Articles
Guide: Exposure in Digital Cameras
Exposure, speed, aperture, ISO ... it's all rather complicated, isn't it? The fact that today's cameras take care of almost everything so well make some of these terms even more of a mystery. One setting remains in your control though: exposure compensation, which is easy to understand--and to use!
Franck Mée
Published: September 01, 2009
Histogram
The histogram is a very useful tool, and sometimes more accurate than the screen on which it's displayed (which can often be hard to see in bright sunlight, just when you need it)--once you know how to use, that is. It allows you to check the exposure level of an image at a glance. For every light level, from the darkest to the brightest possible value, it shows how many pixels there are at that particular brightness level.

Let's look at these three histograms for the gradients from black to white. In the middle, with a normal exposure, the black square isn't quite black and the white square is the only one that's truly white. Along the histogram, each individual square produces a peak of pixels at that brightness. Above, where there is an under-exposure of -2 EV, there are lots more black pixels than any other colour. Below, though, where the image is over-exposed to +2 EV, the white pixels dominate so much that the greys in the middle don't even get a look in.

In general, the histogram for a real photo should be central, and roughly symmetrical. If it's too far to the left, the image will be too dark; too far to the right, and it will be too bright. Some photographs, though, will naturally lead to a histogram that leans one way or the other. A nighttime landscape will obviously have a lot of black, while a David Hamilton portrait will be full of light.
The Basics

Adjusting exposure ensures that a photo is neither too bright nor too dark.

The majority of digital cameras set exposure automatically, choosing the combination of settings to ensure this balance themselves.  These tools aren't always perfect, so there is usually an exposure correction feature that allows the user to choose between 'lighter' and 'darker' photos.

More advanced cameras include settings that allow you to control every aspect of the exposure, but we'll come back to look at these different settings another time.

It was a dark and stormy night ...

To take a photo, the sensor in a digital camera relies on millions of individual detectors that measure the amount of light that hits each part of the sensor.  We can compare each one of these pixels to a well gradually filling up with rainwater: if there's not enough water, it won't reach the surface; if there's too much, it will overflow.  In our metaphor, the former would produce a black pixel, while the latter would produce a white pixel.  The aim is to have enough water in all of the wells to be able to accurately measure their contents, minimizing the number that run dry or overflow.

To do this, we can control three different factors: the amount of light coming in through the lens ('aperture', measured as e.g. 'f/4'), the length of time the lens is open ('speed', measured as e.g. '1/100 s') or the sensitivity of the sensor (measured as, e.g. '200 ISO'). 

The aperture corresponds to the strength of the rain falling over the wells: if there's a big storm, then it will be easier to fill up the wells than if it's just drizzling.  The speed represents how long the storm goes on for: as long as it rains with the same intensity, a weeks' worth of rain will fill the wells more than a days' worth.  Finally, the sensitivity corresponds to how wide the well is: narrow, thin wells will quicker than fat ones.

There are, therefore, several different ways of getting the right amount of water in every well.  An intense storm that lasts just a few hours might fill a small well to the same level as a bigger well subjected to constant light rain for a week.  And so a photo with a speed of 1/1000s at f/2 and 800 1SO has the same exposure as one at 1/15s at f/8 and 200 ISO.

In practice

Look at the three different test cards below, and in particular the gradient that runs from white through grey to black at the bottom of each one.  In the middle one, each individual square is clearly defined and separate from the others: there aren't any wells that have overfilled, nor are there any running try.  This is the ideal result.


In the top photo, though, the last five boxes form a single area of black.  The exposure is too low, and not enough rain has fallen in those wells to be able to measure how much water is in them.  In the bottom photo, there is the opposite problem, and the seven rectangles to the left all look white: these wells were too full, and the water has flooded out so that we can't tell how much fell where.

We say the first image is underexposed, because we haven't allowed enough light to fall on the sensor, while the second is 'overexposed', because there's too much light. 

Of course, it's possibly to have both under- and over-exposed sections in a single photo.  It's often the case when you shoot in the shadow of a building outside on a sunny day.  In photos like this, we approach the limits of what a sensor can achieve, and there's no miracle solution to guarantee a well-lit sky and plenty of details in the shadows.

Exposure Compensation

In general, you don't want to have to choose the right combination of settings yourself: the aim is to take a good photo, with has details everywhere and is neither too light nor too dark.

Most of the time, your camera's automatic mode can sort this out for you, but sometimes things go wrong.  To get around this problem, cameras include an Exposure Correction function, that can set different EV (exposure value settings).  You can spot this feature by looking for a control with a square that contains a black + on a white background and a white - on a black background.  On the left, you can see it on a Canon compact and on the right on an SLR from Pentax.

Note that especially on compact cameras, this setting isn't always available in automatic mode.  You might need to switch to 'P' mode, where there are more settings available to the user.

The majority of manufacturers use the same interface: pressing the button above brings up a scale with a cursor that the user can move.  Moving towards the - side produces darker images, while moving + produces lighter images.  The three test card photos above were produced this way: the same camera was set up identically, but the EV was set to -2, 0 and +2 respectively.  Compact cameras will update their displays in real time to show you how the photograph will look with the current settings.  Using the viewfinder on an SLR, you'll have to rely on the scale--and your own experience of how much you need to adjust the exposure by.


Exposure compensation isn't a setting that is used very often, because today's compacts have fairly competent automatic settings.  Nevertheless, this easy setting is worth knowing about.  It can help you prevent the sky from looking like a single block of white or help you bring out details that would otherwise be hidden in the shadows.

Playing with exposure levels can also be an artistic choice: overexposed photos often have a dream-like quality, while underexposed scenes can seem dark and forboding.

In general, exposure is a fundamental concept in photography, and one that we often return to because the three settings that contribute to it have a big role in other areas.

Our RSS News Feeds : 

Add to Netvibes