Reading a Histogram


Today we’re going to be talking about histograms, how to read them, and why it’s important.

In photography, histograms show the distribution of the intensity of light in an image. This is important because the screens on the back of cameras are not great.  I’m sure many of you have been out taking pictures, looked at the screen on the back of your camera and thought this looks pretty good.  Unfortunately when you get home you realize the exposure doesn’t look as good as you thought.  This is where the histogram comes in.  Instead of trying to check the image for the proper exposure you can look at the histogram to see if any parts of your image are over or under exposed.  On Nikon’s you can usually view histograms in the camera by pressing the up or down arrows while viewing a picture and on Canon’s by pressing the Display or Info buttons, but check your manual for specific instructions.

Reading a histogram

When light is captured on an image sensor, its intensity is measured and placed on a scale of 0 to 255 for an 8 bit image(jpeg).  This range is called the bit depth and will vary depending on the type of sensor in your camera.  A 12 bit sensor is fairly standard, but jpegs are compressed to 8 bits.  0 is on the left of the chart and represents black, while 255 is on the right of the chart and represents white.  The higher the bars the more pixels in the image have that level of intensity.

In the image below, you can see the histogram has lower values on the left and higher values on the right.  this is because the picture has more white areas than black areas.  However since the bars are do not hit the left or right edge of the chart, we can tell that there are no over or under exposed parts of the image.

2014-12-05 19.23.43

The image below shows the same picture but we have the red, green, and blue histograms broken out separately along with the overall histogram.  By looking at the full color histograms you can see if any one color is over or  underexposed even when the white shows the image working correctly.  Here you can see that the red spectrum is a little blown out. This because the table the cards are on has some very bright red elements which make up much of the image.

2014-12-05 19.23.56

The histogram below is the luminosity histogram from one of my pictures of River from Photoshop.  As you can see there is a very large spike towards the right side of the image.  This is not a bad thing, it’s because a lot of the image is white and the numbers pixels in snow-white tones go off the chart.  This would only be bad if the tall bar was all the way to the right (over exposed) or all the way to the left (under exposed).

Capture

Summary

As you can see, histograms can tell you a lot about your exposure, and they can be invaluable when you’re out shooting, especially on a sunny day.  For anyone who shoots primarily in jpeg or does not do much post processing, using histograms can help make sure you get the exposure correct in the camera.

Peter

Advertisements

5 thoughts on “Reading a Histogram

  1. I never use the rgb histogram. For what reason would I use the color one over the highlight? Or do you suggest using both? In your example re rgb histogram, would you do anything to correct the redness if it’s overexposed? Or are you pointing that it just as an aid to reading the rgb histogram?

    Great posts once again! Btw I love your shots of River! I have a beautiful but very lazy weimaraner….I could never get a shot of him running around like that 🙂 His coloring and expressions make him a gorgeous subject to photograph tho!

    Like

    • Most of the time the the color histogram is not necessary, but there are some situations where it can prove useful. These situations would be when the color in an image is important to the overall image, think red flower with a light background. With this kind of image it’s possible for the luminosity histogram to show no over or under exposed parts of the image, but you could be over or under exposing the red channel which would make your flower lose detail. For the picture in the post I wouldn’t do anything with the red channel information since it was not a major part of the image.

      Thanks for the thoughts on River, she’s a fun but rather ridiculous dog.

      Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s