facebook rss twitter

Is the perfect lens; just what the industry is crying out for?

by Alistair Lowe on 4 September 2012, 10:45

Quick Link: HEXUS.net/qabluv

Add to My Vault: x

Currently, when we think of buying a high-quality lens, we think of SLRs, of firms such as Canon and Nikon and the huge price-tags associated, which can run into the many thousands, though, exactly why do we look to these high-quality lenses?

When all is said and done, we're not exactly short on megapixels, even when you look at a modern camera phone. LG, for example, is about to launch the Optimus G, which features a 13-megapixel sensor - something in the range of what you would have expected of SLR cameras but a few years ago. OmniVision, a popular maker of camera sensors, is already offering 16-megapixel sensors with 20-megapixels on the way. Likewise, technology has pushed forwards, with Back Side Illumination (BSI) reducing sensor noise and so, why do we not expect to see the same results from a smartphone's camera as we do with professional kit?

One part of this explanation is that SLRs typically feature three sensors, one for each primary colour. Other cameras and smartphones feature colour-filters on a single sensor, creating a slight mismatch between the brightness and the colour on an individual pixel. As megapixels grow, however, this is becoming less and less of an issue as down-sampling can help to realign this disparity. This is exactly what Nokia does with its epic 41-megapixel PureView sensor, with the camera spitting out 5-megapixel images that are near-enough free of this issue.

In order to make PureView happen, however, Nokia had to go to extreme lengths to produce a lens with as little distortion as possible. Whilst we can increase the megapixel count on sensors, it's useless if we can't ensure that the light reaching them is distortion-free. As you increase the pixel count, you require less and less distortion of the light. In Nokia's case, this resulted in a rather bulky lens, which, despite its technological advances, still isn't quite capable of producing SLR-class imagery. Likewise, many smartphone sensors are let down by inferior lenses, due to the requirements for them to be cheap and compact.

Harvard Distortion-free lens

Here is where Harvard University wishes to step in. Boffins at the University have developed a new kind of lens, fashioned from silicon and gold, which, according to its creators, is distortion-free.

The lens is created from a near-flat 60-nanometre-thick wafer of silicon, upon the surface of which is an array of gold nano-antennas, capable of funnelling light at a range of focal points by introducing a variety of photon delays through manipulation of gold particle size. This causes light to bend at different rates across the lens, right on the surface, as opposed to guiding it progressively, as a large glass lens would.

The technology has the potential to replace the vast majority of large, heavy and expensive glass lenses we use today and would be ideal for the smartphone market. By capturing light on a flat-plane, the original image is almost entirely undistorted and the thin substrate reduces noise introduced through imperfections.

We can't wait for this one to hit mass market production!

HEXUS Forums :: 7 Comments

Login with Forum Account

Don't have an account? Register today!
That sounds awesome but like all the other awesome things Universities are doing I expect a long wait before it gets to market.
Interesting, though of course this does nothing to address the issues of optical zoom and aperture control (IMHO FAR more important than ‘megapixel count’), which for me is the key difference between an SLR and a compact camera of any type. Us amateur photographers are all-too-well acquainted with the idea that there is no perfect lens: no lens that can go from 10-800mm (full frame equivalent) across f/1.0 to f/32 with no optical distortion, weighing less than 100g. That's why many of us have multiple lenses to cover the main bases: a decent zoom lens, a wide-angle lens, a macro lens, a wide-aperture prime (like a 50mm plastic fantastic…) and so on.

For me, the ability to control the depth of field and shutter speed with a high quality wide-open aperture is the most important factor.

With that said, all advances that improve the quality of smartphones and compact cameras are welcome!
Digital SLRs do not typically feature separate sensors for each primary colour; the only current SLRs of this type are made by Sigma. All other digital SLRs use a Bayer matrix filter just like compact and ‘phone cameras.
In terms of image quality, other than the choice of lenses, one of the main feature that distinguishes a digital SLR from compact and ’phone cameras is that the SLR wil have a far larger sensor for a given pixel count. This means each individual sensor photosite (pixel) is much larger and hence less succeptible to noise as each photosite “collects” more light.
Digital SLRs do not typically feature separate sensors for each primary colour; the only current SLRs of this type are made by Sigma. All other digital SLRs use a Bayer matrix filter just like compact and 'phone cameras.

Well a Bayer matrix is separate photosites for each colour :p

But your point is correct that camera phones and SLRs use exactly the same type of sensor for the most part.

Back to the article.. yes, lots of people have been trying to do exactly this for ages.
Judging by that 1 micron magnification, could this be used, in conjunction with some sort of smart controller, to negate slight but predictable scene elements movement with longer exposures or camera shake with shorter? If possible, I can see this technology paving the way for a completely new era in digital photography, from P&S to DSLR and even beyond (like astronomy telescopes and digital optical microscopes). How fast can a phase shift be changed through the entire surface and how precise would the steps be?