Sensitivity and colour controls

Dark frame subtraction is used to suppress noise in long exposures. But it can tie up the camera for long periods of time. A 35-minute exposure to record star trails as streaks locked the camera for 70 minutes to allow noise reduction processing.

We’ve looked at the effects of aperture and shutter speed controls in Chapter 5. In this chapter we’ll consider sensitivity (which is determined by ISO settings) and white balance.

The ISO setting adjusts the sensor’s sensitivity to light, thereby determining the amount of light required for a correct exposure. At the same time, it influences the ratio of signal (image data) to electronic noise that is recorded in the image.

Back in the days of film, the sensitivity of the recording medium was set when the film was loaded; with digital, it can be changed on a shot-to-shot basis. Understanding ISO will help you to set your camera correctly for optimal exposures. And that will lead to better pictures.

When you’re hand-holding the camera in poorly-lit indoor environments, raising the ISO to 1600, as happened for this picture, can dramatically reduce the chance of camera shake blurring the photo.

Exposure Balancing

The ISO setting is used in the same way as the aperture and shutter speed settings. If you increase the ISO by one EV (exposure value), for example by moving from ISO 100 to ISO 200, the camera’s sensor needs only half the amount of light for the same exposure.

Suppose your camera set a shutter speed of 1/30 second at ISO 100 and you felt you couldn’t hand-hold the camera at such a slow shutter speed; increasing the ISO setting to 200 would give you a shutter speed of 1/60 second (assuming the aperture is unchanged).

In practice, you need to balance shooting convenience against potential for noise when adjusting ISO settings. Choose a low ISO whenever you can ““ including long exposures. But be prepared to increase the camera’s sensitivity when conditions demand it, taking into account the potential for noise in the resulting shot.

Adjusting ISO settings is wise when you’re shooting at low light levels and need fast shutter speeds to record action. In such a situation, boosting the ISO setting to 1600 would enable you to use a shutter speed of 1/500 second, which would be fast enough to stop action when photographing indoor sports.

DSLR cameras vary in the range of ISO adjustments they permit but all support a range of ISO settings between ISO 200 and ISO 6400. Many extend that range down to ISO 50 and up to ISO 25600 ““ or higher. Most DSLRs allow photographers to adjust ISO settings in 1/2 or 1/3 step increments, with settings usually provided in the Custom Function menu.

Almost all DSLRs also include an Auto ISO setting, which allows the camera to determine the optimum ISO setting for the subject. Most allow photographers to set an upper limit to the Auto ISO range, either in the main camera menu or as a Custom Function.

Most DSLRs enable users to set a maximum limit beyond which sensitivity will not be increased. (Source: Canon.)

Limiting ISO is useful in shutter-priority AE mode for preventing the sensitivity from being pushed to a level that may result in image noise (see above), which increases when the signal from the sensor is boosted.

Suppose you’re shooting a basketball match and using shutter speeds of, say, 1/250 second, it’s usually wise to set the camera to limit the sensitivity to, say, ISO 3200 to keep noise below noticeable levels. When the light level in the scene is bright, the ISO value is lowered automatically.

When the camera’s exposure meter detects low light levels, the camera will automatically kick up the ISO to maintain that shutter speed. But it won’t go above ISO 3200, regardless of ambient light levels.

Image Noise

Photographers typically recognise three types of image noise: fixed pattern noise, random noise and banding noise. Fixed pattern noise is associated with long exposures and high temperatures. It usually appears as ‘hot pixels’ which show up as white or coloured blocks.

Random noise is also granular, often appearing as a ‘salt-and-pepper’ pattern with dark pixels in bright regions and bright pixels in dark regions. It usually occurs with high ISO settings. Banding noise appears as stripes and is introduced by the camera when it ‘reads’ data from the image sensor. (Banding noise is rare with modern DSLRs.)

Fixed pattern noise usually appears as scattered white and/or coloured pixels. It’s removed with long-exposure noise reduction by causing the camera to take a second exposure that records only the pattern of dots and then mathematically subtracting this pattern from the image data. This doubles the exposure time.

Random noise usually appears as granularity with dots of different colours. It’s not easy to remove without reducing image sharpness.

Noise not only changes with exposure settings and camera models, it can also vary within an image. Shadowed regions usually contain more noise than highlights.

Cameras with large sensors that have big pixels can collect more light and are, therefore, less susceptible to image noise than small-sensor digicams with high pixel densities. This is the main reason DSLRs deliver better performance at high ISOs than compact cameras.

Most sensors are designed to work best between ISO100 and ISO 200. However, when you use a modern DSLR camera, noise is seldom visible unless you shoot with ISO settings above 6400 ““ and even then, it may only be seen when the image is enlarged substantially.

Noise Reduction

Because all pixels collect some noise, all cameras perform some level of noise-reduction processing on the JPEG files they produce. Sometimes it’s applied automatically; at other times at the user’s discretion.

More sophisticated cameras provide separate processing settings for long exposures and high sensitivities. Newer cameras use newer technology and can deliver images with less noise at similar ISOs than earlier cameras could achieve.

Long-exposure noise reduction usually involves a technique known as ‘dark frame extraction’, which is applied automatically in many cameras for exposures longer than one second. The camera records the scene and then takes a second exposure of the same length with the shutter closed to capture the noise pattern produced by the sensor. The noise pattern is mathematically subtracted from the image, leaving it virtually noise-free.

Modern DSLRs include powerful noise-reduction controls, enabling users to shoot at ISO 12800 (used for this photo) and produce virtually noise-free images.

While dark frame subtraction effectively removes most of the noise created during long exposures, high ISO noise-reduction processing tends to work by blurring the edges of noise patterns and this can reduce the sharpness of images. Photographers are faced with choosing whether sacrificing some real detail is acceptable if it allows more noise to be removed.

Processing technologies work on two main types of noise. Luminance noise consists mostly of variations in brightness patterns. Chrominance (or colour) noise appears as coloured

speckles. The aim is to identify the type of noise and apply the right amount of processing to minimise it.

We tend to find colour noise more objectionable than luminance noise so most cameras apply more noise reduction to it. When images are edited, most software allows the user to control colour and luminance noise reduction separately.

White Balance

The white balance setting is used to make the colours in a digital photograph look natural under different lighting conditions. It works by adjusting the colour balance so that objects which appear white in reality are rendered white in the image. To do this, the camera has to evaluate the ‘colour temperature’ of the light illuminating the subject and correct for its relative warmth or coolness of white light.

Human eyes are very good at judging what is white under different light sources and our brains apply automatic correction to produce a colour-correct impression. Digital cameras have to rely on auto white balance (AWB) adjustments, which can often fail “” and may even introduce unwanted blue, orange or green colour casts.

AWB adjustment balances the colour data from the camera’s red, green and blue (RGB) channels. These primary colours are found in all light sources in varying proportions, depending on the colour temperature of the light source. With a high colour temperature, the light has more blue; with a low colour temperature it’s redder.

Different types of lighting have different colour temperatures, which allows camera manufacturers to create a range of pre-set colour balances that should, in theory, counteract the colour cast and restore cast-free colours. The illustrations on this page show the effects of the colour corrections that typical presets apply for different types of lighting.

In most DSLRs, photographers can set white balance relative to the Kelvin colour temperature scale. This can save time when shooting with standardised lighting as you can simply dial in a correction to match the lights you use. The table on the following page shows the Kelvin temperatures for some frequently-used lighting types.

Typical colour corrections applied by white balance presets. Top row from left: Auto white balance should reproduce natural colours with no obvious bias; Daylight correction adds a touch of orange to counteract the blue from cloudless skies; Shade correction intensifies the orange correction. Middle row from left: Flash correction adds a hint of yellow to remove a slight blue bias in electronic flash light; Daylight Fluorescent lamps require a fairly strong orange-red correction to counteract strong blues in their emission; Cool White Fluorescent lighting needs a slight magenta bias to remove greenish casts. Bottom row from left: Warm White Fluorescent light is biased towards red and orange and needs a blue filter to correct it; Incandescent lighting is very orange and is corrected by a strong blue filter; Cloudy skies need a hint of yellow to warm a slight blue cast.

Most types of natural lighting (and some artificial light sources) involve a range of Kelvin values, which means the actual colour may fall anywhere within the specified range. Some types of artificial lighting, notably fluorescent lighting, are ‘spiky’, with stronger emissions at specific colours.

The Kelvin values for fluorescent lighting vary according to the different types of fluorescent tubes on sale: daylight, warm white and cool white being the most common. This can make correction of colour casts tricky.

The colour bias of a clear blue sky or open shade can be influenced by the latitude where the photograph is taken. Kelvin values are usually higher in higher latitudes and some degree of fine-tuning may be required.

To overcome problems associated with varying colour balances, all DSLRs provide a manual or ‘custom’ setting, which lets you measure the colour of the illuminating light and use the result to remove unwanted colour casts. The process is straightforward. Simply cover the subject with a plain white object (sheet of paper or white card) and set the lens focus to manual before taking an exposure or white balance reading to capture the colour of the illuminating light. (Some cameras can record the light without taking the shot.)

Success depends on the initial exposure, which captures the colour data. If it’s not correct, subsequent shots may be off-colour. Some cameras alert photographers when this happens ““ but most don’t so you should always check your shots when using this strategy.

Shooting raw files (see Chapter 9) is the best way to deal with white balance in difficult lighting because you can adjust white balance AFTER shots are taken. Raw files also provide a broader range of colour temperature adjustments, which may been needed to cope with subjects containing a mixture of different lighting types (fluorescent, halogen and incandescent, for example).

Adjusting white balance in a raw file is quick and easy. You can either tweak the temperature and green-magenta sliders until colour casts are removed, or simply click on a neutral colour (white or grey) within the image. This reference can be used to quickly correct other shots taken under the same lighting.

Image Tone Adjustments

Most DSLRs include special image tone adjustments (or ‘picture styles’) for imparting particular ‘looks’ to photos. As a rule, these controls can only be used in the P, A, S and M shooting modes and only with JPEG files.

It’s common to have a Standard default setting plus presets for Vivid, Neutral, Portrait, Landscape and Monochrome. Most cameras allow users to adjust the sharpness, contrast, colour saturation and colour ‘tone’ (or hue) parameters within each setting. Sharpness adjustments work mainly on edges and allow photographers to sharpen or ‘soften’ their shots. Contrast and saturation (colour vividness) adjustments are similar, with the minus settings reducing and the plus settings increasing the selected parameter.

The Standard, Neutral and Faithful image tone adjustments all produce natural looking colours.

The Monochrome setting converts the colour image into black and white.

On-screen adjustments of colours along Blue/Amber and Magenta/Green axes are available in all DSLR cameras for fine-tuning colour balance settings.

Examples of some of the digital filter effects provided in many modern DSLR cameras. Top row: the unedited image, Toy Camera, Grainy B&W; bottom row: Reversal Film, Miniature, Bleach Bypass.

Picture style adjustments can often be applied post-capture when converting raw files into editable formats. Some cameras also allow them to be applied in-camera (to JPEGs) and a few allow users to save adjusted settings as new picture styles for future use.

Digital Filter Effects

Digital filter effects have become popular in recent times, particularly in entry-level cameras. They’re usually accessed through the main menu and typically include Toy Camera, Retro, Miniature, Reversal Film, Monochrome and Bleach Bypass settings. These effects can only be applied to JPEGs and some are quite extreme. Different manufacturers provide different levels of adjustment for these effects.

The main problem with all of these adjustments is that the adjusted settings are locked into the image file. The adjustment range is also somewhat limited, compared with the adjustments available in image editing software ““ including raw file converters. We suggest that you use them sparingly.

Make sure you keep an unadjusted copy of the shot by shooting RAW+JPEG pairs. Then, if you don’t like effect, it’s easy to go back to the unadjusted image and start again.

USEFUL URLs

www.luminous-landscape.com/ essays/art2.shtml for a discussion of the factors that contribute to ‘good’ photos. ø  http://www.nikonusa.com/en/ Learn-And-Explore/Article/g9mqnyb1/ Understanding-ISO.html for a Nikon tutorial on ISO sensitivity.

www.cambridgeincolour.com/ tutorials/natural-light-photography.htm for a tutorial on white balance.

 

This article is an excerpt from  Digital SLR Pocket Guide 3rd Edition.