Bracketing exposure can seem brainless especially when you set it to automatic. But, deciding when to do HDR photography requires experience and judgement.
What most photographers have in their mind is: It’s easier to shoot now and delete the unwanted images later, instead of not shoot and end up regretting it.
Although that is true...
But being a good photographer, you want to develop your non-technical skills too - skills to make a good judgment. By knowing when to use HDR photography in your workflow, you naturally become more efficient (and smarter!). It also saves you time by not editing an SD card full of bracketed images unnecessarily.
Now that I have your curiosity, let me have attention too.
By the end of this article, I hope you have a clear idea on when to do HDR photography, how to use histogram to guide your decision and when not to do HDR.
Say Hello To Your Little Friend - The Histogram
The essential tool most photographers use for an objective judgment on when to do HDR is the histogram.
Histogram is the single best tool you can use to decide if bracketing exposure is worth the effort.
Displaying the histogram on the camera’s screen usually involves pressing a couple of buttons on your camera.
I’m a Canon user and it takes me two presses on the Info button to bring up the histogram. Check your camera's manual if you’re not sure.
Although simple to interpret, most photographers look at the wrong histogram. In the next section, I’m going to show you three types of histogram and tell you which one you should use.
Luminance Histogram
This is the histogram you see most often.
Why?
Because in most cameras, this is the first histogram that appears when you press a button. Some cameras can only display this histogram.
The luminance histogram is always in black and white.
Luminance histogram is calculated based on the weighted average of the RGB channels in each pixel. It’s a computed grayscale brightness equivalent of the colors. In simple words, it tells you the brightness (or luminosity) distribution, based on the grayscale version of your image.
Because luminance is calculated by averaging the RGB channels, clipping of individual red, green or blue channel may not show up on luminance histogram.
The major weakness of the luminance histogram leads us to the second histogram that we’re going to discuss - the color histogram.
Color Histogram
Unlike the luminance histogram, the color histogram displays information on the individual red, green and blue channel.
This means, instead of one, you see three different histogram for each channel.
The color histogram displays the individual red, green and blue channel.
It tells you the distribution of brightness in each individual RGB channel. It’s a more useful tool to see if any of the individual color channel is clipped.
You might be very confused at this point.
I know because I was too before I understood how luminance is calculated from RGB. Let’s take a closer look:
Computing Luminance From RGB Data
Your camera uses this formula to convert RGB into luminance histogram for each pixel.
Luminance = (Red x 0.3) + (Green x 0.59) + (Blue x 0.11).
Each color channel is weighted differently because the human eye perceives brightness in green more than red more than blue.
For example, let’s say we have a color with RGB value of 255, 50, 50. To calculate a pixel from RGB to luminance:
(255 x 0.3) + (50 x 0.59) + (50 x 0.11) = 111.5
Remember, this is the luminance of one pixel. Because we have an image with a uniform color, the luminance for this image is the same for all its pixels.
R (255), G (50), B (50).
255 is the maximum value in RGB. A pixel is clipped if it has all or any RGB value of 255. But this is not the case if you look at the luminance histogram.
The luminance histogram shows no clipping of pixels. But in the color histogram, red channel is clearly clipped.
Most modern digital cameras have the ability to display both histograms. Check with your manual on how to operate this.
RGB Histogram
The RGB histogram is the color histograms all superimposing on each other in grayscale.
The RGB histogram above with all individual channel histograms below in Photoshop.
It only exists in image editing software but not your camera. It’s really irrelevant in the context of helping you to decide when to shoot for HDR.
Bracketing For HDR By The Graph
You might be wondering:
How does histogram and when to do HDR photography come together?
In this section, I’m going to show you what kind of histograms warrant bracketing exposure for HDR.
Definitely
To The Far Right
This is when the graph touches the far right of the histogram. In technical term, it’s called highlights clipping, or more casually, blown out highlights.
It happens because the camera’s digital sensor couldn’t perceive the brightest brights in the scene and therefore couldn’t translate the brightness information into any recordable data.
To the far right (highlights clipping). This is the RGB histogram in Lightroom
By default, these uncodable information are all recorded as the same brightness in the camera’s dynamic range. This results in these data being piled up on the far right of the histogram.
To The Far Left
Similar to the above, but in the opposite direction and it’s called shadows clipping.
The darkest darks in the scene fall outside the dynamic range of the camera’s digital sensor. These uncodable information are recorded as the same darkness and these data piled up on the far left of the histogram.
To the far left (shadows clipping).
The data that falls on the far right and the far left of the histogram unfortunately are not recoverable. This means you won’t get any details back in those blown out areas on your image in post-processing.
Definitely, Maybe
The Smiley Face
The smiley face is easily recognized by the U-shape of the histogram.
Some of the smiley faces have both ends touching the far left and right of the histogram and some don’t. When both ends are touching the extreme ends of the histogram , both highlights and shadows are clipped.
Both ends of the histogram touch the extreme ends, clipping both highlights and shadows.
But it’s a gray area when both ends are not touching but very close to the extreme ends. In these cases, the camera’s dynamic range is just about enough to capture the entire range of brightness in the scene.
An image with a smiley face histogram.
If you find the highlights are too bright or the shadows are too dark, you can definite adjust these in post-processing to regain details in these areas.
But there is one potential problem.
If you try to brighten up the shadows too much, you’re not only revealing the details but also image noise.
This is because of low signal-to-noise ratio. I won’t go into the technical details but you can read more about it here.
How about darkening the highlights, would that introduce noise?
If you ever wonder why I call it smiley face, this is why!
As oppose to what you’ve expected, the answer is no.
This is because of high signal-to-noise ratio in the highlights. A technique called exposure to the right (ETTR) uses this principle to overcome problems with image noise in the shadows.
For that reason, I would still recommend bracketing exposure for HDR when your image has a smiley face histogram. This is in regardless of whether either end of the histogram touches the far end.
When To Do HDR Photography?
You have learned which histogram to use (the color histogram!) and when to bracket exposure based on the graph. But none of these makes sense if you don’t know where you’ll actually come across these histograms in real life.
In this section, we’ll explore situations where you can find scenes with high dynamic range for HDR photography.
Landscape
Probably the most common situation to have a high dynamic range scene. Landscape photographers love to capture the beauty of mother nature during the golden hours where the sun is low down near the horizon. The sky is often brightly lit but the foreground is in shade.
In The Shade
This happens when you’re shooting outdoor but standing in the shade. You could be under a roof or even in the shade of a towering mountain. The immediate vicinity around you will be darker than the field away from you in the distance.
Interiors
Most interior photography that includes the view outside through the windows requires either HDR photography or the use of flash to get an overall good exposure. The area outside the windows are always much brighter than the inside.
Into The Light
Some photographers like to shoot into the sun, or the source of light (me!). No doubt, the source of light will always be significantly brighter than the rest of the scene. Bracketing exposure is a must in this situation.
Backlit Scenes
This is when you’re shooting with the source of light coming from behind the subject. You can either expose for the background or the subject but you won’t be able to get both in exposure at the same time. You can either bracket for HDR or light the subject with a flashgun.
When Not To Do HDR Photography?
Similar to how you judge if a scene has a high dynamic range. You use the histogram.
In most cases, if you’re not dealing with scenes outlined above in When To Do HDR Photography, bracketing exposure is often unnecessary.
If you have a histogram that looks like this, you can easily adjust the tones in post-processing without degrading the quality of your image.
You can still bracket the exposure and merge into HDR with an HDR software or exposure blending, but you won’t get much benefit out of it. Nothing will be achieved except wasting your time.
RGB histogram - most pixels are in the midtones and there are no highlights/shadows clipping.
Take Home Message
Whenever you think of doing HDR, take a shot of the scene and bring up the histogram on the LCD screen. Make sure you’re looking at the color histogram instead of the luminance histogram.
If the histogram touches the far left, right or close to both ends of the graph, bracketing for HDR is recommended to capture the entire dynamic range of the scene.
To explore more on histogram, check out my ebook - Dissecting Histogram: An Unwritten Manual from the top menu under Courses.
For more tutorials on HDR photography, check out the HDR resource page!
brilliant!
Thanks! =)
Very good article, thanks!
I hope you find it helpful! 🙂