The Histogram is a simple graph that shows what is going on in your image.
#1 - I have added the bottom gradient bar to help you see what tones are represented across the width of the graph. We have black at the far left and pure white at the far right, and a progression of tones in between. So the horizontal scale represents the brightness or darkness of the tones in your image. You can visualize your image on the histogram as the darker tones are shown on the left and progressively brighter tones to the right, until you finally reach the brightest tones in your image on the far right. Once you get used to it you can tell at a glance a lot about what is going on in your image.
The vertical dimension of the graph simply shows how many pixels exist in the image of each tone. We don’t actually count the pixels. It just means the higher the graph the more pixels of that tone there are in the image. For instance the larger hump in the middle left represents a darker than middle grey tone. That would be something like green evergreen trees while the next big hump to the right of centre would be something like green grass (lawn). The large hump just shows there is a lot of that tone in the image We aren’t much concerned about the shape of the graph in the middle, except perhaps for curiosity sake. What we are primarily interested in is the width of the graph and how it is placed inside the graph. In this case we at least have all the image data inside the box. That’s where we want it. This looks safe, or OK.
One problem arises with some histograms in camera apps. The outer limits of the “box” are often not clearly defined so it is hard to tell when you have reached the end. That really is a poor design. I like to see the edges of the working space clearly defined.
#2 - Here is an important point to realize. The far ends of the graph right and left represent pure featureless white on the right, and solid black on the left. If you are hoping to see some detail in your brightest tones then the little vertical black line on the bottom right represents the farthest to the right the highlights should be. Otherwise they will have no detail. Likewise at the black end of the scale on the left. Anything beyond the little vertical black line will not have any shadow detail. Naturally, those little black lines are not even shown on the histogram. You just have to know more or less where they should be.
In simplest terms that means to be sure you record highlight detail in your image you should have a little safety zone on the highlight end. You also need to pay attention to what tones actually are in your image. It may be that your image doesn’t have any bright tones but has lots of low and middle values. In that case letting the middle tones creep into the area where highlights belong indicates over-exposure.
If you ever have to choose between the highlights or the shadows I would always opt to preserve the highlights. There’s nothing worse than burned out highlights.
Here’s another important point. On iOS devices the histogram shows JPEG values. If you are shooting RAW it is important to realize that the RAW file can contain a wider range of tones than a JPEG and that means it is considered “safe” to let your highlights bump up to the far right of the graph and you just have to have confidence that your highlight values really will be recorded. This is why shooting RAW & JPEG at the same time is a poor strategy. They have opposite exposure strategies. When photographing in RAW mode you want to minimize image noise which is more prevalent with underexposure. This is where “expose to the right” comes from.
As I write I can now only see image file names so I have to remember which order the images come in.
#3 - This is what happens when you make an underexposure. All your image data is dragged to the black end of the scale. The data that has gone beyond the box will all be solid black -bye bye shadow detail.
#4 - this is how the histogram actually looks when you have image data beyond the display region. The dark tones have piled up against the far edge. When you see a pile-up like this pay attention. You have an exposure problem. More about this when I talk about overexposure.
#5 This shows overexposure in action. All your image data is dragged to the right, some of it exceeds the limits of the histogram. Everything outside the box will be cut off and rendered as solid featureless white.
#6 - The part of your image data that went beyond the sounds of the histogram will be cut off and displayed like this. Note the pile-up on the highlight end. Anything that goes beyond the perimeter of the box is simply not recorded - as if it never existed. You can see the significance of that with respect to exposure and image quality.
#7 - People seem to think that anything can be fixed in editing. Here is what actually happens. You take your overexposed picture into an image editor and you adjust the exposure to darken it. All the data that was inside the box will respond well enough to the exposure adjustment. But the part that was outside the box is simple not there to be rescued. It never was recorded in the first place so there’s nothing to recover. You can darken your image until the white areas turn grey and still there will be no detail in them. The image is toast.
#8 - This is what you see when photographing in very low contrast conditions like fog. Increasing the contrast in editing can really help an image like this.
#9 - If you ever see this while photographing it means you’re in trouble. The dynamic range of the scene has exceeded the capability of the camera to record it. Both the highlights and shadows are off scale. It would be a waste of your time to make a photo like this - unless you switch to HDR. Preferably the real HDR with an app that makes 3 exposures, one normal, one overexposed by a set amount, and one underexposed a similar amount. Typically the exposure steps would be 1 stop, or 1&2/3 stops, or even more, on either side of your middle exposure. Then the three images are combined into one extended range image. Ahhh. That’s better.
I’m hoping you can see from this how useful the histogram can be in helping you be sure your exposures will be OK while making exposures with a camera app.
Next I will explore the usefulness of the histogram during editing.