Demosaicing and bit dimensions

Started Nov 16, 2013 | Discussions thread
John Sheehy Forum Pro • Posts: 19,806
Re: Demosaicing and bit dimensions

Mark Scott Abeln wrote:

But this goes against the trend in smartphone cameras (and Nikon Dx cameras) of oversampling images and demosaicing them to pixel dimensions smaller than the sensor dimensions, or ‘binning’.

Existing cameras of all sensor sizes do not have enough pixels to oversample at typical f-stops. You need to use a pinhole on a DSLR, or a telephoto with stacked TCs to get f-stops that cause over-sampling. Most P&S and cellphone cameras do not allow their apertures to get small enough to cause over-sampling.

Over-sampling, to me, is when you have more than 6 pixels in a line-pair cycle, or it takes more than 4 pixels inclusive to render the sharpest possible optical transient at a useable contrast, and these needs double for independent color channels. Proper sampling is when the discrete capture is virtually analog. We're very far from that, with typical aperture sizes relative to our pixel sizes.

Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow