From: Milton Aupperle <milton@outcastsoft.com>

Date: December 6, 2007 9:28:47 AM MST

To: Astro_IIDC@yahoogroups.com

Subject: Re: [Astro_IIDC] The sharpness curve

On 6-Dec-07, at 12:28 AM, Duane wrote:

Tonight's session leads me to a question. I always have AIIDC set to show me the graph 

after it determines sharpness. Most of the time this chart looks like a standard bell curve 

with room on either side. I've had it do other things (rarely) like put a block of images on 

the far left and a stream of images running off to the right, dribbling off at the end. This 

was a night of better seeing.

I have a whole series of curves described in the Astro IIDC manual with an explanation for each type on pages 31 to 34.

In any case, does this graph ever give a hint as to how good the seeing was by its shape 

and is there an ideal place to make the selection (% vs. #) of images to stack (such as, right 

where the images drop off dramatically or where there is a little plateau on the sharp side 

of the bell curve, or where there happens to be a large spike in numbers on the sharp 

side)?

Yes, that's described in the manual too, for each curve, from Type A  to Type E Curves.

What your not grasping is that the curve is relative sharpness between frames, it's not an absolute number. 

Quote from the manual

The base assumption with interpreting histograms and estimating sharpness is that some  frames will be sharper than others. If the image is out of focus and consistently blurry to start  with, it may give you the "Type A" or "Type B" histograms.

You get an indication of how consistent the frame sharpness is is by looking at the Min - Max value range, the STD68 or STD96 standard deviation values and how skewed the curve is. The larger the Min - Max range, the more variance in frame sharpness and the higher you want to increase the Confidence Interval Slider. The Skew of the curves is indicated by comparing the Mean and Median values and if they are the exact same, the histogram is not skewed. Visually a skewed histogram has a long tail or long nose and the peak of the hump is shifted to towards the low or the high end of the histogram. The skew direction tells you whether you have more sharp frames than unsharp frames and you want to use a Confidence Interval that selects (ie highlight in red) the sharpness bars that are to the right  of the central hump. The histograms you will encounter lie in the following main types, however they they will grade into each other. 

End Quote

The other issues is Camera Noise due to using High Gains. That WILL influence the Sharpness Histograms as it's perceived as high frequency data and it will mask other features. And because it's Random, it may affect some frame more than other frames, which affects the frames sharpness value for each frame. To suppress Noise, I've always recommended check marking the 

"Gaussian blur the frames used for sharpness estimation & alignment"

checkbox (see page 28 of the Astro IIDC manual), which will apply a 1.4 pixel radius light blur to the frames used for sharpness estimate and alignment.

Milton J. Aupperle