Affinity Photo masking bright objects

[Bwood]

I’m a relatively new user of AP. Every year I take pictures of Christmas lights at our local museum. 
I have one particular challenge. Inside one of the rooms, I have two brightly lit snowmen. The other portions of the display are quite dark. I can take two pics, one exposed for the snowmen, and another for the rest of the room. 
I have not done much with duplicate photos in two layers, but I did see a tutorial on masking one photo then combining the layers. 
Does this sound like a possible solution. Or can one work on one photo without combining.

Thanks kindly,

Bill Wood

[Ldina]

@Bwood

There are many approaches to deal with images that have high contrast extremes. 

You can use AP's HDR Merge feature (works best with JPGs, TIFFs, PNGs, etc, not as well with RAW). This will take the most detailed areas from each exposure and blend them together. 

Another approach is a New Stack, which aligns the images and puts them into a stack. The stack has an "operator" to blend the images together (Mean, Median, Max, etc). You can blend images manually this way. Or once aligned, you can drag them outside of the stack and work on the layers individually, with masks, etc. More manual work.

Another approach is to expose for the Highlights and use a single image to create an HDR image from a single exposure. The discussion below has more info and a few links to videos. Check out the Affinity Photo Tutorials section of this website (both version 2, and the legacy V1 videos at the bottom of the page). There is a lot of information on Tone Mapping, HDR Merging, etc. 

 

[Bwood]

Many thanks @Ldina  I had thought of a stack. Expose for the bright object, then another exposure for the surrounding display. I will brush up on stacking. There’s a decent tutorial for that. 

[Bwood]

@Ldina  one more question. I’m dealing with a stack of raw images. Is the process of stacking for bright image exposure better with jpegs? 
 

as well. Did you process your stack images first, before you stacked them?  Thanks Ldina 

 

Bill

[Ldina]

@Bwood Doing an HDR Merge with RAW files didn't seem to work all that well for me. I think part of it is that the White Balance might have been different in each exposure. (BTW, I noticed that the James Ritson HDR Merge tutorials I've seen always seem to start the HDR Merge process using pre-processed TIFFs or JPGs.) RAW images seemed to work a little better with a New Stack, rather than HDR Merge, but still required too much work. At least with Affinity, I have found it's better to start with preprocessed images that cover the entire dynamic range you want to capture. 

Pre-processing TIFFs, PNGs or JPGs might be the better approach. First, you can set the WB the same on all the exposures you plan to merge, so they match when stacking or HDR Merging. Second, if you wish, you can fine tune the exposures of each individual images to maximize the tonal range you want to extract, so you have the best possible images to start your stacking operation. I've had pretty decent success even using bracketed, in-camera JPGs (which were whatever the camera delivered to the memory card). As long as you have quality images that cover a wide enough range of bright highlights to deep shadows, with adequate overlap, it should work fine. There might be an edge to individually processing your bracketed exposures, but I can't say that for sure. Three to Five images should probably work for most HDR Merges. If you are Merging in a New stack for object removal, simulating a long exposure (like a long exposure of moving water), or things like that, you will probably need quite a few more exposures. 

When working with RAW images as a starting point, I created a macro that I can apply during a New Batch Job. This way, I can quickly process my RAW images, apply the same WB to the entire group of RAW images, develop them, and Save them to whatever format I want (TIFF, JPG, PNG, etc). This saves a lot of time and eliminates the need to process them one by one. I created one Macro which develops RAW images using a Daylight WB (approximately 5000K), and another Macro that uses a Shade WB (6500-7000K). I also created a 3rd Macro with a WB set for images taken in the woods under a thick canopy of yellow-green leaves (I forget the WB I used, but it was probably around 4500K with a +15 Tint). I think it's more important that all of the exposures you plan to Merge have the same WB, and that the WB is reasonably close to what you want in your final image. You can always adjust the WB of your merged image afterwards, but if all your bracketed exposures have a different WB, you end up with a "mixed light" image, which is very difficult to color correct and keep looking natural (like shooting a portrait lit by daylight on one side and fluorescent lighting on the other...not fun).

Try some in-camera JPGs. Start with 3 exposures, each 1/1/2 to 2 f-stops apart. (Vary shutter speed, not aperture). For example, shoot 3 exposures all at f8, at 1/60s, 1/250s and 1/1000s. The 1/60s exposure should be overexposed with plenty of good shadow detail. The 1/1000s exposure will be underexposed and should have plenty of highlight detail, even in the brightest areas. The 1/250s exposure should be a 'normal', balanced exposure, with some plugged shadow and blown out highlights. Depending on the contrast of the scene, you may sometimes be able to get away with 2 exposures, but some may require 5 or more. Three bracketed exposures ought to be a good place to start. Just make sure your middle exposure is about right, and the other two should handle the extremes. 

[Bwood]

Great detail @Ldina  thanks kindly for the step by step. I’ll process and make sure the white balance is the same for all jpegs. On the last outing last night I took a number of shots, working from blown out whites down to detail in the background of the display. 
 

This Christmas village in my hometown changes every year, and driving home I surmised that the easiest part of the project is shooting colourful lights with no white sources. It’s so easy without white light ;). I’ve already seen a number of photos posted to social of this village that are almost cartoonish with the sheer over processing. Many thanks again. I’ll post on this thread when I come up with something presentable.

[Bwood]

@Ldina  While I’m working on my collection for this museum, I’d like you to look at something. This picture was done by a professional a couple of years ago but I couldn’t figure out how he did it. I asked him but he wouldn’t give away his secrets. We live in a small community. The only thing he told me he used a very low iso. I believe he used Adobe. The whites certainly weren’t blown and the colors are quite natural. I was never able to duplicate it, though I believe it was highly processed. 

[Bwood]

And here’s my foray into stacking. Not perfect but on the right track. As you can see the snowmen were the only light sources in this school house. 7 raw images were batch converted to jpg. The reindeer will be processed again, no stacking but there’s a grey cast which I just noticed.  On a foggy night one of my favourites is our row of older buildings. The fog gave me an interesting sky. 
All amateur, but fun to learn. 

[Ldina]

30 minutes ago, Bwood said:

I couldn’t figure out how he did it.

It could have been done a number of ways, but without original RAW files, it's speculation. (It's sort of sad that he wasn't willing to share with you.)

It could have been a standard HDR Merge of multiple exposures. If the exposure was perfect and shot at a low ISO, it could possibly have been a single exposure. It could also have been a single RAW file that was "developed" multiple times and blended in Photoshop, Affinity, etc. Of, a stacked, blended image. Etc.

One of the "potential" downsides of working in HDR is that we can become obsessed with trying to preserve detail everywhere in the image, from bright highlights to deep shadows. That is especially true if the final image is going be tone mapped into standard dynamic range. If the final delivery is an HDR image displayed on an HDR monitor, we can usually get away with a lot more because we have much more range to work with. Preserving excessive detail is particularly true when our original starting point shows tons of detail...it almost seems criminal to push some of that detail to become slightly plugged or blown out. Our eyes acclimate to the HDR image in front of us, and we use that as a reference for further edits. 

Images with excessive detail everywhere often look a bit fake and unnatural. It's often (not always) better to allow some details to plug or blow out, at least perceptually. They just look better. 

To me, this image falls a bit into that category...more detail in the shadows and blacks than I'd prefer. It's not wrong, per se, but I'd edit it differently...who needs that much detail in some of those shadows? Then again, up North at that latitude, perhaps it does stay brighter at night and you want to show that in the image. Perhaps my monitor is just set brighter than his monitor when he edited it, but the Histogram of the uploaded image suggests otherwise. I'd prefer something a bit darker, especially in the blacks and shadows, which seems more natural to me, plus the added contrast makes those bright, colored lights stand out more, which is the focal point of the image (IMO).

I tried to upload a revised version, but I got an error message (not sure why).

[Bwood]

5 minutes ago, Ldina said:

 

Images with excessive detail everywhere often look a bit fake and unnatural. It's often (not always) better to allow some details to plug or blow out, at least perceptually. They just look better. 

 ^^^^ This. I’m so glad a pro user feels this way. Now with a camera in every pocket so many over process and miss the feel of the photo or artificially manufacture it. Most of my shots are on the darker side, processed for detail UNITL it looks artificial. 
I very much appreciate your step by step guidance in the stacking process. I concur that it works so much better with jpeg rather than raw. 
My background has been broadcasting for 50 years, now I handle web development for non profits. I do most of my own photography for the sites I manage (animal shelters, museums, churches). 
Now my next project is squeezing down large jpgs to png or other formats suitable for the web. (Smaller is better). 

[Ldina]

40 minutes ago, Bwood said:

Now my next project is squeezing down large jpgs to png or other formats suitable for the web. (Smaller is better). 

Yes, an important area that is in a lot of flux at present. I'm a fan of JPEG-XL, but it is questionable whether it will become mainstream or not. AFIV, WebP, HEIF, etc, all have their strengths and weaknesses, the biggest being wide support by browsers, apps, and operating systems at this time. Standard JPG is still usually the "safest" choice, since it is universal, and anybody should be able to view them (in standard dynamic range). HDR is changing the landscape in a big way, but it's currently in a state of flux. It will be interesting to see which formats rise to the top. 

BTW, PNG files are usually quite a bit larger than JPGs, but they do support transparency, which JPGs do not. JPG Quality 85% is very high quality (exported from Affinity, at least). There's rarely a need to go higher than 85% (IMO), but if you need smaller files, you may be able to go with lower quality. It depends on the image, and the final quality you need.