From: Milton Aupperle <milton@outcastsoft.com>

Date: May 9, 2009 10:01:19 PM MDT

To: Astro_IIDC@yahoogroups.com

Subject: Re: [Astro_IIDC] processing Movies.. [2 Attachments]


Mark;


On 9-May-09, at 6:52 PM, Mark Gaffney wrote:

<*>[Attachment(s) from Mark Gaffney included below]


Hi Milton, Firstly I`ve clicked off  "create blended MAP image from  

manually selected pixel areas". I`ve selected a movie taken with the  

Flea 2 & my Mogg 0.3 FR. I`ve chosen a 70% CI & gotten 77 sharpest  

frames. I take it 000001 1.32347 is the sharpest frame & is the image  

in the preview window beside. I think I`ve mastered the art of  

choosing selection areas, deleting & moving them with the blue  

indicator box `til they`re at least 10.0 pixel variance. However,  

after clicking on & highlighting other items from the sharpness list,  

both individually (then clicking "include") or a number of them with  

"Shift" the number of frames it`s working with in the Processing  

window is still 77.


"Include" means use that frame for stacking. "Exclude" mean prevent that frame being used in the stacking process.


How do I "toggle" back & forward between frames  

after selecting "Always display sharpest frame"?


From page 40 of the manual.


The "Always display Sharpest frame" check box allows you to toggle which movie frame is 

displayed. If it is unchecked the movie frame for the selected sharpness list item is displayed 

and if it is check marked then the sharpest (blue value) frame is displayed. This allows you to 

quickly compare the current selected frame to the sharpest frame so that you can decide if it 

should or should not be included in the stack. 


If it`s as simple as  

it seems then I`m not getting much difference between frame 1 & 77?  


Yes it's that simple and maybe there isn't much difference between the frames.


Why don`t you have a mechanism for selecting only the sharpest frame?  


The first item in the list  (0001) is the Sharpest Frame and has the highest value. It's already selected as the sharpest frame and it's in blue in the list.


From page 40 of the manual:


The "Set Sharpest Frame" button allows you to select which frame is to be the sharpest frame, 

which is the one that is used for pixel comparisons. The "Include" button will cause the selected 

frame(s) to be marked green as included and the "Exclude" button will cause the selected 

frame(s) to marked red as excluded. At the bottom of the list is the phrase "84 of 94 accepted" 

which indicates how many frames have been accepted out of the total number of frames in the

list. This value is updated as you include or exclude frames.


Do you consider the Preview>Grab selector to be sufficient for this?  


Sufficient for what - I have no idea what context your talking about here?


I`m finding that none of the TIF files then processed from my 22  

selection areas nearly match the sharpest frame. Is this normal or is  

there something I`m omitting?


Yes you may see some drop in frame sharpness from an individual frame. The atmosphere can move +/- several pixels erratically due to turbulence and your adding these multiple frames together. Also your using Noise suppression which can soften things too.


Is it usual practise for people to use  

only the sharpest frame as I`m now doing?


People stack multiple frames to suppress noise and improve signal to noise ratios. This smoothes out noise so that you can sharpen it a lot more afterwards. If you sharpen a single frame, you wind up really amplifying the noise.


I`m attaching another  

comparison between sharpest frame & processed TIF file & I`ll  

hopefully be able to send you the movie,  image selection area & log  

file privately. The sharpest frame & TIF file have of course been  

converted to jpeg which gives a reduction of resolution anyway.


Unfortunately the movie you sent has been h264 compressed which obliterates all the fine pixel details (it also becomes a bit blocky because of the DCT processing), is 533 frames and 426x320 pixels in size. So it isn't what you were working with originally and it's been compressed by 1: 142 times (i.e. if it was uncompressed it would be 199 megabytes not 1.4 megabytes). Also Apple's H263 h264 is really really slow to decompress in high quality, normally I would be running at 30+ fps if it was bayer or monochrome, but this only runs at maybe 4 fps and is chewing through 100% of the CPU.


It stacked fine, but detail wise it isn't all that great because the H264 has removed all that.


The color balance on it is really good, and one only has to increase the saturation to around 3 to pull up the faint colors differences between titanium content in the mares.


The log you sent shows that your using "Very Light" sharpening with noise reduction. Uncheck the noise reduction unless your using sharpness at medium levels. I usually use that combination to blur stars when stacking DSO objects.


HTH..


Milton Aupperle