How to get a perfect exposure using the histogram

Saturday, January 30, 2016

How to get a perfect exposure using the histogram


Description

This is the second in a series where Harley talks about histograms as they’re used in photography. Having problems getting a good exposure based on the preview image of your camera? In this episode we look at how to use the histogram to get the exposure you want for your images.

This episode looks at exposure as it relates to the histogram: what is "proper" exposure, how different types of images impact exposure and the histogram and finally how to see exposure problems on the histogram and what to do to correct them.

Previous video: How to easily read a camera's histogram
Playlist of histogram related videos.

For a written transcript, go to go to How to get a perfect exposure using the histogram

Music under Creative Commons License By Attribution 3.0.
Intro/Exit: "Hot Swing" by Kevin MacLeod at http://incompetech.com
Light switch effect: http://freesound.org/people/AlienXXX/sounds/151347/

Transcript

Struggling to get the exposure you want? Today at the House of Hacks we’ll look at how you can use your camera’s histogram to get the perfect exposure.

Hi Makers, Builders and Photographers. Harley here.

The histogram is an important tool to understand the actual exposure of an image. When working with people in classes and workshops, I've noticed many of them look at the picture on the back of their camera trying to evaluate the exposure but don't have the histogram turned on. The problem is the image displayed on the back of the screen does not represent the actual exposure saved in the raw file. This preview is filtered based on several camera settings such as picture mode and white balance. Also, that screen is tiny and uncalibrated. It's really only a rough approximation of the actual image. Now don't get me wrong, the preview is a valuable tool; it’s great to see the overall composition. However, it's nearly useless to evaluate the exposure.

In this previous video, I explained what all the information on a histogram means. Today we'll look at the exposure as it relates to the histogram: what is "proper" exposure, how different types of images impact exposure and the histogram and finally how to see exposure problems on the histogram and what to do to correct them.

Somewhere in the history of photography, someone decided a "proper" exposure had a histogram that looks like this. It assumes that on average the majority of the luminance information is in the center of the image. Another way of saying this is that 50% gray will make up most of the image.

So, the exposure computers in cameras look for a peak in the image and try to adjust camera settings to put the peak in the center of the histogram. And this works well if you have an evenly lit, average subject. In an image like this, this rule works.

But this isn't necessarily valid for images that fall outside of average. And for artistic purposes, it’s not at all unusual to want something intentionally different.

For example, in low-key images, where a large percentage of the image is dark, this peak should be skewed to the dark side of the histogram on the left. But the exposure computer will try to push it to the center, making what should be dark, a washed out grey. This isn’t necessarily a bad thing, and in a future video I’ll talk about why you might want to intentionally do this. However, if you try to get the exposure correct right out of the camera, it doesn't give you what you really want.

Conversely, in high-key images, where a large percentage of the image is bright, the histogram should have the peak skewed to the right. But again, the computer will try to push it to the center, making what should be bright look dingy. Frequently you’ll encounter this problem shooting bright scenes outdoors, like at the beach or in the snow.

Unless the overall dynamic range of the scene is greater than your camera, generally both these problems can be fixed in post-production. However, both these problems are easily detected and corrected in-camera using the histogram. If you’re aware of the conditions you're shooting in and the look you're trying to achieve, you can know where the peak is supposed to be. Knowing where you want it, you can look at the histogram to know if it's in the correct spot or not.

If you find it’s not where you want it, and you’re using an automatic mode, use the exposure compensation setting on your camera. It allows you to tell the camera to adjust where the peak rests relative to center. By setting it to the plus side, you're telling it to move the peak to the right to brighten the image, or from the camera's perspective, you want to overexpose the image. By setting it to the minus side, you're telling it to move the peak to the left to darken the image, or, again from the camera's perspective, to underexpose the image.

If you’re in manual mode, simply adjust one of your exposure settings, f/stop, shutter speed or ISO, to make your image brighter or darker. If you're using your camera's exposure meter, the computer’s “perfect” exposure will be with the indicator right in the middle. If the image with this exposure is too dark for your scene, adjust your settings to move the indicator to the right. Or if the image is too bright, just move the indicator to the left.

Once these adjustments are made, make another image and re-evaluate the histogram. Repeat this process until the peak is where you really want it.

A significant problem in digital photography, that wasn't so much an issue with film, is clipping. With digital imaging, once a part of the image gets to a certain brightness level, it can't go any higher and is simply cut off. Everything brighter than that level is set to that maximum. It's clipped off. This causes a loss of information. The same type of thing happens in the dark regions. There's a level that means "black" and once that level is reached, anything less than that is set to black.

Imagine a number line that's really big in both directions. This is reality. But a digital sensor can only capture a region of this infinite line that has 256 values. Anything less than 0 is treated as 0 and anything greater than 255 is treated as 255. The number line is just cut-off, chopped, or clipped. Within a certain range, using exposure controls and ND filters, you can adjust where on the line this camera defines the start and end points, but you can’t change the size of the window.

Fortunately, clipping is easily detected by the histogram. If you have a tall vertical line on the extreme left or right edge, that image was clipped. If you see this, and you know you shouldn't based on your subject, you need to adjust your exposure either up or down, depending on which side of histogram the problem lies. Some cameras will flash areas of an image that are over and underexposed.

Just keep in mind, there are some cases where this can be expected to happen. A landscape image with the sun in it, or something with specular highlights, will probably have a small amount of clipping. This is normal and to be expected since these are the brightest areas of the photo. But if it's a wedding dress or white shirt, that's definitely a bad thing because any clipping will result in a loss of detail in these areas.

In conclusion, I’d love to hear in the comments below about how much you’ve actually used the histogram in the past. And, if you have used it, how it’s helped you.

Here are two playlists. One has all my photography related videos and the other contains other histogram related videos.

If this is your first time here at House of Hacks: Welcome. I’m glad you’re here. We’d love to have you subscribe. Through these videos I hope to inspire, educate and inform makers in their creative endeavors. Usually this involves various physical media like wood, metal, electronics, photography and other similar materials. Thanks for joining me on our creative journey. So subscribe and I’ll see you again in the next video.

Now, go make something. It doesn’t have to be perfect, just have fun!