disperser 19July2022 Photo Post-processing

Photo Processing Tools

Warning: this is all about photo processing; if not interested, just look at the photos or watch this slideshow:

Slideshow of Photo Processing Tools (40 photos)

Readers occasionally comment favorably on my photos . . . and I typically downplay any thoughts that my skills as a photographer are of any note. This is especially true when I look at what some people do with their photos.

At the center of it all is the postprocessing of photos. There are numerous classes and countless free videos dealing with processing photos of all manner of subjects . . . animals, pets, architecture, landscape, macro, and the list goes on and on.

What you see in many of these tutorials are tools employed to process not the photo as a whole, but individual adjustments — shadows, darks, light, highlights, midrange, hue, tonality, white balance, vibrance, saturation, clarity, texture, etc. — applied in different amounts on different parts of the photo (through masking).

That’s a photo taken 18 years ago with the Nikon D100 camera. In the past 18 years, the capability of photo processing software has greatly increased. I could be wrong, but I attribute it in part to the continuing hardware and software advances of digital cameras; there’s just much more information to work with than what you had with film.

This is where someone will call me out as not knowing what the heck I’m talking about. They could be right, but my statement is in part referencing the limitations of what I could do. I don’t have an example of the original processing of this photo, but I have an example from 2013 . . .

Compare that to this version that I just processed . . .

I cropped it a bit more, and it’s sharper, perhaps too much so if you look at it at 100% magnification (the other is perhaps too soft). Here’s the monochrome version I just processed . . .

Monochrome photos, just like color photos, have a lot of options and ‘looks’ . . . if you like the above look, great; if not, I have hundreds of other looks I could offer up (but won’t).

Those are examples of global adjustments and the nine years difference doesn’t produce drastically different results. Global adjustments act on the whole photo, and as such, there’s a limit to what you can achieve even now, especially on photos with wide dynamic ranges (the difference between the darkest areas and the lightest areas).

True, some adjustments affect only shadows, or only highlights, or only one color, and so on, but they act globally as opposed to adjusting only one area of a photo and excluding a different area.

This next photo is another example of global adjustments, albeit these are directed by AI software. That means the software guesses what the photo is supposed to look like based on having ‘learned’ from having scanned many other similar photos, and makes adjustments based on what it has learned.

That’s literally a couple of clicks. I got a similar result years ago by moving a number of sliders around, but it took more effort.

In these next two pairs of photos, I masked the sky and treated it differently than the rest of the photo. So, it’s no longer global adjustments because I make different adjustments for different parts of the photo. Still, they’re not refined adjustments because within each of the areas (sky and ground) I apply adjustments to the entires masked area. You can see that the very peak of the mountain in the first pair gets a bit blown out.

That’s one of my favorite shots from the site, and the above is the best that I’ve been able to do with a photo that had a fair amount of issues. Below is a previous attempt using strictly global sliders and manually adjusting levels of shadows, highlights, etc.

It’s a matter of preference as to which photo you like best, but for me, the current processing is much better. (keep in mind WP doesn’t flatter photos; to see how they look, click on each of them)

Some photos can pass with very minor adjustments, and global processing is sufficient. Here are three pairs of photos where the AI and global tweaks produce decent results without resorting to localized adjustments (except for the puma’s eye).

Here’s an example where I used multiple masks because global adjustments weren’t cutting it.

There are a number of ways one can create a mask; by subject, color, luminance, manual selection, and so on. I’m not saying that’s the best result for that photo, but it shows how you can apply different adjustments to different areas of the photo.

Sometimes the photo might just need a bit of noise reduction and sharpening . . .

Here is another example where you can’t just brighten everything and where selective adjustments can bring out details in the shadows without blowing away the highlights or messing with the bright areas.

Again, one could argue about tinting or white balance, but the point is that localized adjustments give a more balanced look to the photo.

Sometimes the photo is just ‘soft’ and needs a bit of punch. This is true for most of my photos because I shoot with little to no on-camera processing. This is another case where it was beneficial to process the sky separately.

The most difficult scenes to process (for me) are those with red rocks . . .

When it comes to red rocks, there is a fine line between too little and too much . . . and also to get the saturation and contrast right. In this shot, the distant hills were processed separately.

This next shot is an example of first using a global approach and then using multiple masks and trying to balance the scene without making it unrealistic. The big problem in this photo is the dynamic range; a very bright horizon and sky and a very dark (from a passing cloud) foreground. I should have bracketed the photo and then merged them . . . or waited for the cloud to pass.

I think I (maybe) managed to overcome the challenge. This is the first time I processed this photo and was happy enough with the results to share them, and that goes both for the global and the masked versions . . .

Processed with masks and local adjustments

I think I used four or five masks . . . I like the results from using masks a bit more than the global adjustments version, but — as always — it’s subjective.

How about the monochrome version of it?

The fact that I normally process multiple photos for a given post is why — when comparing the results of the global adjustment version (which took minimal time) to the results from using masks (which took considerably longer) — I’m not moved to put the effort into masking and local adjustments.

That’s especially true when the photos are just going into a blog post that — on a good day — maybe 15 people will see. More importantly, of those 15, probably only 2 or 3 will give the photos more than a passing glance.

So, yes . . . adjusting the scene using masks yields (marginally) better results, and that would matter if I printed the photo to either sell or hang on my wall.

Here are a few recent photos processed a bit more carefully and using masks. Not that different from what I’ve shared, but if compared side-by-side, one could discern subtle differences.

This last photo is an old one (2004) . . . and I should have masked and tweaked more components, but there’s something else I consider when processing photos . . . the amount of time spent versus the expected payoff. In this case, I deemed it good enough.

I also ran the above in Topaz Sharpen AI and DeNoise AI, and it may be a bit overdone, but I was rushing because my interest in the effort was waning. What I should have done is make hue/tint/color adjustments for each of the masks, which I did not do.

Maybe something like this . . .

. . . but who’s got the time?

That’s it. This post has ended . . . except for the stuff below.

<><><><o><><><><><o><><><>

Note: if you are not reading this blog post at DisperserTracks.com, know that it’s copied without permission, and likely is being used by someone with nefarious intentions, like attracting you to a malware-infested website. Could be they also torture small mammals.

Note 2: it’s perfectly OK to share a link that points back here.

<><><><o><><><><><o><><><>

If you’re new to this blog, it might be a good idea to read the FAQ page. If you’re considering subscribing to this blog, it’s definitely a good idea to read both the About page and the FAQ page.

12Comments

Add yours

1

JenT on 19July2022 at 5:23 AM
Reply

Those are great examples. A couple of the landscape ones reminded me much of the postcards that my dad used to send me when he was a traveling salesperson and I was a wee girl. I think it’s the level of saturation.

LikeLike
- 2
  
  disperser on 19July2022 at 8:38 AM
  Reply
  
  Thanks, JenT.
  
  There’s actually not that much increase in saturation. It’s more playing with highlights, shadows, black and white points, clarity, contrast and exposure. The AI do a great job of balancing all of those and while they occasionally increase something called “vibrance”, they hardly ever increase saturation, and if they do, it’s minimally compared to other adjustments.
  
  Before the advent of AI, the “auto” settings were restricted to just a few tweaks, but even there, saturation is like a blunt instrument. The ‘Auto’ settings is how I learned to edit photos . . . I would apply the auto adjustment and see what they did, then played with the sliders to see how they each affected the photo.
  
  But, the AI tools are much better than I am at balancing the settings, especially Luminar AI.
  
  Combined with noise removal and sharpening, there’s a definite perception of increased saturation, but I prefer calling it “punch” because, as I said, the actual saturation is rarely bumped, and is sometimes reduced to offset the other settings.
  
  Whether I overdo my overall adjustments is another matter. I can look at a photo at different times of day and think it’s not “punchy” enough (dull) or too “punchy” to where it looks unnatural. And, looking at it on a screen is way different than on a print, and then it matters what kind of print (matte, glossy, etc), and it matters on what kind of screen (my screen is color-calibrated, but most people’s aren’t).
  
  Still, when I compare to photos in magazines and other websites (again, the medium matters), perceptually, I’m more restrained than what is professionally published. They are obviously better than I am at cranking up the visual impact without overdoing it . . . because they mask the crap out of photos. It’s interesting watching videos of professionals manipulating the look and feel of photos. Of course, they spend as much time on one or two photos as I do on 40 or more.
  
  One thing I cannot control is the calibration of the devices people use to look at my photos. Depending on their settings, that can either help or hurt how the photos come across. Things like HDR settings, brightness, and any setting that makes one display different from another, and so on. I’ve also gone to presenting larger version because the resolution of tablets and phones have increased and a larger photo usually presents better than a smaller photo.
  
  Lastly, those photos are all from different cameras (D100, D200, D7000, and D7500), and each camera has its own color rendering engine. For instance, the D100 always appears “more saturated” than the others. The last photo is from the D100, and I typically struggle with making adjustments on D100 photos because I end up pushing the already “punchy” colors over the top (look at the greens; I had to desaturate them and reduce the luminosity).
  
  However, when properly exposed, I think the D100 outputs more pleasing photos than the other two. In part, because it’s a 6MP camera with larger pixels (photon collectors) which aids in collecting more light, but at the cost of resolution.
  
  Anyway, probably way more than what you wanted to read (and likely already know).
  
  LikeLiked by 1 person
3

oneowner on 19July2022 at 6:15 AM
Reply

I think you could write a book about masks. The word itself is deceiving. In PS, a mask is used to block an effect on a layer whereas in LR it’s used to apply an effect. If you can wrap your head around that you can find out that masking is probably the most powerful tool you can use in post-processing, And it goes way back to early chemical processing in photography. Great post with interesting before/after illustrations.

LikeLike
- 4
  
  disperser on 19July2022 at 8:51 AM
  Reply
  
  Before we moved, I had at least three books about masks . . . and I must confess I never finished any of them. In part because the tools changed, and some of the books — while helpful in describing the concept — were already outdated relative to the software. But also because they were really boring.
  
  FYI, I tried working with the Lightroom masking tools, but it’s kind of tedious and too difficult for someone inexperienced (me) to manipulate and refine. In its defense, I’ve not spent much time with it, nor looked at any tutorials. I rarely if ever mask in Photoshop because — while probably more powerful — it’s even more of a pain to use.
  
  Most of the above photos where I made fine adjustments of multiple masks in adjacent areas, were masked and tweaked using u-point technology (Nik Collection). I think that’s a less intuitive but easier to master and apply adjustment tool than Lightroom’s masking tools. Again, probably because I’m more familiar with Nik (even though I rarely use its masking function). Lightroom has more ways to select positive and negative masks, but I couldn’t make them work as well at the boundaries between masks. Again, that’s on me, not the tool.
  
  I have to thank you for restoking my interest in masks. And, if I really get into it full bore, my readers will thank you (some, at least) because I’ll have fewer photos in each post.
  
  On the other hand, I’ll have even more photos that will never see the light of screen. Thanks for the comment.
  
  LikeLike
5

oneowner on 19July2022 at 2:06 PM
Reply

Masking In LR is getting more and more powerful thanks to AI technology that has been available in PS. The simple “select sky” and select subject” work fairly well but I admit they always need tweaking. Also, if you only want to select a portion of a subject – you’re on your own. The brush tool is your only friend. In the future, I see complex selections being made with little input from the user. I would welcome tools like that but would still want a way to manually control the selection. Control freak?! Who, me???

LikeLike
- 6
  
  disperser on 19July2022 at 2:16 PM
  Reply
  
  The current Sharpen AI is sometimes scary-good at selecting the subject out of a picture. Photoshop is getting there as well.
  
  The problem I saw with Lightroom is with the brush. No matter the settings, I could usually still see the boundary of its application.
  
  Have you tried Nik Collection’s U-Point selections (control points)? I think you had the free version. I like how it works for controlling selections.
  
  The latest version (which I’m not sure I’ll get because I think the upgrade price is too expensive for what you get) includes luminosity and hue options to refine selections.
  
  I enjoyed playing with masking and localized controls in the above photos, but they took a fair amount of time. For the landscape scene, selecting the “sKy” grabbed all the way down past the mountains and hills and into the valley. I then had to have a number of negative masks (exclusion masks) and had to spend a long time working on the boundaries. Admittedly, that was a difficult scene to mask because of the close gradation of luminance.
  
  LikeLike
  - 7
    
    oneowner on 19July2022 at 8:33 PM
    Reply
    
    I was an early user of the NIK Collection but it’s been years since I last used them. The U-Point technology was cutting edge at the time but I don’t know if the newer versions have kept up with the tech. I would consider using plugins again if they had better integration with LR. Ideally, you could import a RAW file from LR to the plugin and it would return a RAW file back after processing. That’s tough to do. Even PS can’t save a RAW file. But I’m sure a plugin can make a modification to a RAW file and return that file back to LR without the saving process. That’s probably going to infringe on Adobe LR rights. The one plugin I’m using now, Topaz Sharpen AI, returns a TIFF file so I use it after all other LR adjustments are made. Though it’s possible to make further adjustments on TIFF files in LR, I don’t like to do it. At the Museum, I shoot in RAW, edit the RAW files, and give them finished TIFFS and jpgs. Other than for use in the photo studio, they have little use for the RAW files except to archive them.
    
    LikeLike
  - 8
    
    disperser on 19July2022 at 11:04 PM
    Reply
    
    I’m not sure which version of Sharpen AI you are using, but the version I have can read RAW, and while it can’t output the native RAW file (because, as you said, infringement), it can output DNG files.
    
    I’ve not done a comparison to see how much is lost, but CAMERA RAW can read those files, recognizes them as being from the D7500, and all the adjustments appear to work as normal. That’s the same for the other Topaz AI programs, and for DxO’s PureRAW and PhotoLab (although they also have the option of exporting something called Linear RAW which I’m unfamiliar with and have not looked up).
    
    I’m assuming they export a DNG with the corrections applied, so I don’t know how much of the original information is retained. I’ve not run tests, but I’m assuming I wouldn’t (easily) be able to return an exported file to its original settings.
    
    Luminar AI and Neo can read RAW, but they output TIF/PSD/JPG/PNG. I think it’s Nik collection that can
    
    It’s true that the NIK plugins as standalone, will only accept JPG files, but as a plugin in Lightroom, it can edit TIF files and also has an option to save them so that you can go back in and modify the original edits (as opposed to editing on top of the existing edits) .
    
    I assume that might eventually be expanded to accept RAW since DxO has integrated the Collection into PhotoLab (but still as a plugin).
    
    When I output TIF files, I don’t use any compression, so, in theory, I’m not suffering from a loss of data, but, of course, it’s not the same as the original RAW. (NEF is based on the TIF format, but not the same)
    
    What’s odd to me is that Photoshop exports something called Photoshop RAW whereas Lightroom can export DNG files. The reason I find it odd is that DNG was developed by Adobe, so I’m not sure why Photoshop can’t export to DNG (unless it’s well-hidden and I can’t find it).
    
    However, regardless of what’s exported, it’s not the original file (in my case, NEF), so once I’m reading the NEF in any program, I’ll always be working with a rendering based on what they get out of the NEF file.
    
    There’s one other thing, and I could be wrong, but . . . when you say you work in RAW (or when I work in RAW) what I mean is that the NEF file is interpreted by whatever software I’m using, and, depending on the software, I may not see the same thing.
    
    Not that anything is lost, but Capture One and Camera RAW will render the same RAW file differently. So, I can still access all the data, but I’m not technically editing the RAW file; I’m editing a rendition of it, so I might be editing from a slightly different starting point depending on what program I use to open the NEF file (but the data I’m reading is the same).
    
    If I just open the NEF file in Photoshop without making any changes in Camera RAW, I’m editing the rendering based on the default RAW settings. Meaning, I think I have more control in processing the photo if I first make adjustments using Camera RAW (I don’t have Capture One), and then make final adjustments once in Photoshop proper. Although, I think you can still fire CR up from within Photoshop after the fact.
    
    Lightroom, on the other hand, incorporates Camera RAW’s processing options in its interface. The Develop mode could be called — and I think used to be called — Camera RAW. The adjustments it has are the same as Camera Raw. So, in Lightroom, you are, essentially re-rendering the RAW file whenever you make any adjustments.
    
    I know you know all this, but in case anyone is interested, I just wanted to make the distinction. Bottom line, even though editing a RAW file, different software packages will give you different starting points based on their interpretation of the original RAW file.
    
    As a side note, Nikon’s NX Studio does work with the NEF file directly since it’s their format. That means that when I open a file in NX Studio versus Lightroom, if I put them side-by-side, they do not look exactly the same. Close, but the colors, shadows and highlights, and noise ( at the pixel level) are different. Not much, but noticeable upon inspection, especially in the shadows.
    
    I believe I have access to the same data, but the interfaces don’t render the camera’s starting point as exactly the same. My understanding is because the RAW formats are proprietary and software companies have to reverse-engineer them to be able to import raw formats. Although, that could be changing: https://helpx.adobe.com/camera-raw/nikonraw.html.
    
    However, I also see rendering differences within NX because I have the option to be “camera compatible” or use the “latest picture control” which basically changes some of the base settings; same data, different rendering. That points to the NEF format as not being static, but having had updates through the years.
    
    And, of course, none of this matters once I start editing.
    
    LikeLike
9

doesitevenmatter3 on 19July2022 at 4:37 PM
Reply

These are wonderful and a joy to look at, Emilio!
I like hearing about, and seeing, the work you put into your photos!
Even I, a not-a-photographer, learn things from your posts about your photography.
(((HUGS))) and Happy Not-Monday! HA! 🙂

LikeLiked by 1 person
- 10
  
  disperser on 19July2022 at 4:40 PM
  Reply
  
  Thank you, Carolyn. Just playing around with some old photos as a way to keeping current with the latest tools for editing files.
  
  Happy two-ess-day to you as well.
  
  LikeLiked by 1 person
11

Ggreybeard on 19July2022 at 11:27 PM
Reply

I used to use masking quite a lot – but sometimes it can be tedious and these days I tend to tweak globally using the curves feature.

Good to see you squeezed in a few hummers there.

🙃

LikeLike
- 12
  
  disperser on 20July2022 at 9:25 AM
  Reply
  
  My motto is . . . dazzle them with quantity, not quality, so most of my adjustments are global.
  
  And yes, most of my photos right now are hummers. It’s too hot to go out and do anything other than sit on the patio early morning before the heat and humidity build. So, birds and bugs; the hummers usually put on a better show than the rest of the birds.
  
  LikeLiked by 1 person