Breaking News

HDR Photography: Friends Don’t Let Friends Make Boring Pictures

Explore the vibrant world of HDR/DRI photography with this easy-to-follow tutorial. Make your photos sing in Photoshop and keep friends guessing how you did it.

Food for Thoughtful Photos

 
The few remaining film purists of the world like to tell budding photographers “all you need is some Velvia, a pinhole, and a vision, maaaaan,” but they’re just being stubborn. Manual processing is quickly becoming an obscure hobby and Nikon has discontinued its pre-digital flagship SLR, the F100, along with almost all of its other film cameras. If you haven’t already heard, let me help the cause: we live in a digital age.
 
And it’s never been easier to crank out terrible images by the truckload. One look at any friendship/networking website will show how far “vision” is from the minds of the kids shooting poorly-lit self-portraits in bathroom mirrors. Perhaps the film purists were on to something after all.
 
Modern photography is an evolving, exciting, and intimidating hobby that is largely treated as a convenience. It’s much easier to fire off some flash shots than it is to manipulate a scene in a truly pleasing way. To make things more difficult, in addition to a great eye for situation and concept, photographers also need powerful computers, imaging software, and some degree of technological expertise – all in addition to the gear inside their camera bags.
 
If you’re a hobbyist trying to produce clean, professional looking pictures – or someone who’s simply curious and just pressing buttons – it can be very difficult to nail down satisfying results.
 
This Photoshop tutorial is a primer to get you cranking out stunning images right now. What it won’t do is cover the basic functions of your camera – there are sites all over the web providing proper introductions to gear. This is a very specific software solution that some photographers have used to harness a difficult variable: uncontrollable light.
 
Welcome to the fascinating world of HDR and DRI photography.

[50mm f/4.5 ISO 100. Six image DRI composite. I don’t carry a flash kit and the available light here was terrible. I asked my model to pose as still as possible for 60 seconds while I shot a wide dynamic range of JPEGs. Later, I combined the necessary details from each image with a DRI technique in CS2, and saved what would have been a wasted night out.]

[12mm f/8 ISO 250. Six shot composite. Without creating an HDR image, the “COFFEE” text would have disintegrated into the white sign from the long exposure. Increasing the dynamic range also helped to dull the lens flares and naturalize other lighting flaws.]

HDR 

HDR stands for High Dynamic Range.

 
High in HDR means “beyond the normal value.” This begs the question: What specifically is the normal value? The normal value refers to the image your camera is capable of capturing when you take a single picture – and if you email me philosophizing about RAW format, I’m going to reply with pictures of a peanut butter sandwich wearing sunglasses. I’ll do it, too.
 
Any single picture you take might look bright, or dark, or perfectly natural, just like you saw it through the viewfinder. The picture could look like any infinite blend between these three basic values of Light, Dark, and Natural, depending on the settings you dialed in or the automatic settings the camera itself chose – but it is impossible for your camera to capture all these different values at once. Thus, the word High in High Dynamic Range ultimately means that an HDR image exceeds what a non-HDR image can contain.
 
The Dynamic Range part of HDR is related to film exposure times, lens aperture values, and the way our eyes perceive change. If someone showed you two fundamentally opposite exposures of the same picture, what qualities might make one seem dark and the other light?  Would you judge them by the depths of their shadows and the brightness of their highlights, or by the absence or presence of illumination?
 
You have a natural understanding of a visual dynamic range. Night is by definition different than day – shadows appear deeper and more dramatic in moonlight, but definition and detail emerge under direct sun. Depending on what things look like outside, there is an infinite opportunity for the natural lighting of the night-and-day continuum to blend. The beauty of HDR is its ability to overcome some of these complexities by increasing important visible details.

DRI

 
DRI is an acronym for Dynamic Range Increase.
 
This is how we’re going to obtain our HDR image: by increasing the dynamic range. DRI is not a means to an end as much as it’s an umbrella term for the various methods used to create a HDR image.
 
Now that we know the terms, let’s learn when and where to use them.

Situational Relevance

 
HDR/DRI photography is extraordinarily useful for:
  1. All kinds of stationary night photography – especially photos including neon signs or bright light.
  2. Daytime indoor shooting without a strobe kit.
  3. Any high-contrast environment where you’re forced to properly expose the subject and blow out the background; or conversely, ignore a poorly-lit subject to keep the background.
Unless you’re being clever, HDR/DRI methods are useless for:
  1. Action photography.
  2. Fidgety models.
  3. Shooting non-structural subjects outside on a windy day. Everything capable of movement will appear blurry in your final photo.

Equipment

 
Here’s what you’ll need:
 
Digital camera with a Manual option. I’m using a Nikon D80 with a 12-24mm f/4 AF lens.
Adobe Photoshop. I’m using CS2.
Tripod. The camera needs to be kept still. Mounting your camera will ensure your scene’s duplicated composition across multiple shots. It will also keep everything steady during longer exposures.
 
When you bought that new camera it probably came with a trial version of some random imaging software. You may outgrow this software quickly depending on how ambitious you get with your editing. Using a program like GIMP is an appealing option because it’s free and it supports the open source ideal – but unfortunately for everyone except Adobe, its practicality is often proven limited for users doing more than basic photo retouches.
 

(AKA: Free Editing vs Editing Freedom)

 
I prefer Adobe Photoshop. It’s an extremely powerful corporate standard graphics app with a great GUI. It’s like duct tape for pictures – with enough willpower, time and bad language you can fix anything. You’ll also find a knowledgeable web community out there who has already solved every problem you will ever have with the software, and that type of comfort gives me a warm and fuzzy feeling worth about $500 USD.
 
Now that you’ve got your gear, let’s go find your scene.

 

Say “Quality,” Not “Cheese”

 
Paying attention to the world around you and caring about the different ways people interpret art is still the most important part of making good pictures. Go out of your way to find something interesting and singular to your city. Explore odd angles and the tiny details of objects that other people would never notice. Use symbolism and metaphor. Innovate!
 
To show off the usefullness of the HDR/DRI technique, I shot a simple night scene which was made complicated by neon lights and text.  You could shoot a similar scene using a model, too, and it would look great – but make sure to tell him or her to pose in a comfortable position, and be prepared to remain as still as possible for several minutes.
 
Below is the type of edited image we’re working towards. All the neon text is readable, the lights look natural because the bulbs are defined, and we can even see the night sky. The scene is balanced and the important details are preserved.
 
Read on to see how I got this:
 

 

 

HDR Shooting Guide for Night Scenes

Objective:

Obtain six useful pictures that together represent a wide dynamic range.

Step 1:

Set up your tripod and mount your camera.

Step 2:

Set your camera’s Program Mode to Manual.  Modes like Auto, Aperture Priority, Shutter Priority, and any of the predefined specialty modes won’t work.  The idea is to use your shutter speed as the only variable, which means you’ll need to manually set your aperture to a constant value.

Step 3:

Set your camera’s aperture to a low number, like f/5 – f/8.  Anything lower and you might be overexposing or seeing depth of field problems.  Anything higher and you might need to bump up your ISO, which will give you grainy images. f/5 – f/8 is a perfectly conservative value that should get you home with smooth, usable pictures.

Step 4:

Try out your autofocus (AF).  Compose your image, select your area of interest, and press the shutter button halfway to focus on your subject.  Try to focus several times; if AF is finding your subject consistently, that’s great, move on to Step 5.

If AF is occasionally grabbing different areas of interest, you will get unusable images.  Switch to manual focus (MF) and find your subject the old fashioned way.  The less contrast between your subject and background, the more likely you will need to use MF.

Step 5:

Set your ISO to its lowest possible number. I used ISO 100.

Step 6:

Set your shutter speed somewhere between 10-30 seconds and snap a test picture.  The correct shutter value is going to vary widely according to your available light. You want the shot to be really bright but not all white. Too much white is useless. If you have lights in your scene, you want them blooming uncontrollably.
 
Your goal is to fill all the shadows with detail and to expose the really dark areas. If everything is still really dark after a 30 second exposure, you’re probably isolated from civilization and shooting under a cloud-covered New Moon. Bump up the ISO and try again, or set your shutter to Bulb and wait it out.
 
Here’s my brightest image. Notice that the text on the sign is gone and the bulb definition is eliminated – but the sky is defined:
 

[12mm 30 sec. f/6.3]

Step 7:

Once you get a perfect long exposure – perfect meaning all your scene’s shadows are filled with informative, detailed lighting, where everything looks unnaturally bright – start speeding up your exposures.
 
For example: after shooting my brightest image at 30 seconds, I sped up my shutter to 15 seconds, then 5 seconds, then 2 seconds, then 1 second, and finally up to 1/15 of a second. Eyeball each result and keep the increments that work for you.
 
Here’s the darkest exposure I kept. The text and bulbs are visible, but unfortunately there’s nothing else to see. Adding the details from the other pictures will change that:
 

[12mm 1/15 sec. f/6.3]

You want to shoot a total of six JPEGs. When you’re editing, you can use them all or you can use only two of them – but having six pictures to work with will give you some flexibility.
 
Below is an example chart of shutter values and their corresponding subjective descriptions. You may find that you need to use different shutter values to get the look you need, and that’s okay. We’ve already dialed in our constants, so don’t change anything but the shutter speed to get results. Be careful not to shake the tripod as you make adjustments.
 

Example Shutter Speed

Desired Look of Picture

12 Seconds

Very Bright: Shadow depth eliminated or severely reduced. If available, sky and background are very visible.

8 Seconds

Bright: Shadow depth significantly reduced and lights still blooming. Sky and background potentially fade.

4 Seconds

Normal: The most balanced compromise. Out of your six pictures this one is the most evenly exposed.

1 Second

Slightly Dark: Shadows are thick and image details are reduced. Lights no longer bloom, but aren’t yet perfectly exposed.
1/3 Second Dark: Shadows dominate. All details are scarce, except on objects and structures near a light source. You can clearly read neon text.

1/6 Second

Very Dark: Image is way underexposed, but the subtlest details of the lights emerge. You can clearly see bulb shapes.
 

Step 8:

 
Once you’ve got six pictures that represent a wide dynamic range, get a cup of coffee, go home, and copy the JPEGs to a new folder called “PICS” on your CS2-ready machine.
 
There are several ways to get the result we’re looking for in Photoshop, so let’s set our standards low and look at the not-so-good method first.

The Short Answer:

For Realism, Avoid Adobe’s “Merge to HDR”.

Photoshop CS2 has a built-in feature that allows you to merge many files to a single HDR image. It requires very little effort to do and it kinda shows in the end. You also need a black belt in Histogram to use it.

 

 

 

 

 

The “Merge to HDR” Guide

Step 1:

Open CS2 and click to bring up the Merge dialogue box.

Step 2:

Select Folder under Use, and Browse to the PICS folder where you copied your six images. Uncheck the box to Automatically Align Source Images. You used a tripod so your images are already aligned.  Click OK and you’re off.

Step 3:

Depending on your computer’s specs and the size of your files, it will take some time for CS2 to run the voodoo down. When it finishes, it will spit out a nasty looking preview image with your source files on the side.  Everything is 32 bit and useless to us. To see it properly we need to convert it to 8 bit glory from the Bit Depth pull down.  Select 8 bit/Channel and click OK.

Step 4:

Now you’ve got something worse looking than before.

The HDR Conversion dialogue has appeared, and there is a Method pulldown where you can select from several ways to butcher convert your image to 8 bit. Play with all of the options. I dare you to find anything useful that doesn’t leave you with Saturday Morning Cartoons.

Our Specific Problems:

  1. There’s no natural definition in the neon “VIPS” sign, or in the text under the light on the left side of the building.
  2. Depending on what method you select, the sky will either look unnatural or turn black.
  3. All the colors are plasticky and generally horrendous.

Our Non-solutions:

Your first choice of conversion method is based on modifying the Exposure & Gamma values – but so what? You could’ve shot this in RAW instead of JPEG if you wanted to change your natural exposures artificially. Combining the desirable traits from each of your six images is your goal, and that really has nothing to do with this.

The second and third choices are Highlight Compression and Equalize Histogram. These run predetermined conversion formulas that you cannot adjust manually, which makes them both useless and cartoony. You can’t specify any kind of numeric value, and CS2 doesn’t ever ask for one.

If there is a correct answer to the “Merge to HDR” automation riddle, I suspect it lies in the fourth choice: Local Adaptation. Here, you’re given the ability to drag sliders around on a histogram, and if histograms are your forte, you may feel right at home. I use them to confirm captured data (reading extreme highlights/extreme shadows) in my camera, but I don’t try to manipulate them in Photoshop because they rarely yield the results I need.

Step 5:

Ignore the “Merge to HDR” automation feature of CS2 altogether, and try the much-better method of creating HDR images on the next page!

 

Increasing the Dynamic Range Manually

 
Some call it Tone Mapping, others call it Contrast Reduction; I call it DRI Awesomeness. This technique produces excellent results for most HDR applications. It takes more time and attention than the Merge to HDR automation, but “when you shoot it, you’ll know where that extra money went.”
 
In a nutshell, we’re going to take our six images and split them into two groups of three. We will blend the three brightest photos and save the resulting image as “Light,” then we’ll blend the three darkest photos and save that result as “Dark.” Finally, we’ll blend these two new groups of Light and Dark into a single image called “Even.

 

The DRI Guide

Step 1: Rename to ease workflow.

Locate the PICS folder you created and rename each image with a whole number on a scale of 1-6.

1 = the severely underexposed picture.

6 = the brightest picture.

Please don’t email me saying “Well, I think I’ve got a 2.5 here and 3.7 here and maybe two sixes.” Look at them side by side. If you followed the shooting guide they should all look different, and we need to give them hard values of 1, 2, 3, 4, 5, and 6.

Step 2: Open the three brightest pictures.

  1. Open CS2 and click to bring up a dialogue box. Navigate to your PICS folder.
  2. Click on the image marked 6.jpg and then [Shift+Click] on 4.jpg to select images 6.jpg, 5.jpg, and 4.jpg all at once. These should be your brightest images.

 

Click Open. Your screen should look similar to the shot below, with three new windows containing your three brightest images.

Step 3: Make the three pictures as layers in one project.

Let’s stack them in a useful way; i.e., your brighest layers on top of the darkest.

  1. Select the MOVE TOOL (V) from the Tools palate.
  2. Click on the image anywhere in the 5.jpg window, and drag the entire 5.jpg onto 4.jpg.
  3. Make sure you line the pictures up perfectly! It helps to turn on the Snap feature using [Shift+Ctrl+;].
  4. Close the 5.jpg window. You should now have only two open windows, 6.jpg and 4.jpg, and it should look like this:

Step 4: Make the three pictures as layers in one project, continued.

Repeat Step 3, but simply substitute image 6.jpg for 5.jpg.

Now you should have one window open that contains all three pictures with your brightest image on top and the darkest on the bottom.

Look at your Layers Palate in the bottom right corner of CS2. Your darkest layer should be labeled Background in the Layers Palate. It’s locked, so we need to duplicate it so we have something to edit.

  1. Right Click on the layer text that says Background and click Duplicate Layer.
  2. In the dialogue that pops up change the As: field to 4 instead of Background Copy, and Click OK.
  3. While you’re in the Layers Palate, rename Layer 2 to 6 and Layer 1 to 5.

You should have something like this:

On the next page we’ll use the Color Range feature to blend these together.

Step 5: Selecting the Color Range

Here’s where the magic happens:

1.  Disable/Uncheck the eyeball in layers 6 and 5.

This is important; if you don’t disable the two layers above 4 you won’t be able to see your work.

2.  Now select layer 4 in your Layers Palate.

3.  Click .

4.  In the Color Range dialogue box choose Shadows from the pulldown and Click OK.

You’ll get some dancing lines all over your image. This selected area is the area to be edited, but its default definition is too strict and we need to make it look more natural. To do that, we’ll Feather our selection.

Let’s discuss the Feather Radius number you’ll use. Some clever DRI photographers like Andreas Tofahrn create HDR images in a similar way to this. Andreas uses the function in Photoshop instead of Feather. I Feather my images because I prefer the results. No matter what method you choose, an important concept to consider is his rule of “halving the width for each processed layer” when choosing the Sample Radius or Feather Radius value. I will halve my value when necessary, but I didn’t use that method here because this should be organically reductive.

5.  Click [Select ► Feather]. I used a constant Feather Radius value of 100, but you should use what looks good on your pictures. Enter your value and Click OK. You will see the selected area change significantly:

6.  Now we need to create a Mask from this selection. Look at the Layers Palate and locate the Add Layer Mask button. There’s a fancy green arrow pointing to it in the above image. Click Add Layer Mask, and your layer 4 will link to a ghostly rendition of itself the Layers Palate; it shows your saved shadows, and it looks like this:

Step 6: Repeat Step 5 using layer “5” instead of layer “4”.

  1. Select layer 5 and activate its eyeball so you can see what happens.
  2. Click [Select ► Color Range].
  3. In the dialogue pulldown choose Shadows and click OK.
  4. Click , and enter the value you used on layer 4, or use half that value if the results look better. My starting value was 100, and I stayed with it.
  5. Click the Add Layer Mask button in the Layers Palate.

You should get another ghostly thumbnail, this time linked to layer 5 in the Layers Palate.

Step 7: Repeat Step 6 using layer “6” instead of layer “5”.

Try this one from memory. If everything goes well you will see all your layers linked to masks, with something like this in the Layers Palate:

Notice, too, how the exposure of the sky has brightened and building details are starting to even out.

Step 8: Save your image as a new JPEG.

Click [File ► Save As] and select JPEG from the Format pulldown. Name it “Light”. CS2 will flatten your layers automatically and save your file.

Finishing the Image
 

Step 1:

Now that you have your Light.jpg, you need to make your Dark.jpg. Use the same DRI method we’ve covered, but instead of combining your three brightest pictures (6,5,4), combine your three darkest (3,2,1). When you finish, choose Save As, and name that result Dark.jpg.

The last step is to combine the two extremes.

Step 2:

  1. Open your Light.jpg and Dark.jpg.
  2. Place the Light.jpg on top of the Dark.jpg using the Move Tool (V) just like you’ve done before, and make sure they’re lined up.
  3. Use the Color Range/Masking DRI method. You don’t need to duplicate the background on this final step.
  4. Save the finished image as Even.jpg.
  5. Crop it, tweak the colors to your liking, and tell everyone how Bjorn3D changed your life.

The Inevitable Drawback:

All images are different. I’ve tried to show you a method that will always preserve the readability of your text and give you clean bulb lines.

Here, you’re paying for all this detail with the occasional loss of brightness. I think it’s a fair trade.

Final Thoughts

HDR lighting upped expectations of what video games can look like, but many people still don’t know how practical the concept can be for photographers. The goal is not to make some super k3wL off-the-wall image in Photoshop and unleash it on the denizens of Flickr; they’ve already got that problem ad nauseam. The goal is to make a realistic photo that captures details you could never get in a traditional photograph, and to show it to someone and hear them say, “Great exposure, how did you do it?”

If the “Merge to HDR” or DRI technique interests you, let’s see it. Get out your brainstorming caps and post your best image in the forums. Have fun with this and happy shooting!

[14mm f/8 ISO 100. Six image DRI composite used to expose the ceiling as well as the neon.]

[50mm f/4 ISO 100. Two image DRI composite. The brightness of the sky was balanced enough with the neon to capture a good single shot; however, the “Savannah Georgia” text was unlit, and required a second longer exposure to make it appear.]

Check Also

Geforce Now and the Apple Mac Book Pro (M1 Pro) – a match made in heaven?

Jump to section 1. Nvidia GeForce Now – the basics 1. Nvidia GeForce Now – …

NVIDIA introduces DLDSR – an intelligent downscaler

We have written a lot about Nvidias DLSS-technology that uses AI to upscale images to give a boost to the frame rate while trying to still offer great image quality. While I think it offers great image quality I know there are gamers who would would love to get a way to use AI to instead downscale a higher resolution image to give better quality at the lower resolution. Well, they are in luck as Nvidia now has presented Deep Learning Dynamic Super Resolution, DLDSR (phew, say that fast a few times).

Leave a Reply

instagram default popup image round
Follow Me
502k 100k 3 month ago
Share