Machine learning masters the fingerprint to fool biometric systems

Fingerprint authentication systems are a widely trusted, ubiquitous form of biometric authentication, deployed on billions of smartphones and other devices worldwide. Yet a new study reveals a surprising level of vulnerability in these systems. Using a neural network trained to synthesize human fingerprints, the research team evolved a fake fingerprint that could potentially fool a touch-based authentication system for up to one in five people.
Machine learning masters the fingerprint to fool biometric systems

The subtle science of wok tossing

Wok tossing is essential for making a good fried rice — or so claim a group of researchers. The researchers became interested in the physics of cooking and, seeking to satiate their curiosity, they focused on stir-fry, which lies at the heart of Chinese cuisine. The team set out to isolate the key elements of this venerable cooking technique.
The subtle science of wok tossing

Prototype of robot dog nose

Every day, thousands of trained K9 dogs sniff out narcotics, explosives and missing people. These dogs are invaluable for security, but they’re also expensive. Researchers have made the beginning steps toward an artificial ‘robot nose’ device that officers could use instead of dogs. The heart of the system would be living odor receptors grown from mouse genes that respond to target odors, including the smells of cocaine and explosives.
Prototype of robot dog nose