Guarding the Beaches of LA, by Stuck in Customs on Flickr

Old solution to this problem

If you were facing such a scene in the old days of film, you had to analyze the scene and determine what the key tones were. Often, you needed to hold the highlights and allow some clipping in dense shadows or other dark areas. Sometimes the key tones were dark and you were forced to allow blown highlights. And in some cases, you would decide that the scene couldn’t be compromised. In other words: you couldn’t capture it the way you wanted, so you didn’t!

Thank the spirits we now live in a digital age! Now you can capture this scene using HDR.

My two cents in the HDR argument

Before I move on, I want to address the fact that HDR is frowned upon by some people. There are several reasons for this. Some claim that bracketing or multi-exposure shooting is counter-skillful (elitists). Others think that using HDR is “cheating”, creating a falsified representation of “reality”.

Ignore those people.

HDR is not cheating. Nor is it the path to enlightenment. It’s just another tool in your photography toolbox, to be pulled out when you feel the situation demands it.

Enough ramble! Let’s get on with the subject!

What is HDR photography?

HDR photography is the principle of using multiple shots to capture a high range scene, and then merge these shots together digitally on a computer during post-processing.

The process involves using specialized software tools. These tools will use the different exposures to create an “optimal” representation of the scene, ensuring that the clipped shadows or highlights are excluded from the resulting image. A little simplified, this enables you to expose detailed information in the shadows from the brightest exposures, and details in the highlights from the darkest exposures.

Old suitcase scene analysis

As you can see in the above image, the scene has a high range between the sun on the wall and the suitcase interior. When the interior is correctly exposed, the wall light is completely blown out. And when the sun on the wall is correctly exposed, the photo is underexposed overall, and almost all detail inside the suitcase is lost. To capture this scene, HDR is necessary.

HDR is in its infancy

HDR is still in its infancy. This means that although we’ve had it a few years in still photography, the tools and techniques are going to grow a lot more mature in the future. Camera manufacturers will definitely make use of HDR in some sort in future models to increase their cameras’ ability to cope with high dynamic range scenes.

And also, the tools for post-processing will grow much more mature. Still, some of the tools we have today are already relatively sophisticated. With proper knowledge we are able to create high quality HDR images.

Building the scene

Our own vision is rather limited for taking in a scene as a whole. But our mind has a wonderful way of compensating for this. Our eyes rapidly move around the scene, registering it bit by bit, and our mind then combines these visual impressions into one experienced scene. The brain is also an expert in completing known things where only a part is visible. If we see a car where half of it is hidden by a corner, our brain completes the car for us.

When moving around like this, our eyes quickly adjust to the light and tones of the different areas. And that is pretty much the same as we do with multiple exposures for HDR, except that the camera captures the whole scene in every shot. But we expose for limited areas with each shot.

So, in a sense, one might argue that HDR is more “natural” than capturing a scene with one shot!

Capturing the scene

First and foremost; if you have the choice: use a tripod! Slight movement from hand-held shooting can be compensated by the software, but you will be so much better off by making sure your image data is of the highest possible quality!

How many exposures you need will depend on the given scene. Often, around five exposures are needed. Today, most SLR cameras have a clipping indicator you can enable on the screen, and most also have the option to show a histogram.

Start out by under-exposing the scene. Under-expose to such a degree that the histogram is jammed up on the left side of the histogram, clipping the shadows. Slow down the shutter 1 full stop, and take a second shot (yes, the shutter; the aperture is fixed during HDR shooting!). Slow down another full stop, and take another shot.

Continue shooting in full-stop steps until the darkest part of the histogram is around the center of the scale. (Yes, that means that I should have shot at least one, maybe two stops more in the example shot!) This ensures you record all details in the darkest shadows.

HDR shooting sequence with histograms

Pitfalls when shooting for HDR

There are some pitfalls when shooting for HDR. Among the most common are ghosting, flare, banding and haloing. Ghosting is caused by movement, flare by light from a bright light source. Haloing and banding usually occurs when there are sharp edges between light or dark in the scene. Common examples are around the edges of a window in interior scenes, and at the edge between a dark building wall or roof and a bright blue sky.

I will return to these pitfalls in a follow-up article on HDR tonemapping.

Merging into a HDR image

There are several tools on the market for generating HDR images. I will focus on Photomatix as this is the oldest (and probably best-known) tool out there.

In this introductory article I will not go into details on how to tonemap. I only adjusted a little, mostly the saturation slider. You can experiment with this a little on your own. In a follow-up article I will delve deeper into the technicalities of producing the best possible result using tonemapping.

In Photomatix, click the Generate HDR image shortcut. In the following dialog, browse and select the sequence of photos.

The next dialog is the Options dialog. I recommend that you always check the “Align source images” choice, even when a tripod was used.

Then click the “Generate HDR” button. Photomatix then merges the photos and generates an HDR image. The resulting temporary image then shown is unprocessed and does not show the full range. Your next step is to click the “Tone Mapping” button. This will produce an image where you see the full details in both shadows and highlights.

As stated previously, this article will not cover the tone mapping process. Feel free to play around with the various sliders and see how they affect the photo.

Old suitcase HDR (non-optimized), by Marius Waldal

As you can see above, the various exposures have merged into one photo where both the highlights and the shadows are held, and all details are shown.

I hope this introduction was helpful and that you are hungry for more! I will revisit this subject in a later post with more detailed info on how to perform successful tone mapping!

If you liked this post, you may also like:

1. How to simulate cross-processing in digital photography
2. How to create a black&white image with selective colors in Photoshop
3. Introduction to digital black & white photography

Red tractor, by Marius Waldal

What is cross-processing?

Cosplay-Zoku, by bass_nroll on Flickr

Today, anyone (that means YOU!) can cross-process their photos. What’s even better: it’s actually relatively simple! I will guide you through the process using Photoshop. I have to use my old CS2 version since Elements 8 for Mac does not include the Curves tool (bummer).

How to cross-process digitally

How much work you do, and how hard you adjust will always depend on how you want the resulting image to look. In this tutorial I will strive for a “LOMO-look”, so the adjustments will be fairly thorough. If you want a more subtle change, feel free to downscale my adjustments a little. Trial and error is the key anyway.

We’ll start by increasing global contrast. If you chose a high-contrast photo, you may choose to skip this step.

Open your desired photo and create a new curves layer (Layer -> New adjustment layer… -> Curves). I name this layer “Contrast”.

In the Curves window, create a shallow S-curve to increase overall contrast:

Increasing global contrast

Now, repeat the process of creating another curves adjustment layer. Name this layer “Color”.

In the “Channel” drop down where it says “RGB” you now choose “Red”. Adjust the curve like you did in the last step with a classic S-curve.

Adjusting the red channel

Change the channel to “Green” and repeat the previous step.

Lastly, for the channel “Blue” you create a reverse S-curve by lowering the highlights and heightening the shadows:

Adjusting the blue channel

The image is now cross-processed, but we still need to make some adjustments to make it more “LOMO-like”.

LOMOification – Sharpening

We’ll start by sharpening the image, creating a “harsher” look.

Start by merging your layers (Layer -> Merge visible if you’re using CS2).

Next, convert the image to Lab color (Image -> Mode -> Lab color).

Now, open the Channels palette and click on “Lightness”. The lightness channel is now the only selected channel:

Use the Unsharp mask to sharpen the image (Filter -> Sharpen -> Unsharp mask…). I usually go for an amount of 40% and a radius of 40 pixels, and a threshold of 0 levels. Try different settings and find the look you like.

Unsharp mask values

Change back to RGB-mode (Image -> Mode ->RGB Color).

To make the photo even more “LOMOified”, you might want to add a vignette. The LOMO is (in)famous for generating quite a bit of vignetting. I must admit that it has its charm. So if you want, go ahead and add a vignette as a last step!

LOMOification – Adding a vignettte

Start by duplicating your background layer (Layer -> Duplicate layer). I name my layer “Vignette”.

With this layer selected, choose the “Elliptical Marquee Tool”. Place the marker about 1cm down and right of the top left corner, click and drag the ellipse all the way down to 1cm above and left of the bottom right corner. The ellipse will now cover most of the image, having a margin of 1cm to all sides.

Next, click the “Add layer mask” icon at the bottom of the Layers palette (the grey square with a white circle inside). On the Vignette layer in the Layers palette, a Layer mask thumbnail appears (and yes, the ellipse marquee disappears). Make sure the Layer mask thumbnail is selected by clicking it.

To get a better visual response when doing this step, I prefer to hide the background layer temporarily (clicking the eye at the far left of the layer in the Layers palette).

Adding vignette using Layer mask

Create a gaussian blur (Filter -> Blur -> Gaussian blur…). Set the amount of blur you want. I feel that around 30 pixels are fine. Click OK.

Now do the following: Unhide the background layer by clicking in the empty box where you clicked the eye a few moments ago. The image now looks just like it did before we started creating a vignette.

Select the background layer by clicking on it (important!).

Create a new fill layer (Layer -> New Fill Layer -> Solid color…). Call the layer whatever you want, or accept the default.

Click OK, and a Color Picker appears. Pick a totally black color and click OK. Now, in the Layers palette, click the Opacity field and reduce opacity to your desire. I usually go for an opacity between 40 and 50%.

Creating the vignette

And you’re done!

Compare the original with the cross-processed one. Like the effect?

Original vs cross-processed

Not all photos respond equally well to cross-processing. Trial and error is again the key. With more experience, you will learn to see which of your photos that may benefit of cross-processing.

Afternote – How much to adjust?

There is no right or wrong here, as this depends on the effect you want to achieve and the photo in question. Both the light, mood and colors of a photo will affect the adjustment amount. The amount shown on the screenshots here is a guide. Play around a little and decide what works best for you.

Was this tutorial easy to follow? Was anything unclear? Please leave a comment and I will do my best to clarify!

As always, we would love it if you share your photos after following this tutorial. Hope to see you in the comments section!

If you liked this post, you may also like:

1. HDR (High Dynamic Range) photography – An introduction
2. How to create a black&white image with selective colors in Photoshop

Round and round, by Marius Waldal

There are several ways to accomplish the same thing, but the method I describe here is maybe the most “photoshopish” way.

Find yourself a photo that has at least one element that stands out with a bright color (or several colors). Or you can download the test photo from here.

Original image

What we need to do is create a new layer where we will do the work, leaving the background layer untouched. Create a new layer by clicking on the menu item Layer -> New Adjustment Layer -> Hue/Saturation

This brings up a dialog asking for a name for the new layer. Just accept the default.

Then, in the Adjustments palette, make sure “Master” is selected in the dropdown, grab the Saturation slider and slide it all the way to the left, effectively removing all color from the image. (This is not the recommendend way of creating a black & white image, but it serves our purpose here.)

Painting the color back

I prefer to zoom in quite a bit and work on the edges first, leaving the “fill” for last. Then I can use a pretty large brush to paint the rest.

And that’s it! It’s really simple and if you’ve been reasonably careful at the edges, your final image should look something like the large image below.

A few words of advice

To maximize the effect of selective colors, make sure you don’t color too much of the image! The effect is usually at its best when only a small part of the image is colored. And that part should preferably be brightly colored.

Be warned also that color in itself is a strong attention grabber. If you color a part of the image that is unrelated to what you really want to convey with your image, you increase the chance that your message is lost!

Yes, this means that the example working image I’ve used here is not a good example. The colored part takes up too much of the overall space, effectively robbing all other parts of the image of the attention. On the other hand, there are no other parts of the image that needs attention. So it works for illustration. The feature photo at the top is much better in that regard.

Also, be sure to check out the Flickr photos at the bottom of this article!

Final image

Great Flickr photos with selective colors:

Strawberry and milk, by Nate Flickinger on Flickr

Even though the berry takes up rather much of the frame, I think this photo works because the berry is the main attraction, and the splash does not lose it’s impact by the fact that the berry is red.

De Verleiding, by soleá on Flickr

I think this photo is great for selective coloring for several reasons:

• It is simple, so there are few elements to disturb the message
• The implied line from her eyes to the apple is a strong leading line
• The apple is small in the frame, yet it is the main subject. Being the only colored object emphasizes its importance even more

A couple of other photos coming (awaiting photographer approval)!

If you liked this post, you may also like:

1. How to simulate cross-processing in digital photography
2. HDR (High Dynamic Range) photography – An introduction
3. Introduction to digital black & white photography