From the course: Photography Foundations: Exposure (part 1)

How to read the histogram

From the course: Photography Foundations: Exposure (part 1)

Start my 1-month free trial

How to read the histogram

- Now that you're intentionally over and underexposing, you have a great level of control over a number of different things exposure wise. You can ensure that you don't overexpose your highlights. You can make certain that your shadows have the level of detail that you want. But here's the bad news, while it's fantastic that you have the little screen on the back of your camera for reviewing your images, it's very important to understand that that screen is only good for judging composition. Really it tells you nothing at all about color or exposure because the image on the screen is brightened and saturated by the camera to make it easier to view in bright light. So just because an image looks good on the camera, that doesn't mean that you've actually got a good exposure. In the old days when facing a difficult exposure situation, photographers had to bracket their shots. They had to shoot the same image multiple times with different exposure settings. And there are still a lot of times when that's a good tool for a digital shooter. But fortunately as a digital photographer you have an additional tool in the form of the histogram. You might have seen a histogram display on your camera or in your image editing application, and it might have looked intimidating and math-like. But don't worry, it's actually very simple. Here's an image of a gray scale ramp. I did not shoot this, I generated in Photoshop. So I told Photoshop to make a gradient that goes from complete black to complete white. Here is the histogram of that gradient. A histogram, again, it can look intimidating but it's actually very simple. It's nothing more than a bar chart. It's a bar chart of the distribution of tonal values within an image. So on the left, over here on the far left side, I have black. Over here on the far right side I have white. In the very middle, I have middle gray. So what's happened is that Photoshop has gone through this image and counted the number of black pixels and made a bar for that on the very left hand side. And then it's gone through and made a count of almost black, like 99% black, and it stacked up a bar for that and it's done that all the way through to white. And so if you think about each column in this histogram as being just a column in a bar chart, it starts to make a little more sense. Now in the ramp you can see what appears to be an equal distribution of black to white but on the histogram you can see that I've got a lot more black than I have middle gray. And I don't have as much middle gray as I have white. Shouldn't the histogram just be a straight line all the way across? No, because of the way our eyes work. Our eyes do a fantastic job of expanding our sensitivity to very dark tones and very light tones. So we're actually able to see more shadowy tones and more really bright tones than we are able to see middle gray tones and so the histogram is reflecting that. The histogram is showing us a graph of the perception of our tones in the image. Our eyes have this ability because it's very important that we be able to peer into dark shadows and look into a bright light and still see detail so that something doesn't eat us. That's rarely a risk when you're working in Photoshop. So what I can do with this histogram is tell immediately if I've got tones where I want them. There is no correspondence between a geographic point in your image and a point on the histogram. If I flip this gray scale ramp like this, what does the histogram look like? It looks exactly the same because I haven't changed the distribution of tones in the image. There's still just as much black as there was before. There's still just as much middle gray. There's still just as much white. So the histogram ends up looking the same. So now let's look at a histogram with more tones in it than just black and white. I have here a Canon SLR. I've got this cable coming out of it because I'm capturing the feed from the back of the camera so that you can see it. And as you look at that, you'll see the view that I'm visualizing here with my camera. It's all out of focus right now but that doesn't really matter. I've still got all the tonality of the one. I'll just go ahead and focus it. There we go. I have table that has a gray tablecloth on it that's pointed towards a gray wall, and I've got the live histogram feature up on this camera. When I'm shooting in live view on this camera, I can ask it to give me a real time histogram that changes as things in my scene change. So what have we got here? We've got a histogram of an image that's mostly gray. So you'll see there's that big hump of data right in the middle of the frame. That is representing the tonality of the tablecloth and the background. There's a smaller hump to the left. That's most likely the shadow over there on the left side of the background. And you'll notice there's some darker tones at the rear of the table. There's also a seam that you can see that has some darker tones in it. And so those are all being represented there. What I can see from the histogram is that there is nothing in this image that is black. I know that because there's no data over there on the left hand side. Even though there's a really dark shadow in the upper left hand corner of the table, it's not actually black. It looks black to my eye but the histogram is telling me it's not, it's just a very dark gray. I can also see that I have no white tones in the image because there's no data on the white side of the image.

Contents