAR—short for "Light Detection and Ranging"—shoots lasers at objects to generate data, which Lin visualizes with 3D mapping software. The results suggest what the ruins probably looked like when they were new.

Albert Lin and crew in Peru
Thomas Hardy, Adan Choqque Arce, Joseph Steel, Duncan Lees, Albert Lin, and Alonso Arroyo launch the LIDAR drone at Wat'a in Peru.
National Geographic

“It’s like a window into a world that we’ve never had before,” Lin tells Mental Floss. “It’s shooting millions of laser pulses per second through a distance of air. By digitally removing the top layer of everything above the ground—trees, brush, cacti—you’re washing away the past. All of the sudden you’re left with these fingerprints—experiments in how we organized ourselves through time.”

For the six-episode series, Lin and the expert storytelling team were dispatched to the South Pacific, the Middle East, the Andes, the Arctic, and other destinations. Lin explains that while most of the sites are known to archaeologists, they’ve never been so precisely mapped in three-dimensional detail.

In the first episode, Lin travels to Nan Madol, an enigmatic complex of temples and other structures on the Micronesian island of Pohnpei. With the help of local researchers and indigenous leaders, Lin and the team scan the ruins and digitally erase trees, water, and forest undergrowth to unveil the complex's former grandeur.

“Technology and innovation have always been that gateway to go beyond the threshold, and see what’s around the corner,” Lin says. “Seeing these worlds for the first time since they were left, it’s almost like reversing the burning of the library of Alexandria. We can take the synthesis of knowledge of all these watershed moments of our human journey, and imagine a better future.”

Lost Cities With Albert Lin premieres Sunday, October 20 at 10/9c and resumes on Monday, October 21 at 10/9c on National Geographic.

Samsung Is Fixing Bug That Lets Anyone Unlock Fingerprint-Protected Galaxy S10s

stevanovicigor/iStock via Getty Images
stevanovicigor/iStock via Getty Images

Users of the Samsung Galaxy S10, fear not: A fix is on the way for your device’s faulty fingerprint reader. According to Engadget, Samsung Electronics told Reuters in a statement that it is “aware of the case of the S10’s malfunctioning fingerprint recognition and will soon issue a software patch.”

Soon after the device’s initial release, consumers discovered that a 3D-printed fingerprint could unlock the phones. Then, UK user Lisa Neilson reported to The Sun that any human fingerprint would work on her phone. Though we don’t know exactly why the technology is malfunctioning, it has happened through the use of third-party plastic or silicon screen protectors—Neilson had purchased hers on eBay.

“It’s a real concern,” Neilson told The Sun, in large part because people nowadays store much more than contacts and photos on their smartphones. “Anyone can access it and could get into the financial apps and transfer funds.”

Samsung users can protect their phone's security with an old-fashioned number code. Though not as hassle-free as the fingerprint mechanism, the code will work just fine.

The company hasn’t disclosed when we can expect a solution, though it confirmed to Engadget that an internal investigation is underway and customers should stick to using authorized Samsung products in the meantime, rather than third-party screen protectors.

[h/t Engadget]