NHK shows downsized Super Hi-Vision video camera

May 28, 2012 by Nancy Owano report

(Phys.org) -- NHK this week placed on exhibit a shoulder-mount camera, developed in cooperation with Hitachi, capable of shooting what NHK calls super high vision (SHV) video in 7680×4320 resolution. Super Hi-Vision is NHK's preferred name for ultra high definition television (UHDTV). The powerful prototype was part of NHK Scientific & Technology Research Laboratories (STRL) Open House event in Japan earlier this week. The camera is an innovative development, as a compact Ultra High Definition camera using a single-chip color imaging sensor to produce “closest to being there” video. NHK says that the compact head is compatible with commercially-offered still camera lenses.

The prototype is the result of the NHK researchers’ investigation to see if they could produce a portable, lighter-weight Super Hi-Vision camera. They achieved their goal, with the size, weight, and shape of their prototype camera head similar to those of current broadcast cameras.

This video is not supported by your browser at this time.

They found the answer using single-chip color imaging, drawing on results they got in earlier studies. “Single-chip color imaging” refers to a system for acquiring color images with only a single image sensor chip. The single chip sensor uses the Bayer color filter array, so that only one color component is acquired per pixel. An NHK technique involving an “up converter” estimates the other two complementary colors to turn the output into full-resolution video. In turn, NHK has succeeded in showing it is possible to construct a camera creating video with a resolution equivalent to SHV for each of the RGB colors through its up-conversion technique.

In detailing the process, NHK’s STRL said the upconverting is applied to each frame recorded. First, it is conducted for the G color, which uses more pixels than any other . The process interpolates pixels based on the correlation among sets of six missing pixels of each of the longitudinal and lateral lines. Then, based on the obtained green image, an estimation is made for the red and blue pixels.

Next on the research to do list, says NHK: “We hope to improve picture quality and functionality by developing a camera control unit that performs signal processing specifically for this head.”

NHK’s R&D will have another chance to brand itself as ahead of the broadcast technology game during the 2012 Summer Olympics in London, putting Super-Hi Vision to work on public screenings. Opening and closing ceremonies will be among the events filmed by cameras recording 16 times the detail of high definition. In 1964, NHK R&D began on a Hi-Vision (HDTV) system, the same year of the Tokyo Olympic games.

Explore further: 50 Cent, Intel team up on heart-monitor headphones

More information: via DigInfo

Related Stories

NHK shows off a TV that watches you

Jun 01, 2011

Most people are comfortable enough with the idea of watching TV, but how comfortable will they be with the idea of the TV watching them back?

Recommended for you

HP sales inch up while profit drops

3 hours ago

Hewlett-Packard on Wednesday reported that its quarterly revenue was nudged up by improved computer sales, but its profit dropped as the veteran technology firm tried to renew its momentum.

Giant tablets aimed at families

4 hours ago

Costing a little more than an iPad but standing more than twice as tall, a new pair of giant tablets wants families to share cozier group experiences with technology.

Restaurants experimenting with pay-in-advance tickets

6 hours ago

With restaurant patrons increasingly jumping on the Internet to make reservations, some high-end eateries here and across the country are adding a new tech wrinkle: having their clientele pay for their meal in advance using ...

User comments : 6

Adjust slider to filter visible comments by rank

Display comments: newest first

Sonhouse
4.3 / 5 (3) May 28, 2012
Single-chip color imaging refers to a system for acquiring color images with only a single image sensor chip.

Wow, glad they explained that one. Could never have figured that one out by myself:)
Standing Bear
1.6 / 5 (5) May 28, 2012
cute way to do it, but seems like counting animals by enumerating the legs and dividing by four.....cummmbersome! Better to have a three layered sensor chip, but supposed that is 'copyrighted--patented--trademarked' or some such job killing monopolist megacapitalist trick to hold down innovation.
Noumenon
1.7 / 5 (6) May 28, 2012
but supposed that is 'copyrighted--patented--trademarked' or some such job killing monopolist megacapitalist trick to hold down innovation.


Your statement here is remarkably ignorant. Copyrights, patents, and, trademarks, ensure that a company can protect it's rights to profit off of it's innovations. Without which, there is little motivation for financial investment to support such innovation, development, and employment of people to begin with.

Capitalism doesn't "hold down innovation", the absolute exact opposite is the case,... it makes it possible to begin with. My god how this new generation of lefties are stupid.

Hitachi obtains well over 1,000 patents per year, and NHK has a division that just deals in patents.
Noumenon
1 / 5 (3) May 28, 2012
,... because there was an existing patent, other companies are forced to INNOVATE new ways, thus improving technology on and on. Basic lesson over.
LuckyExplorer
1.5 / 5 (2) May 29, 2012
Who the hell needs this resolution exceeding most high end professional photo cameras?
Image size and distance to the image for the best view have such a mismatch that, in praxis, the system is not very useful. It just drives costs and causes bandwidth problems to the transmission infrastructure,...
EWH
not rated yet Jun 15, 2012
The sensor will have to be big to keep the pixels big enough to have low noise in moderate light. At that resolution you are looking at maybe 3 micron pixels at most for a sensor that can be manufactured at reasonable cost, which is smaller than ideal, almost cell-phone quality - 5 microns would be much better. A three-chip design might make more sense, as you get 3x the area for 3x the cost rather than the ~10x the cost it would be to make a single chip with that area. It also would allow 40% smaller image circle lenses, which makes them up to 80% lighter. The lenses will be the limiting factor at such high resolution with reasonably sized chips so there is an optimum pixel size for the system - bigger pixels means less stringent demands on the lens and better noise performance, but also means bigger lenses; smaller pixels means cheaper everything, but worse noise and lower effective resolution for the system once the lens is no longer able to resolve to the level of the sensor chip.