COVID-19 INFORMATION What you need to know about the coronavirus if you are living in Japan or planning a visit.
tech

University develops super-resolution technology for smartphones

6 Comments
By Mayuko Uno

Kogakuin University says it has developed a super-resolution technology capable of playing video on a smartphone in real time.

It is a result of a joint research conducted by Seiichi Gohshi (professor of the Department of Information Design, the Faculty of Informatics, Kogakuin University) and Fujitsu Ltd. The new technology was employed for the "Xevic" image processing engine of Fujitsu's arrows NX F-02H smartphone.

Unlike commonly-used "reconstructed super resolution" and "learning super resolution," the super-resolution technology being researched by Gohshi uses an original method called "nonlinear signal processing method."

"A nonlinear function is applied to the edges detected by applying a high-pass filter to input images," he said. "Then, the data is compressed and added to the original images."

Though it is similar to a conventional enhancer function, it is different in that a nonlinear function is used to supplement high-frequency components, Gohshi said. As a result, it becomes possible to reproduce images with high-resolution components that surpass the "Nyquist frequency (half of a sampling frequency)," which is the theoretical limit of super-resolution.

Also, the new method does not need repeated calculation, which is required for general super-resolution technologies, and uses a simple algorithm. So, it is suited for real-time processing.

The research on the super-resolution based on the nonlinear signal processing method started in 2012 in cooperation with Keisoku Giken Co Ltd. And the technology has already been commercialized as "FE super-resolution unit," a commercial-use device that expands high-vision video data to 4k video data.

This time, the university further simplified the algorithm, and Fujitsu made improvements to task division and time control. As a result, real-time operation was realized by using smartphone software without dedicated hardware. The new technology can play 4k video data with a frame rate of 30fps without increasing the power consumption of a smartphone. The university also confirmed that it is possible to increase the frame rate to 60fps.

The university started the joint research in March 2015 when it received a proposal from Fujitsu.

"We receive many inquiries about the technology from overseas organizations," the university said. "But there are problems related to intellectual property management in some cases, and we hope that it will be used for improving the technological strength of Japanese companies. So, we are dealing with joint researches and developments only with Japanese companies."

The technology is also expected to be used for medical equipment and monitoring cameras.

© Japan Today

©2020 GPlusMedia Inc.


6 Comments
Login to comment

That's gunna be a hot smart phone!

1 ( +1 / -0 )

But there are problems related to intellectual property management in some cases, and we hope that it will be used for improving the technological strength of Japanese companies. So, we are dealing with joint researches and developments only with Japanese companies

I think he wants to say that some foreign companies might rip off the technology but Japanese ones are trustworthy.

2 ( +2 / -0 )

capable of playing video on a smartphone in real time

Can anyone please explain with an example what that means?

1 ( +1 / -0 )

If a Japainese firm rips the tech, it will still be Japain's tech.

I won't mind that much to be cheated by my own brother; to be cheated by YOU will be a different kettle of fish.

0 ( +1 / -1 )

@HighLama

Can anyone please explain with an example what that means?

Almost all videos you watch have gone through some sort of optimization algorithms/processing beforehand. The software libraries that do these sorts of video data optimization are called "codecs". Usually you need the correct codec to A: create a video file of smaller data size, with the same apparent visual clarity or B: play back a video file created by A.

If you take a 4K camera, and tried to pump its video recording output directly to an existing smartphone....the playback would be REALLY sluggish, because the phone needs to run the existing optimization algorithms to render video data of that clarity. I think they are saying most phones get around this by having dedicated circuitry to run these algorithms as fast as possible to enable smooth, immediate playback. But the new algorithms that the university developed means you play 4K video without requiring a dedicated little circuit on the phone's circuitboard, which should reduce fabrication costs and complexity slightly while still delivering the same or better performance for 4K video.

0 ( +1 / -1 )

Japan should patent it, i don't want the korean or any country 4 dat matter to just come and stole it without paying the damn thing.

0 ( +1 / -1 )

Login to leave a comment

Facebook users

Use your Facebook account to login or register with JapanToday. By doing so, you will also receive an email inviting you to receive our news alerts.

Facebook Connect

Login with your JapanToday account

User registration

Articles, Offers & Useful Resources

A mix of what's trending on our other sites