Astrophotography photo stacking (need help)

[sean_lulgjuraj]

 

 

 

<p>Hey y'all,<br>

So I have some questions about photo stacking, I can't seem to figure it out. Just wondering if anyone could help to simply explain what the process is. I can only seem to find a lot of convoluted explanations.<br>

So first I understand the <b>dark frames</b> you take about 20, lens cap on and with all the same settings as the light frames. And take them right after you shoot your lights.<br>

Then <b>Bias frames </b>also take about 20, same settings as light and dark, lens cap on and shortest possible exposure. take them right after you shoot your lights.<br>

<b>Flat frames </b>so I've heard a few things, from my understanding you use the same settings as the lights, darks and bias, but you cover the lens with a white T-shirt or something then smooth out the folds then take 20 shots but at a bright light or at the sky?<br>

So my 2 questions are, first regarding the flat frames, what is the easiest way to do this? during the day? in the morning? what will produce the best results, and is the least complicated.<br>

My second question is I do not understand the light frames, I heard It would be ideal to have between 20-40 light frames, but if you are taking 40 frames of the exact same scene at say 45 seconds or whatever, are you not going to get trailing when you stack all of the photos together?? or do the stacking programs correct for this?<br>

Thanks</p>

 

 

 

[MichaelChang]

<p>Trailing in long exposure will depend on the focal length of your camera or telescope lens, so you will empirically choose the longest possible exposure time without noticeable trailing (or elongation of stars) - the stacking program does not (and can not) correct this. </p>

<p>Flat frames should be taken at the end (or beginning) of your shoot because its purpose is to cancel, in stacking, lens artifacts such as smudges and vignetting. <br>

<br>

Regarding number of dark, flat and bias frames and their respective importance: </p>

<ul>

<li>Dark frames are the most important, and temperature dependent, but the only time multiples are necessary is when you do automatic subtraction in software - only one will suffice if you subtract manually through opacity adjustment. This is because random noise in your light frames are averaged while the signal (stars) is additive, and a single dark frame will contain all the fixed-noise artifacts for the purposes of subtraction and removal, and its random noise component will also be averaged. The manual method will save a bunch of time.</li>

<li>Only one flat frame, for the same reasons as above will be necessary.</li>

<li>I've found that bias frames are only necessary for the most exacting and critical astrophotography as its components are already included in dark frames; really not necessary for wide angle or less critical shots. </li>

</ul>

<p>A lot of this stuff really comes down to what works for you either derived experimentally or doing it strictly by the book. In the end it's still about the picture's composition and aesthetic merit, and only geeks such as ourselves will be bothered by those slightly imperfect, albeit identifiable artifacts, if the photo contains technical flaws. </p>

<p>Here's a picture I made that's filled with flaws, no bias frame or flat frame, but I like it anyway - it adds "character".</p>

<p><img src="http://d6d2h4gfvy8t8.cloudfront.net/2617978-lg.jpg" alt="" width="800" height="532" /></p>

<p> </p>

[sean_lulgjuraj]

<p>Cool thanks Michael! I understand that trailing depends on focal length and shutter speed, but what I was asking about the light frames is how many light frames should you take? I heard around 20 to 40. So If I'm going to take for example 20 light frames at 30seconds per frame on my wide angle lends at 11mm which will not cause any trailing. When you stack those 20 frames in deep sky stacker how will there not be trailing? Your taking 20 pictures of the same scene every 30 or so seconds 20 times then stacking them together should'nt that in itself cause star trails to appear? how to people get crisp stacked photo's of the night sky with out star trails, maybe I'm not understanding something here.</p>

[MichaelChang]

<p>Sean, what you're referring to is called field rotation. <br>

Specialized (and expensive) software can take care of that, but the usual approach is to avoid trailing altogether by using a tracking mount - the camera-head rotates to track the stars - which can be manually controlled or automated.<br>

Using a tracking mount also has the advantage of allowing longer exposure times before tracking error produces trailing.<br>

Inexpensive equatorial mounts such as the EQ1 or EQ2 can produce pretty decent results for wide angle shots. These are manual mounts but accessory motors are available to semi-automate the tracking. <br>

</p>

[sean_lulgjuraj]

<p>Ahh I see! thanks again~! do you have any experience using the EQ1 or EQ2. I did a bit of research and a lot of people recommend the astrotrac system but it looks like it's around 800$ I also heard the EQ5 is good as well</p>

[MichaelChang]

<p>I have used both the EQ1 and 2 mounts for visual observation but not astrophotography; they are low cost mounts adequate for the most basic wide angle shots of the sky. </p>

<p>Generally, the more you pay, the higher the tracking precision, but also the more demanding it will be on your skill and patience starting with polar alignment and elements of the entire process. It all depends on your budget and what you want to achieve. </p>

 

[Glenn McCreery]

<p>Another approach is to use Starry Landscape Stacker, which is available as an app for Mac computers for $5. More information is available at https://sites.google.com/site/starrylandscapestacker/home</p>

<p>It works by compositing a series of exposures and shifting the sky to align the stars. The compositing also greatly reduces noise. If a foreground is included in the images, the foreground may be identified, and not shifted with the star field. I find that the program works well. Here is an example using six 10s images taken with a Canon 5D II using a Canon 17-35mm lens at 17mm and f:2.8. I did not try reducing the size of the image to 700 pixels maximum dimension, since the resolution would not be sufficient to show the quality of the result. So far, I have only used this method with a wide angle lens, but it should work at higher magnification, as long as most of the stars stay within the field of view for the multiple images, and the lens is well corrected for distortion.</p><div></div>

[sean_lulgjuraj]

<p>So what would I need to purchase? I'm thinking of getting the EQ1 or 2 just to experiment with it, so I need to buy a motor for it to automate the tracking? or is there a way to manually set it yourself?</p>

<p>Thanks Glenn! damn i wish I had a Mac, that would have been convenient </p>

[MichaelChang]

<p>Sean, YouTube has a number of tutorials on "<a href="https://www.youtube.com/results?search_query=equatorial+mount+setup">equatorial mount setup</a>" to get you started. </p>

<p>You won't need a motor drive to start; you can use its manual control. </p>