Curve/ball phone-cam tech 'to match professional kit'

American boffins say they have cracked one of the main problems of camera design - the limitations imposed by the use of flat film or photoreceptor arrays. They say that their new curved photoreceptors, modelled on the human retina, could offer mobile phones able to take pictures as good as those from today's bulky professional …

[Colin Mountford]

Solution looking for a problem?

Nowadays they compensate for that kind of problem is done through software compensating for the lens distortion with generally good results.

The main problem with a slim camera is the "film" is not as far from the lens.

This means the image is projected onto a smaller area on the film.

To compensate the density of pixels increases and the pixel size decreases.

As pixel size decreases they essentially become more sensitive to light (I'm sure there is a more technical explanation). This sensitivity means higher error ratios causing more noise in the picture.

So until they solve the noise concern this seems pointless.

Sounds like a solution looking for a problem to me.

Now if he really wants a market he could turn it into a coat and include both sensors and emitters then potentially we have an invisibility cloak!

[adrian lynch]

missing the point

This assumes that the poor quality of phone cameras is due to a technical limitation. Actually the problem is size (as Colin Mountford says) and component quality.

Try comparing a £300 SLR lens with a £10 phone lens. Which one is going to give better results?

Which is why even a dirt-cheap, 3 year old 2megapixel holiday camera will knock the nuts off the best phone camera.

I'm interested in Colin's invisibility cloak though.

[DT]

typical

@Colin Mountford

I don't get the el Reg attitutude of "Uh, progress sucks, because it doesn't look like they've fixed everything in one go.

You're right about the depth of the 'phone limiting the pixel size, but that's only because of the constaints of current lens. However, wouldn't a curved surface have a greater area, facilitating larger pixels (especially if it didn't require mounted at a point focus, a larger CCD) thereby addressing the noise issue?

Anyway, who said this technology was exclusively for camera 'phones or a miraculous cure all? Yes, it's only a concept at the moment , yes it might only be a stepwise improvement. It doesn't mean they shouldn't have bothered because you couldn't think of a useful application for it.

Reduce the number and required precision of lenses in larger cameras could make them cheaper, smaller and with less distortion. What's not to like about that?

[Anonymous Coward]

@Colin Mountfield: distortion vs focus

"Nowadays they compensate for that kind of problem is done through software compensating for the lens distortion with generally good results."

If I'm not mistaken, you're confusing distortion and focus issues. Barrel and pincushion-type distortion can indeed be mitigated by software, but this is a loss of focus at the edges of the image, due to the receptors not being in the focal plane, and a loss of focus is a loss of information for which you cannot compensate.

By the way, it's not the first time I've read about methods for making a curved photoreceptive surface.

[Mike Moyle]

Curved like a retina...?

So, how about an actual artificial retina?

Power and connectivity issues still remain, I know... but if they can get this part working, it puts us that one step closer.

That possibility, alone, should be reason enough to be working on this technology, whether it solves the shorter-term issue of 'phone camera sensitivity or not.