what is binning and its purpose?

I have recently been reading about binning with CCD cameras. What is the
purpose? Say I have a camera that is 1024x768 full resolution and I want to
bin 2x2. My understanding is that the resolution now becomes 640x480, but
the sensitivity increases by 4 times. True? However, isn't my image size
halved with 2x2 binning? therefore, if I wanted the same image size as full
res, I'd have to apply a 200% enlargement and since bicubic and other
algorithms have loss, I lose resolution. Instead of using binning, wouldn't
I get the same result with full res by halving the focal length?

Sorry, binning is a bit confusing to me, and I'm trying to understand the
point of it.

thanks,
Jack

On Sat, 18 Jun 2005 00:03:12 GMT, "Jack Weinstein"
wrote:

I have recently been reading about binning with CCD cameras. What is the
purpose? Say I have a camera that is 1024x768 full resolution and I want to
bin 2x2. My understanding is that the resolution now becomes 640x480, but
the sensitivity increases by 4 times. True?

Actually, that camera will have pixel dimensions of 512x384 when it is
binned 2x2. "Sensitivity" can have different meanings, but it is pretty
reasonable to say the sensitivity is increased 4 times since each pixel
now has 4 times the area, and therefore collects 4 times the light in
the same time.


However, isn't my image size
halved with 2x2 binning? therefore, if I wanted the same image size as full
res, I'd have to apply a 200% enlargement and since bicubic and other
algorithms have loss, I lose resolution. Instead of using binning, wouldn't
I get the same result with full res by halving the focal length?

No, the image size is the same because the sensor size is the same. What
you have changed is the pixel scale: you have half as many pixels along
each axis, but each pixel is twice the size. Basically, you have reduced
your resolution.

Sorry, binning is a bit confusing to me, and I'm trying to understand the
point of it.

A common reason for binning is if your pixels are much smaller than the
optical resolution of your system. For example, my ST8i (with 9um
pixels) has a scale of 0.8 arcseconds/pixel with my LX200. But my
typical seeing is around 3-4 arcseconds, so I'm oversampled. I usually
bin 2x2, giving me 18um pixels and an image scale of 1.6
arcseconds/pixel, which is a better match to my conditions. Each of
these larger pixels is more sensitive, and reading each only adds a
single unit of readout noise. Binning is also used commonly for
collecting the RGB data for a LRGB image, since the binned exposures can
be collected with a shorter exposure time, and the reduced resolution
isn't very obvious in the color channels. It is also useful when framing
an image, since the exposure time is shorter and the download time is
faster.

_________________________________________________

Chris L Peterson
Cloudbait Observatory
http://www.cloudbait.com