Post Processing Simulated LE Brackets

In the last post I covered the basic approach to simulating a Long Exposure (LE) image using Photoshop. Although the post was based on a Magic Lantern Lua script to auto generate the images in camera, any camera is capable of generating the required input images, eg using an intervalometer or simply pressing the shutter multiple times.

Once you have ingested the image brackets into your computer, eg I use Lightroom; the first thing to do is process one of the images, ie from the RAW state, and sync the rest of the bracket set to this master.

The next step requires the brackets to be sent over to Photoshop as layers in a single document.

As mentioned in the last post, you could now use a Smart Object statistics approach, but this is heavy on time and memory. So I prefer to use a opacity merge approach, ie each layer’s opacity is adjusted according to its position. Layer 1, top or bottom, is 100%, layer 2 is 50% and, in general, layer n’s opacity needs to be set to 100/(n).

Although you could do the opacity changes by hand, I have created a Javascript based on an original script written by Neil Farquharson at www.verdantvista.com. You can download my version from here: https://gist.github.com/pigeonhill/309a7be818d4d523d50684e0fecd2d6c

The script needs to be placed in the Photoshop scripts folder.

This script, which you can run via File and Scripts in Photoshop, first prompts if you wish to parse the opacities into blocks. If you leave the number at the default 100, the opacities will be set as a single block, ie with 25 images the 25^th layer’s opacity will be set to 1/25. This assumes you haven’t taken more than 100 images!

If you input, say, 5, the opacities will be set in blocks of 5, eg [100% x {1, 1/2, 1/3, 1/4, 1/5}]….repeated across the number of layers you have.

If you use the block approach, you will need to merge each block manually and then rerun the script to merge the merged-blocks.

According to what approach you use, and the number of layers in a block, you will get a subtly different look. In other words you have choices.

In the following example there were 15 input images, each taken at an exposure of 0.6s: giving a simulated LE exposure of 9 seconds.

 As usual I welcome any feedback.