Reading Histograms-- Part I

Article and Photography by Ron Bigelow

www.ronbigelow.com

Photoshop CS2 Used in this Tutorial

Background

Figure 1: Histogram

The Histogram is one of the most important tools in Photoshop (see Figure 1). Histograms help photographers understand the tonal or color structure of an image. For instance, a Histogram can show whether an image is composed primarily of dark tones, light tones, or a wide distribution of tones.

Histograms display tonal distributions (grayscale histograms) or color distributions (channel histograms). There are a total of 256 tones/colors in a Histogram. The darkest tone/color is assigned a value of zero. The lightest tone/color is assigned a value of 255. The horizontal axis displays these tones/colors from the darkest on the left to the lightest on the right. The vertical axis shows the number of pixels that have each tone/color. In Figure 1, for example, it can be seen that a large number of pixels are in the middle of the distribution. Therefore, the image from which this Histogram was created has a lot of midtones.

By viewing an image's Histogram, a photographer can understand the tonal/color distribution in an image and can analyze the tonal/color impacts of an edit.

Accessing Histograms

Figure 2: Histogram -- Compact View
Histograms can be accessed a couple of ways. Choosing Window/Histogram will bring up the Histogram palette. This Histogram can be viewed a number of different ways by using the Histogram palette menu (choose the pop-up menu in the upper right hand corner of the Histogram palette). Figure 2 shows the Compact View. Figure 3 shows the Expanded View. Figure 4 shows the All Channels View. One of the advantages of the Histogram palette is that it can be used at any time in Photoshop and with any of the tools.

 

Figure 3: Histogram -- Expanded View
Figure 4: Histogram -- All Channel view
Figure 5: Levels Histogram
A histogram is also integrated into Levels. Choosing Layer/New Adjustment Layer/Levels brings up Levels and the Levels Histogram (see Figure 5). The Levels Histogram can only be used with Levels.

Histograms and Images

People will sometimes try to evaluate a Histogram by looking at just the Histogram. For instance, they may look at a Histogram that has the tonal values shifted heavily to the left and make a statement such as, "This image is too dark" or "This image is underexposed". Such an evaluation would be invalid. Histograms can not be properly understood by themselves. In order to understand a histogram, it must be viewed in relation to the image it represents. In other words, both the image and the Histogram must be viewed at the same time.

Grayscale Histograms

Figure 6: Canyon Image

Grayscale Histograms are used to evaluate the tonal distribution of an image. The information shown in an image's grayscale histogram can be used to determine what tonal edits need to be made or to evaluate the results of tonal edits that have already been performed.

Figure 6 shows an image of a canyon and the surrounding, snow dusted mountains. This image has a full range of tones. There are bright whites in the snow, a small amount of blacks in the foreground rock, and a large amount of mid tones in the mountains.

 

Figure 7: Histogram of Canyon Image

This tonal distribution is reflected in the image's histogram (see Figure 7). The Histogram shows the whites to the right, a small amount of the blacks on the left, and the large amount of midtones in the middle. One might suspect that a small amount of clipping occurred in the brightest areas because the right side of the histogram appears to cut off a very small amount of the highlights. This is acceptable in this image. This image was taken under a heavy overcast sky. The flat light insured that there would be little if any detail in the brightest areas of the snow. In addition, the snow is so far away that any minor loss of detail in the brightest areas of the snow will not be visible.

In short, the dynamic range of the scene matched very well with the dynamic range of the camera. In addition, the exposure was correct. As a result, this histogram shows that the tonal distribution of the image is very good and no major, tonal edits are needed.

Figure 8: Tuffa Image
Figure 8 shows an image with an entirely different tonal distribution. This image has primarily light and dark tones with very few midtones.
Figure 9: Histogram of Tuffa Image

This tonal distribution is reflected in the image's histogram (see Figure 9). The Histogram shows the light tones to the right and the dark tones to the left. There are very few midtones. On the right side of the image, the histogram shows that there are no whites or extremely light tones. This is verified by looking at the image. The left side of the histogram shows a very small amount of clipping of the darkest tones. This is probably not a major problem with this image as the amount of clipping is so small.

As with the prior image, this histogram shows that no major, tonal edits are needed.

Figure 10: River Image
An image of a river and bridge are shown in Figure 10. This image should have a full range of tones. The small patch of sky is very light and there is a white sign on the bridge. There are dark tones in the shadows on both sides of the riverbank. However, the image seems to be rather flat.
Figure 11: Histogram of River Image
The histogram of the river image (see Figure 11) shows why the image appears flat. There are gaps on both sides of the histogram indicating that the image lacks both dark and light tones. Setting both the highlight and shadow values of the image will help increase the contrast of the image and make it appear much more vibrant.
Figure 12: Levels Used to Adjust the Tonality
Figure 12 shows a Levels adjustment that was used to add some life to the flat image.
Figures 13 -- 14 show the River image before and after the Levels adjustment was made. Figures 15 -- 16 show the associated histograms. As can be seen, making the tonal edits has improved the image. The Histogram in Figure 16 shows that the image has both blacks and lighter tones after the edit.
Figure 13: River Image before Tonal Adjustment
Figure 14: River Image after Tonal Adjustment
Figure 15: Histogram before Tonal Adjustment
Figure 16: Histogram after Tonal Adjustment
Figure 17: Rock Image
Figure 17 shows an image of a rock slab. This image does not have any pure blacks, and it totally lacks any highlights. Most of the tones are midtones.
Figure 18: Histogram of Rock Image

This is demonstrated in the rock image's Histogram (see Figure 18). The Histogram shows a lot of midtones and no very dark or light tones. This is as it should be. Thus, this image shows that the image has the proper tonality and no major tonal edits are required.

This image points out a very important point that was mentioned above: Histograms can not be properly understood by themselves. In order to understand a histogram, it must be viewed in relation to the image it represents. If a photographer were to have viewed the Histogram in Figure 18 without looking at the image, the photographer might be tempted to conclude that the image needed to be adjusted to increase the contrast and bring in both dark and light tones (as in the prior image). This would have been a mistake and would have created an image that had way too much contrast.

Figure 19: Rock and Plant Image
Figure 19 shows an overexposed image of a plant growing on a rock slab. The image appears washed out, and the colors are lifeless.
Figure 20: Histogram of Rock and Plant Image
The histogram of the overexposed image is shown in Figure 20. It can be seen that the histogram is shifted to the right. As a result, there are no dark tones and the midtones are too light. This image needs a tonal adjustment to compensate for the overexposure.
Figure 21: Curves Used to Adjust the Tonality
Figure 21 shows a Curves adjustment that was used to adjust the tonality of the image.
Figures 22 -- 23 show the rock and plant image before and after the Curves adjustment was made. Figures 24 -- 25 show the associated histograms. As can be seen, adjusting the tones has significantly improved the image. The Histogram in Figure 25 shows that the image, after the tonal edit, now has blacks and that the midtones have been darkened.
Figure 22: Overexposed Image before Tonal Adjustment
Figure 23: Overexposed Image after Tonal Adjustment
Figure 24: Histogram before Tonal Adjustment
Figure 25: Histogram after Tonal Adjustment
Figure 26: Statue Image
An image of a couple of small statues is shown in Figure 26. This image is composed of very light tones and a small area of mid grays.
Figure 27: Histogram of Statue Image
The Histogram in Figure 27 shows the preponderance of the light tones. Despite the similarity of this Histogram and the one for the overexposed image above (i.e., the histogram is biased toward the right), this image does not need any tonal editing as the histogram of this image shows the tones that would be expected based on the image content.

Articles

Histogram -- II