Communications of the ACM
Nus Engineers Invent Tiny Vision Processing Chip For ltra-Small Smart Vision Systems and Iot Applications
National University of Singapore researchers have announced the development of a microchip that captures visual details from video frames using 20 times less power than existing chips, which runs on a much smaller battery than conventional devices.
Credit: NUS News
Researchers at the National University of Singapore (NUS) say they have developed EQSCALE, a microchip that captures visual details from video frames using 20 times less power than existing chips, and runs on a much smaller battery than conventional devices.
The team says this could result in technologies that are powered continuously by a millimeters-sized solar cell without the need for battery replacement.
They also think EQSCALE could pave the way for cost-effective Internet of Things applications.
The video feature extractor captures visual details taken by a smart camera and reduces them into a smaller set of points of interest and edges for further analysis.
The researchers note EQSCALE also performs continuous feature extraction using just 0.2 milliwatts of electricity, which they say is a major advancement in the level of miniaturization for smart vision systems.
"Energy-quality scaling allows correct object recognition even when a substantial number of points of interests are missed due to the degraded quality of the target," says NUS professor Massimo Alioto.
From NUS News
View Full Article
Abstracts Copyright © 2018 Information Inc., Bethesda, Maryland, USA
No entries found