Your average Histogram really isn’t very glamorous. Not only is it unglamorous but you might think that thanks to the ability of the LCD preview on the back of the camera to show us what we have taken, there’s no real need to bother with it at all, writes Andrew James.
In the early days of shooting digital this was my attitude. Why look at a Histogram when there is an instant ‘polaroid’ on the back of the camera? Accuracy is the simple reason for reading the histogram.
What is the histogram?
In essence it is simply a graphical representation of the pixels contained within an image. To be truthful, it is actually representative of the JPEG preview so it still isn’t 100% accurate but because it is a graph it is more accurate than just looking at the LCD panel. If you just look at the LCD you can easily be fooled by a lot of different factors. For example:
- The brightness the LCD is set at
- Anything potentially reflecting off the surface of the panel
- How good or more likely, bad, your eyesight is
I’m also not saying DON’T use the LCD panel as a guide. I’m saying use it as a guide but in conjunction with the Histogram for a more accurate assessment of exposure. Because the histogram is a graphical representation of the exposure, it’s easier to make an assessment on the fine detail of under and overexposure. When you take a look at the Brightness Histogram on the back of the camera the graph is representative of 256 RGB (Red, Green, Blue) tones. So you start at 0 and that is Black. This then goes through the range of tones right up to 255, which is white.
When you look at the graph, you can roughly divide it into thirds on the horizontal axis. The first third is the Shadows, the second third the Midtones and the final third the Highlights. The vertical axis is simply representative of the number of pixels at each value. So if your ‘peak’ runs off at the top it really doesn’t matter.
When you are starting out it is sometimes easy to forget which side of the vertical axis is the shadows and which is the highlights. I always tell people to remember it as Black & White. Black comes first on the left and white last on the right. Simple! If we were reading the Histogram above we can see that it is neither under or overexposed, although it leans towards the midtones and shadows. We’ll come back to what we might do about this later.
With many modern cameras you also get a Histogram that will show the tonal range within the different Red, Green and Blue Channels. It works in the same way but separates the different colour channels. This next diagram is simply the RGB Histogram for the same image as the Brightness Histogram above.
Histograms and Live View
It’s also possible to see the Histogram when you are using Live View. When you switch into Live View press the info button again until the Histogram appears. This can be useful for a brief check but I wouldn’t suggest leaving it on the entire time as it partially obscures the image view. It’s a bit like trying to drive a car with a map across part of the windscreen.
Many of you may also use your Highlight Alert warning (blinkies), which flashes areas of the image on the LCD preview when the camera thinks you are in danger of overexposing parts of the scene. This is useful but don’t rely it on it over checking the Brightness Histogram as the Highlight Alert Warning will flash but when you look at the Histogram you can still be well within the graph. It’s called a ‘warning’ and that’s what it is doing, warning you that you are getting close to a problem area!
Also remember that the Histogram is just a tool and the exposure is your choice and you may be happy to overexpose or underexpose for creative reasons. Some overexposure and underexposure is also inevitable. A silhouette, for example, will mean you have an image with parts that go to pure black while a landscape with the sun it it may mean there are specular highlights – a bright spot of light that appears on shiny objects (like a mini-starburst). If these are small areas it’s fine but large areas of overexposure are best avoided.