Wondering how to read the histogram on the back of your camera? Not quite sure what highlights and mid tones are? Let’s take a walk through this basic tool used throughout the image capture and processing process.
While there are several ways to represent the color in an image, by far the most used form is to record each pixel or point as a combination of three values; a number for the red component, one for the green and one for the blue. The color of a pixel is a mixture of these three components and the intensity or brightness the scale of the values.
You often hear about “bit depth” in image files and camera sensors. This indicates the range of numbers for each color component. Eight-bit depth allows numbers for each component that are 0-255 in range. Twelve-bit depth allows numbers for each component in the range of 0 – 4095. The actual number of possible different colors when you combine the values from the three components is fairly large. For an image with an 8-bit depth the total possible combinations works out to about 16 Million colors; for the 12-bit depth, you get more than 68 Billion possible color combinations.
Getting the most from your camera means properly exposing the image with your sensor – but how do you know you are doing so? Proper exposure means not presenting too much light which will overwhelm the sensor or too little light which will not provide enough energy for proper recording. Providing this insight is where the histogram becomes useful.
Given the bit depth of the sensor, each pixel can take on only certain values. A histogram is a tool that shows you how many pixels in the image are recorded at what value. For historical reasons histograms are expressed in an 8-bit range – numbers 0 to 255. Each point or bin in the histogram represents how many pixels in the image have a particular value. The #10 bin represents how many pixel values had the value 10, the #20 bin is similarly the number of pixels that have the value 20.
For sensors or files with more than an 8-bit depth, the histogram represents a scaled version from the smallest to the largest value. Histograms are often done multiple ways – you can have an individual histogram for each color component or a single histogram representing the combined energy in the pixel; these are called luminance histograms.
So how do you read a histogram?
A well-recorded image is one where you have numeric values spread across the range of possible values. Below is an example of an under-exposed, over-exposed and properly exposed image. In the over-exposed image the histogram appears to be chopped off at the top. This tells you that the image is too bright and that the sensor had to limit or clip the image causing loss of information. Similarly, an underexposed information means the scene was too dark for you to record values and again you have lost information.
Highlights, Mid-Tones and Darks.
In discussing image processing terms like “lights”, “darks” or “mod tones” are often used. These are simple references to the ranges of light in the image and correspond to different parts of the histogram.