Click it and Unblock the Notifications
Latest Updates
-
Heatwave Alert: Why Your Phone and Laptop Are Overheating Right Now -
National Anti-Terrorism Day 2026: How Rajiv Gandhi’s Assassination Sparked A Nationwide Call For Peace -
International Tea Day 2026: Here's What Drinking Tea First Thing In The Morning Does To Your Gut -
Horoscope for Today May 21, 2026 - Curiosity Rises, Plans Shift -
Paneer Lababdar Recipe: Creamy Restaurant-Style Curry Made Easy -
Mouni Roy’s Cannes 2026 Patola Gown Took 300 Hours To Craft — The Story Of Gujarat’s GI-Tagged Weave -
Bread Pizza Recipe: Your Instant Snack Hack -
India's Hottest City Hit 47.6°C Today — This Is What Heatstroke Looks Like -
Exclusive: Rubina Dilaik Said Yes To The Ward In Seconds: Here's The Raw Truth Behind Why -
PM Modi Turns Viral ‘Melodi’ Nickname Real With Melody Gift To Meloni, Inside India’s Iconic Toffee Origin
A Revolutionary Face Recognition Device
The revolutionary gadget, according to the engineers, work in a simple concept. Lin Huang, from the university's Department of Engineering, says that every face has special features that define people, yet faces can also be very similar. The researcher adds that this is what makes computerized face recognition for security and other applications an interesting, but difficult, task.
According to the researcher, some thing like that - a face recognition softwares have existed for years' now, but the problem with all the earlier days software had many technical glitches. The system, according to them, were accurate, but needed a lot of computer problem, which some how became a big hinderance.
As compared to the earlier face recognition systems, which marked major facial features, like the eyes, nose mouth, the new system, breaking all other prototypes, will have a one-dimensional filter to the two-dimensional data from conventional analyses, such as the Gabor method (which is based on neural networks).
This allows them to reduce significantly the amount of computer power required without compromising accuracy. The team tested the performance of their new algorithm on a standard database of 400 images of 40 subjects. Images are grey scale and just 92 x 112 pixels in size.
They found that their technique was not only faster and worked with low resolution images, such as those produced by standard CCTV cameras, but it also solved the variation problems caused by different light levels and shadows, viewing direction, pose, and facial expressions.
It could even see through certain types of disguises, such as facial hair and glasses.
AGENCIES
Continue with Google