Police photograph and manually tag suspects' tattoos as part of the booking process, but the FBI says computers could classify them much better, leading to more "hits" when trying to identify criminals and also corpses.
For example, one person could label the Detroit Tigers' "D" logo as the team emblem while another might think it's just a stylized "D." So the FBI and the National Institute of Standards and Technology challenged R&D organizations to work on better image recognition systems optimized for tattoos. IEEE Spectrum
reports on the results:
In June, the six groups (that participated in the challenge) reported on how well their algorithms performed in five different types of searches. The algorithms did well in three of these searches, achieving success rates of 90 percent and above in detecting whether a given image contained a tattoo; identifying the same tattoo on the same person, over a span of time; and identifying a small segment of a larger tattoo.
The algorithms performed poorly—with hit rates as low as 15 percent—at two tasks: identifying visually similar tattoos on different people, and searching for similar tattoos across a variety of media, including sketches, scanned prints, and computer graphics.
The tattoo image algorithms are similar to those used in facial and other image-recognition technologies, says Anil K. Jain, a professor of computer science and engineering at Michigan State University, which licensed an algorithm developed by its researchers to MorphoTrak three years ago. The algorithms are all based on extracting key points in an image. However, where fingerprints have ridges and valleys, and faces have eyes and noses, tattoos have no standard features to identify and compare, he says.
"FBI Wants Better Automated Image Analysis for Tattoos" (IEEE Spectrum)
Nissan, to show off its autonomous parking tech, outfitted an inn in Hakone, Japan with “self-parking slippers,” autonomous floor cushions that tidy themselves, and a TV remote control that straightens itself on the coffee table. While obviously a marketing gimmick, self-knolling anything is quite appealing to me. ProPILOT Park Ryokan (Nissan)
Boing Boing pal Eric Paulos, an engineering professor and artist at UC Berkeley, has a history of high-tech provocations, from his early work with machine performance group Survival Research Laboratories to his controversial art installations such as a vending machine for pathogens. Above is the performance/prank Eric recently staged to open his Critical Making class: […]
With Vaunt, Intel is taking steps toward solving the Glasshole paradox: how to get consumers to wear wearables that don’t make wearers seem like bad clichés of wearable users.
Many of us enjoy the aesthetic of vintage electronics, but trying to use most hardware from the 1950’s isn’t necessarily practical. This is especially true where speakers are concerned. While most of us can appreciate the old-school feel of retro speakers, they have a hard time matching the convenience and power delivered by today’s Bluetooth speakers. […]
Python is one of the most popular and versatile programming languages used by developers today, making it an ideal first choice for those looking to kickstart a career in programming. While you could go back to school or sign up for a pricey coding bootcamp, you can learn the essentials of coding with Python at […]
Going back to school isn’t necessarily an option for everyone. Between the time commitments and steep tuition rates, there are obstacles aplenty as far as furthering education is concerned. However, that’s not to say it’s impossible to learn new skills. Excel with Business lets users access thousands of hours of online learning in Microsoft, business, technology, […]