On The Level: Make Your Photos Pop with Histograms and Levels

Mon, 12 Nov 2007 08:30:00 -0000

If your lightbox is letting you down and you can’t find a sunny window to save your life, you might find yourself in dire need of some digital post-processing to help out your photos. Fortunately, the “Levels” tool found in most photo editing programs can whip virtually any photo into shape with just a few clicks of your mouse.

Using the Basic Functions of the Levels Tool:

In Photoshop, Levels is found under Image>Adjustments. In the Levels function, you will see three arrows below the histogram of your images. The arrow on the far left controls the darkest third of the image’s pixels, or “shadows”, the middle arrow controls the grey tones, or “midtones”, and the arrow on the far right controls the brightest third, or the “highlights”.

1. To adjust your histogram, first evaluate the distribution of the pixels to determine whether your image is currently low-key, high-key, low-contrast or high-contrast, versus what you want it to be (see description below).

2. Simply click on the arrows and slide them back and forth to adjust the tonal range. If you have a low-key or high-key image, use the opposite arrow to adjust the image i.e. if your image is high-key, move the shadows arrow to the right, if your image is low-key, move the highlights arrow to the left. Use the midtones arrow and move it left or right accordingly until your photo has improved to your satisfaction.

3. If your image is low-contrast, slide both the shadow and highlight arrows in towards the middle to create more contrast. High-contrast images are not so easy to repair. The distribution of pixels in a high-contrast image means that there are fewer grey tones that can be manipulated. A high-contrast image can be darkened or lightened overall, but not evenly redistributed. A very high-contrast photo may need to be re-taken.

4. Click ‘OK’ to save your changes. Then go back into Levels and have another look at your new histogram. The pixels should be more evenly redistributed in a centered, bell-shaped curve. You can continue to adjust the image using the arrows until your desired effect is achieved.

Tip: When an image’s pixels are redistributed over a wider tonal range, some tonal values wind up empty. This is called “combing”, and appears as a finely striped histogram, with the blank or empty vertical stripes representing the tonal value that is missing. Repeated use of Levels can lead to heavy combing and appear as banding or “posterization” in the photograph. Always use Levels judiciously and try to make all adjustments in one turn to avoid combing.

What are Levels and Histograms and How can They Help?

The “Levels” function is a tool used by digital editing software to manipulate the tonal values or brightness levels of an image. The standard 8-bit digital image contains 256 colours which are mapped as discrete black, grey, or white tones on a histogram, with 0 = black, 255 = white, and 254 shades of grey in between. When you take a photograph, the pixels of an image are sorted into one of these 256 tonal values, and stacked to make a vertical bar. When the bars are lined up together in numerical order of tonal value on a graph with 0 (black) on the far left, and 255 (white) on the far right, they create a curve on the graph. This curve (or curves in some cases), is the image’s histogram.

Some digital cameras have a function that allows you to view an image’s histogram directly on the camera screen. Even if your camera doesn’t have this function, a program such as Adobe Photoshop will be able to produce your image’s histogram in the Levels function. By examining the shape and placement of the histogram’s curve, you can determine whether your image is properly exposed or not, and manipulate the tonal values to achieve a better image.

A perfectly exposed, evenly toned image will display as a bell-shape centered in the middle of the histogram, with the bulk of the image’s pixels (top of the curve) lying in the midtones or grey tonal value range. The ends of the curve will meet precisely at 0 and 255, because a properly exposed, evenly weighted image will have fewer perfect black or perfect white pixels than mid-tones. However, very few photos will display a perfect curve. See here:

Depending on where the peak or peaks of the histogram fall, you may have a high-key, low-key, low-contrast, or high-contrast image:

High-Key: In a high-key image, the bulk of the histogram’s curve will fall on the right side of the graph, indicating the majority of the tonal values lean towards 255, or perfect white. This may indicate an over-exposed photograph that appears pale or having glare. See here:

Low-Key: In a low-key image, the bulk of the histogram’s curve will fall on the left side of the graph, indicating the majority of the tonal values lean towards 0, or perfect black. This may indicate an under-exposed photograph that appears dark or muddied. See here:

Low-Contrast: In a low-contrast image, the histogram’s curve will be tight and narrow and bunched in towards the centre of the graph. The ends of the curve will not meet 0 and 255. This indicates a photograph with not enough contrast that may appear washed out, flat or dull.

High-Contrast: In a very high-contrast image, the histogram’s curve may be inverted, with a peak at either end of the graph that dips in the centre. This indicates the majority of the pixels in the image are either very dark or very bright and results in a photograph that is very high-contrast with few midtones. See here:

The types of histograms described above are not necessarily “bad” histograms. There times when it is desirable to have an unbalanced tonal range — for example, a scene of a snowy day would realistically be a high-key image.

Whereas a night scene would automatically create a low-key value range.

A soft and romantic image might need to be low-contrast, while high-contrast would enhance a bold and exciting image. The “Levels” function comes into play when you do need to adjust the distribution of tonal values across your image.

Search results (tags) for: "levels"

On The Level: Make Your Photos Pop with Histograms and Levels