[opendtv] Re: Image quality
- From: Craig Birkmaier <craig@xxxxxxxxx>
- To: opendtv@xxxxxxxxxxxxx
- Date: Tue, 15 Nov 2005 09:00:00 -0500
At 2:28 AM +0100 11/13/05, Olivier Houot wrote:
>Some experiments bypassing the eye optics using laser interferences
>have measured the resolution of the retina to be around 150 cpd .
>Since the optics blur that down to 30 cpd it suggests a pont-like
>source is spread over several sensitive cells (through diffraction,
>diffusion and aberrations).
There is an additional process in human vision that in fact
correlates with what Oliver describes above.
From section 3.2.1 of the Task Force Report on Digital Image Architecture:
>Although high spatial resolution vision is restricted to the fovea,
>the visual system acquires high resolution images over a wide
>portion of the field of view. This is achieved through involuntary
>eye movements; high frequency tremor, slow drift, and rapid saccade.
>
>Research has determined that it takes several hundred milliseconds
>for the eye to acquire a high spatial resolution image, synthesized
>from a number of overlapping views. Slow drift and rapid saccade are
>the mechanisms used for repositioning the fovea to acquire these
>multiple impressions. The tremor appears to be a mechanism to remove
>high frequency spatial noise. The tremor's oscillation occurs at a
>frequency range of 40 to 80 Hz over an area approximately equal to
>the size of a single cone.
So the human visual system actually oscillates to spread the photons
out over several sensors when acquiring a high resolution view. This
can accomplish two things:
- A reduction in noise (i.e. sampling errors)
- A reduction in gaps between the sensors that could cause gaps in
the acquired image.
With electronic image sensors we use tiny lenses on the chips to
focus light from a larger area onto the individual sensors. This
process pulls in photons that might otherwise have hit the gaps
between the sensors. And we use optical low pass filtering to spread
out details that are smaller than a single sample so that they do not
simply disappear. This filtering contributes to the loss of contrast
in these high frequency details - I believe this is one of the
factors that influences the modulation transfer function of the
system.
The human visual system appears to have some similar attributes to
help it capture high frequency details with low SNR.
>Besides, the various eye automatic movements alone would allow the
>spot to cross through several cells.
>
> So once it is no longer possible to resolve the individual sources,
>the only limit remaining is the combined power vs distances of the
>sources.
>
>And since the human rods have been shown to be sensitive to a single
>quantum of light, you could see a supernova tens thousands of light
>years away, and a hypernova even further (except those tend to favour
>gamma ray output)
Now go out on a clear night and look up into the sky why moving your
head constantly. Only the brightest stars will be discernable. Allow
the eye to remain fixed on one region of the sky and you will see
many more stars as the photons trickle in...
Regards
Craig
----------------------------------------------------------------------
You can UNSUBSCRIBE from the OpenDTV list in two ways:
- Using the UNSUBSCRIBE command in your user configuration settings at
FreeLists.org
- By sending a message to: opendtv-request@xxxxxxxxxxxxx with the word
unsubscribe in the subject line.
Other related posts:
