The Histogram

The Histogram

Histogram - how to use it

In this article I would like to discuss the histogram. Whatever it's for. How to read it correctly and what it's good for you.

What is a histogram in photography?

The histogram is a graphic representation of the brightness values of a digital photo. This allows you to see if your photo has been properly exposed. So whether it's too bright or too dark, because you can't see that exactly on the camera display or on the monitor. A histogram can't tell you if the photo is "good".

Histogram vs. camera display

The display of your camera is deceiving you. Sometimes it's too light, sometimes too dark. In direct sunshine you can't see anything on it. It looks much brighter at night. You can only use the display (view photo) for the screen layout. In no case should you use the display to judge the brightness / exposure. Always use the histogram for this.

How to read the histogram

The histogram of this photo looks like this:

On the left side the dark color values are displayed and on the right side the light ones. The higher the strokes in the histogram, the more color values of a brightness are present in the image.
From left to right a histogram shows the black values, then the depths, then the mid tones, the lights and on the left the white values. In this photo the brightness is well distributed. We have a big "mountain" in the middle which runs out almost evenly at the edges. Thus we have many midtones in the picture.

Directly at the left margin pure black is shown while at the right edge pure white is shown. As you can see, there is no pure white and no pure black in this photo. It would look like the histogram was truncated at one end. In these two color values there is then no "drawing" left. This kind of thing should be avoided if possible. You won't be able to correct these points in the photo later on in the image editing process (or only very limited), because it is only pure white or pure black for your computer. All the details have been lost. And where there's nothing left, you can't get anything.
In a correctly exposed photo there is no pure black or pure white, or only in very small areas.

To illustrate this a little better, I have "blemished" a copy of this photo:

This photo is underexposed and therefore too dark. The histogram is cut off at the left end. This can easily be recognized by the thin bar on the left edge. That's the amount of pure black in the picture.

The lower photo is overexposed and therefore too bright. This time the histogram is cut off at the right margin. In this example, pure white is in some places in the sky.

A few other examples

The curve of the histogram does not always have to be in the middle of the curve. This is not possible and varies greatly depending on the motif.
Here I have two other examples how a properly exposed photo can look like:

This photo is quite dark at the bottom, but not quite black. The histogram shows a lot of values on the left side, but it is not cut off at the left edge. So everything is still o.k. There is no pure black in this picture. The details in the picture are still there.

This bright photo is also correctly exposed. The values in the histogram are strongly pronounced on the right side, but not yet to the edge. There is no pure white in the picture. The details remain unchanged.

Differences between JPEG and RAW files:

To understand how the histogram affects both file formats, you need to know how the files are structured.

Generally speaking, JPEG files are finished images. While RAW files are raw data of the photo. These "digital negatives" still have to be processed / converted with a RAW converter. You can't display / view a RAW file itself.

The cameras generate a small JPEG version of RAW files for preview. And image processing programs such as Lightroom or Photoshop also generate a JPEG as a preview.

And what does all this have to do with the histogram?

With a JPEG file, the image information has already been edited. Cameras that take pictures in JPEG format process the images internally without the user noticing anything about it.
So the photos are "finished", so to speak. And also the histogram resulting from the image is "finished". Thus the photo offers very little reserves if you want to edit it later.
For example, if the photo is too dark and you want to lighten it up, you get noise and the quality drops rapidly.

The situation is completely different with RAW files. These raw data provide sufficient reserves. As mentioned above, RAW files are not yet developed. The information is stored with up to 16 bits per pixel, depending on the camera model. Simply put, the way the light falls on the camera's sensor. First software on the computer or in the camera creates an image from it.

A RAW file therefore requires much more storage space than a finished JPEG image. But it contains all the information in the RAW file and therefore a lot of reserves to edit the image.
You can say that if the histogram is truncated from a RAW file on one side, you still have enough room for corrections when editing the image. Usually these are one or two f-stops.

The RGB Histogram

In addition to the normal histogram, each camera and of course Lightroom / Photoshop has an RGB histogram. This displays the brightness values for each color channel individually. So for red, green and blue. This is because all colours are created in the digital photo. Here you can determine exactly which color values in the image are too light or too dark. In general, this histogram can be used in the same way as the normal luminance histogram. However, there is a more detailed breakdown of which color values need to be corrected. Sometimes it is unavoidable that a color channel is at the end of the histogram. For example, the blue sky in photographs in the blue hour. More than one color channel should not "erode".

Histogram in practice

As written at the beginning, it is about exposing our picture correctly. The histogram tells us whether it is too light or too dark. In general, it is better to brighten underexposed areas of the image than to darken over-exposed areas. So you should avoid getting overexposed parts of the picture.
Turn on the "overexposure warning" on your camera. In the camera display you can already see the bright spots in the picture, shown as white flashing areas. Of course, the overexposure is also displayed on the histogram. As I said, you can save something from these areas with RAW files, but you don't have a chance with JPEG.
You should expose your image so that no areas are overexposed. Ideally, no area should be underexposed.

If you can't do this with your subject, the image has too much dynamic range for your camera. In this case, you should use gray gradient filters - primarily in landscape photography - and expose your image so that the white values of the histogram do not erode. The picture can be enhanced with Lightroom / Photoshop if the black values are cut off.

Another possibility would be to make several exposures and later assemble the image at the computer, to make a HDR image (High Dynamic Range) of it. As you know, opinions on HDR pictures differ widely. One likes this style, another doesn't like it at all. For me, HDR recordings are a means to an end to control difficult lighting conditions. I also use multiple exposures. One may not look at the finished photo, however, that it was necessary to take several photos. It must look natural to me.

The histogram is an indispensable tool for photography and, of course, for image processing. The sooner you get acquainted with it, the sooner you will be able to take better technical photos.