![]() ![]() We’ll be adapting our shooting and post-processing approach to still allow single image results while also accommodating a new workflow for people who want better, lower noise results. That has given us the opportunity to test and develop lighting methods. Since I originally wrote this post we’ve had a few night photography workshops in Bodie (see link above for info). I already have some ideas on how we might adjust our nighttime shooting practices, especially for those precious hours we have when we get permits to shoot at night in Bodie. The goal will be to get better results without too much impact on valuable night shooting time. How many images are needed at a minimum? How many optimum? Are different exposures needed for the foreground landscape? How many and at what settings? Should we shoot an HDR bracket, a sequence of images, or both: multiple HDR brackets? Some of these were time-lapse or star trails sequences that I re-processed to produce a single image result for the first time. Obviously I’ll gain experience and be able to fine tune the process and results, but the initial results are very encouraging. My sense is that you need enough stars to make the alignment work, and the physics is such that you don’t always get enough reflected stars to pull that off. Not all of my star reflection shots worked. Then I tried stacking 20 files from a single-exposure star trails or time-lapse sequence, shot later that night at a different focal length: I had to process the reflection and the sky separately, since the stars move in different ways in each, then merge the results. It turned out really well for a first pass. I noticed that the program has an HDR setting, so I pointed it at three bracketed Milky Way shots taken one stop apart in exposure. Next I had to try something a little more interesting, like a Milky Way reflection. We capture star trails sequences of the Methodist Church on most of our workshops in Bodie, so that was a natural subject to start with. I recently built a fast Windows PC, so I downloaded Sequator to see how it performed. On a Mac, there’s Starry Landscape Stacker. So programs have been created to stack the starry sky while masking and preserving the static landscape. Landscapes present a particular challenge, aligning the stars as they move through the sky would blur the landscape portion of the image. Go for 8s or less and stars will look much rounder.20 images of Joshua trees stacked with Sequatorįor quite some time astrophotographers have used a program called Deep Sky Stacker to align and combine multiple starry images and create a better result. So we can establish that you used roughly 23s exposure (20s in reality) from rough star shape measurement. In above calculations (14 - 6) x 2.87s = ~23s You said that you used 18mm and 20s exposures? Brighter stars measure 6px x 14px, which would give ratio of x2.33. Let's say that your stars are 3px wide, and you want to maintain 2:1 star shape - longer diameter being twice long as short diameter of corresponding ellipse - this will mean 3px max, or about 8.6s. For image to "move" one pixel, it takes - 2.87s. You are shooting right at equator - declination 0, so you need to use sidereal rate of 15"/s. In your case working resolution will be 206.3 x 3.7um / 18mm = 42.6"/pixel ![]() Resolution is 206.3 x pixel size in um / focal length in mm Resolution that you are going to show your images at (or work with) - depends on pixel pitch and focal length of lens, but also any binning used. Rule of 400 or 500 are really rough guide lines - actual max duration depends on couple of factors.ġ. So there has not been any improvement stacking the images, if anything the stacked image looks worse ? (ignoring the blury corner of the house). With regards the image quality of the stars in the original jpg and the stacked raw images, the original looks sharper and to be honest looks better to me. This way you will get both stars to be sharp and house to be sharp. Use stack that you already have, but replace section of blurred house with regular house from a single frame or stack of frames where house is stationary and stars move. The way you deal with this is to produce two stacks, or one stack and use single frame. ![]() If you are stacking on stars and keeping them stationary - house will move frame to frame and you will get motion blur. If you were to stack on "house" to get it sharp - your stars would trail. Each frame will contain different position of these two compared to other frames. ![]() You have two different "layers" of objects in your frame - moving sky and stationary foreground. That is quite normal for stacking, and it is not considered bad stacking result. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |