A Brief Primer
During many of my tutorials and discussions, I toss around the term histogram. It occurred to me when writing Part 4 of my series on The Three Cs of Image Editing that maybe a little explanation of the histogram is required.
Here is an example of a histogram. For the purposes of image editing, a histogram is a graphical representation of the pixel tonal value distribution in an image from dark to light as you travel from left to right. The above histogram is from Canon’s proprietary software. It shows a distribution of tonal values for each of the Red, Green and Blue channels as well as a gray histogram detailing the luminosity or brightness of an image.
Most digital cameras and most image editing software leverage histograms in some fashion. As a tool for evaluating and editing images, it is rapidly becoming indispensable. By looking at a histogram, you can tell instantly if you have over or underexposed an image. You can also tell if the overall image is going to be dark or light. In addition you can tell if the image is lacking in overall contrast or, conversely if it has too much dynamic range for the camera.
The following are several histograms based upon interpretations of this single image when modified in an editing program.
If you think of a histogram as black on the left and white on the right and there being 256 levels of gray in between (assuming you are shooting a JPG file with your camera), you can get a pretty good idea what your photo is going to look like and, to some degree how it will edit in your software.
Most digital cameras show histograms, or can be set to show histograms on the LCD screen. They are valuable tools to give you instant feedback on your exposure as well as your scene’s dynamic range and composition of values. If you check it from time to time, you may discover that you are getting too far from correct exposure or that your scene requires you to choose exposure for the brighter parts or darker parts of the image depending upon which subject is more important.
Software today relies on histograms to guide you as you make adjustments to the black point and white point of an image as well as the tonal values. In the fourth example histogram you will notice that the curve is less smooth. It has jagged peaks and pronounced stair steps from having its black and white point pulled left and right from the top histogram. This is an indicator of potential posterization. Here, you can see potential printing problems in the histogram though your eyes may not discern it on the monitor.
As you edit with your software, pay attention to your histogram.
- Avoid creating a wall in the white or black areas of the histogram.
- Avoid pulling your image too far to both sides and creating stair stepping.
- Remember to use the histogram as a guide and not as a proofing tool. The final proof is in the image itself. Some incredible images have ‘rule-breaking’ histograms.
- Finally, remember that a histogram isn’t good or bad. It simply reports on your image and therefore is just a tool to help you divine your creative vision
Rikk Flohr Copyright © 2007