Histograms 101

Got a digital camera?  Do you use Adobe Photoshop? (or Photoshop Elements, or Nikon Capture NX, or Lightroom, or Aperture, or just about any other digital image editing app?) Then you’ve probably seen a histogram, whether you know it or not.  If you’ve read any blogs or forums about digital photography, or listened to any podcasts about digital photography, you’ve probably heard that you should “check your histogram” or “keep an eye on the histogram” to get an idea of how good your exposure was.  If you’re like me, right about now you’re probably wondering just what the heck a histogram is and what you should be looking for when you “keep an eye on it,” but you don’t want to ask.  Not to worry; I’m here to help!

So, what’s a histogram?  Here’s what Dictionary.com has to say about it:

“–noun Statistics.

a graph of a frequency distribution in which rectangles with bases on the horizontal axis are given widths equal to the class intervals and heights equal to the corresponding frequencies.”

 Oh, that’s all it is?  Right.  Got it.  Perfectly clear now.  

Not. 

Let’s see if I can untangle that a little bit for you.  As it pertains to a photograph, a histogram is a depiction of the relative brightness levels of the pixels in the image.  That probably didn’t help, did it? 

OK, let’s try an example.  Here’s a simple black to white gradient:

And here’s the histogram that goes with it:

As you can see, the histogram is almost completely flat, and extends all the way from left to right.  That’s because there are an equal number of pixels at each brightness level in the image it represents.  Now, look at this gradient, in which I’ve shifted the midpoint to the right, so it has more dark pixels than light:

Here’s the histogram for this gradient: 

You can see that it “piles up” at the left side, at the lower brightness levels, reflecting the fact that there are more dark pixels in this gradient.  Here’s a gradient with a left-shifted midpoint, which increases the number of light pixels:

 And here’s the corresponding histogram:

Now the graph “piles up” on the right side since there are more light pixels than dark. 

Here’s a picture I recently took in Japan of some trees after a fresh snow lit by the rising sun: 

 And the histogram for this shot:

 

Based on what we now know about histograms, what can we tell about this image?  Well, we can see that there are quite a few dark pixels (evidenced by the peak at the left end of the graph), not too many in the medium-dark range, and quite a few in the medium-light area.  Also, the fact that the peak on the left-hand side piles up against the edge of the graph indicates that there are areas of the photo in the shadows that have completely lost detail (they’ve gone entirely black, commonly called "blocked-up shadows").  On the other hand, since there’s no peak on the right end of the graph there should be details visible in all the highlight areas.  Since there’s some “room” at the right end of the graph, it shows that I could have increased my exposure by about half a stop or so without losing much detail in the highlights. 

That’s enough for now; in the future, I’ll get into some more details about how you can use a histogram to make adjustments while you’re out shooting and increase your chances of getting that “perfect shot!”

 

See more of my photography at http://lokahiphoto.com.

Looking for more great photography related info? Check out the This Week in Photography blog and podcast.