All currently editable chapters are in the Working Version section. -- All finished versions are in the Finished Version section.
Please create an account and log in to start editing

From Digital Foundations

Contents 1 Download Materials For Chapter 8 2 8. Tonal Range 3 Visual References 4 Results of Chapter 8 Exercises 4.1 Exercise 1: Minor adjustments to the original file 4.2 Exercise 2: Understanding the histogram 4.3 Exercise 3: Adjusting the image with Levels 4.4 Exercise 4: Adjusting the image with Curves 4.5 Exercise 5: Targeting saturation levels 4.6 Exercise 6: Sharpening the image

[edit] Download Materials For Chapter 8

You can use your own digital image in RGB color mode for the exercises in Chapter 8, or you can download our file here.

You can download our final file here.

[edit] 8. Tonal Range

The tonal range is the change in value from black to white. It is the set of grayscale information in an image. Sometimes tonal range must adjusted so images have a full range of values in the shadows, midtones, and highlights. Adjusting the tonal range addresses these common problems:

Watch Out: All monitors are different. If you consistently see a color cast in all your images and in an area you know to be a neutral grey (such as the grey surround in Photoshop’s Maximized Screen mode), use the buttons on your monitor to calibrate it until the monitor can display a neutral grey.

1. The image is too light or too dark. There may be a lot of detail in the light areas or in the shadows that can be made visible or printable through an adjustment.

2. Contrast is too low or too high. A low-contrast image has a flat tonal range. A high-contrast image has very light highlights and very dark shadows, and very little detail in the midtones.

3. The image displays a color cast — evidence of a hue in areas that should be neutral gray or white.

[edit] Visual References

http://commons.wikimedia.org/wiki/File:Niepce_1826.jpg
View from the Window at Le Gras, Nicéphore Niépce, 1826, Saint-Loup-de-Varennes, France. Captured on 20 × 25 cm oil-treated bitumen.

In this first recorded photograph, the exposure time was eight hours! Notice the limited tonal scale and high contrast between the dark and light values.

http://en.wikipedia.org/wiki/Image:Lange-MigrantMother02.jpg
Migrant Mother, Dorothea Lange, 1936. Silver gelatin print.

This photograph was commissioned by the Farm Security Administration (FSA). Florence Owens Thompson looks towards the future with worry, as her children bury their heads into her shoulders. Notice how the range of tonal values expresses the details in Florence’s face and on the blanket on her lap. The FSA (1935 – 1944) was part of the New Deal, a set of programs initiated by Franklin Delano Roosevelt to stimulate and revitalize weak economies. Starting in 1935, the FSA hired photographers, such as Lange, Walker Evans, and Marion Post Wolcott to document America during the Great Depression.

[edit] Results of Chapter 8 Exercises

On the left is the original image, on the right is the final image made in Chapter 8 Exercises. Notice minor differences in the contrast and value of the image tones. You should aim for minor image adjustments, not large changes in the tonal range or color intensity.

[edit] Exercise 1: Minor adjustments to the original file

1. For this exercise, open any image from your digital camera or scanner in Photoshop. You may also use the file on the wiki called rgb-trees.psd.

2. If the image needs to be rotated or cropped, make this adjustment now. To rotate the entire document 90 degrees (clockwise or counter-clockwise) or 180 degrees choose Image > Rotate Canvas > and select the amount of rotation from the submenu.

Tip: You can rotate the Crop tool’s marquee area by placing the mouse just outside one of the corners. It will change into a curved arrow to indicate rotation. Once you press return, the Crop tool will rotate and crop the image.

3. To crop the image, use the Crop tool to click and drag around the area of the photograph that you intend to keep. The Crop tool displays the cropped area by setting a dark gray cast on the parts of the image that will be cropped out of the scene. Use the anchors along the side edges of the crop box to adjust where the crop box is located in relationship to the image. When the crop area looks appropriate, press Return on the keypad or click on the checkmark in the Options bar to commit to the crop operation. If it seems like the Crop tool is sticking to the edges of your image, click and hold on an anchor then hold down the Control key, which temporarily turns off “snapping” (a feature in the View menu that can be helpful when moving multiple objects into position). Let go of the mouse before letting go of the Control key.

Tip: If you clicked with the Crop tool and simply want to start over, the Escape key on the keypad will return you to normal editing mode.

4. At this point, it is not a bad idea to check the Image Size dialog box (Image > Image Size) to evaluate the resolution settings. Ask yourself, at what size do I plan to print this image? If your image comes from a digital camera, there is a good chance the resolution is 72 DPI and the file’s pixel dimensions are very large. To change this in the Image Size dialog box, be sure to uncheck Resample Image before adjusting the resolution to a higher number, such as 300 DPI. You should see that as the value of the resolution (measured in dots per inch) increases, the width and height of the file decreases and the amount of pixel information (in the top boxes, which should be grayed-out) remains the same. If all of this is not happening for you, and you are trying to increase the resolution of the file, something is not right!

Notice the button for Resample Image is not checked in the Image Size dialog box on the left. In the Image Size dialog box on the right, where the resolution value was changed to 200 DPI, the width and height that the image will be when it is printed is reduced to 4 by 6 inches and the number of pixels (in the top part of the box) remains the same.

[edit] Exercise 2: Understanding the histogram

Now we will take a look at the tonal range within the image. This can be done in any color mode, but to keep this process easy for the first time, we’ll change the image to grayscale.

1. Click on Image > Mode > Grayscale to convert the image from RGB color mode to grayscale.

2. Click OK through the Discard Color Information dialog box.

3. Use File > Save as and add “_gray” to the end of the file name. Be sure to save it as a Photoshop native, or .psd file type.

4. Click on the Histogram tab within the Navigator and Info panel to the right side of the screen.

Note: Color images are rendered digitally by compositing separate color channels (red, green, and blue, for example), each with corresponding grayscale values. So, the Histogram displays information about grayscale values, even if they correspond to color information.

5. Observe the way the Histogram panel conveys information about the grayscale tones in the document.

There are 255 levels of gray in any 8-bit image. Consumer scanners and digital cameras capture 8-bit images. The overall graph displays the amount information within the image (y-axis) at the various levels of gray from black (on the left side of the x-axis) to white (the right side of the x-axis).

Note: There are professional scanners and cameras that capture 16-bit images, yielding more options for adjusting the tonal range; but 8-bit images are more common and easier to work with, so we will remain focused on them.

Look at the histogram to make the following observations: A. Does the histogram start and end at the beginning (dark values) and end (light values) of the x-axis? This would mean that there actually exists image information in the darkest shadow areas and the lightest highlight areas. If the graph ends before the edge of the histogram’s frame, it means there is no information at that end of the spectrum. There is probably a noticeable lack of detail in the image where the graph is clipped.

The histogram for this image is clipped on the shadow side.

B. On the x-axis of the graph, where is most of the image information stored? In other words, where are the spikes in the graph? This should give you a sense of how dark or light overall the image appears.

Imagine in the histogram above that the midway point is where 50% gray occurs in the image. In this image, the highest spike appears somewhere between the blackest shadow and 50% gray.

Watch Out: The triangle-in-the-exclamation-point icon is an alert. In Photoshop, it usually means that there are colors in the image that are out of that color space’s gamut, but in the histogram it is meant to alert you that the panel is showing cached information. Clicking on the icon will refresh the histogram information. In this exercise we are not changing anything, so it is not important.

C. Does the histogram have any gaps where information does not exist? This means that there is no image information in those areas, where gray values between black and white are expected. This is usually a result of over-tweaking an image with tonal adjustments. Small gaps are not noticeable and are a reasonable result of increasing contrast in an image, especially when certain areas have bright highlights.

In the above image, the histogram has no gaps. In the next exercise, we will be making changes to the histogram and you will see gaps as a result.

[edit] Exercise 3: Adjusting the image with Levels

For this exercise we will complete the first step (Levels) on the grayscale image that was used in Exercise 2. Then we will use the color version of the file again.

1. Click Image > Adjustments > Levels, which we will use to adjust the tonal range of our image. The Levels dialog box displays the histogram that we just viewed in the previous exercise. Shadows are represented on the left side, midtones in the middle, and highlights on the right.

2. Move the left input level sliders (the small triangles just beneath the graph) to the beginning of the shadow information. This sets the darkest tonal range in the original image at 100% black.

3. Move the right input level slider to the beginning of the highlight information. This sets the lightest tonal range in the original image to 100% white.

4. To adjust the contrast, move the contrast slider (the triangle below the middle of the graph) to the left.

The image on the left with the Levels dialog box on top of it is the "before" version of the file. The image on the right is the "after" version. Notice that the shadows are considerably darker and it appears there is more contrast between the dark and light areas of image information.

[edit] Exercise 4: Adjusting the image with Curves

1. Download rgb-trees.psd from the wiki and open it in Photoshop. Look at the Histogram panel to see information about the grayscale values in the image.

2. From the upper right pull-down menu in the Histogram panel, choose All Channels View to see the histogram for the composite RGB channel as well as the single red, green, and blue channels that comprise the image. Even though the image is seen in color, the overall scale of gray values should be evaluated. Notice the graphs in the Histogram panel for each of the three separate channels (ask the same questions as we posed when evaluating a grayscale image in Exercise 2).

Look at the individual histograms for the red, green, and blue channels. Notice that there is more highlight information in the red channel, while all three channels peak around the same point in the shadow areas. Also notice that the red channel has the most color information across the x-axis, while the other two channels have steeper slopes towards the start and ending points of the curves.

3. Click Image > Adjustments > Curves. Once again, the histogram is presented in the Curves dialog box. Curves, like Levels, can be used to adjust the tonal scale within the image.

4. This time, don’t touch the RGB composite curves. Instead, adjust each of the red, green, and blue graphs so that there is image information where the deepest shadows and lightest highlights appear. To do this, start by using the pull-down Channel menu in the Curves dialog box to select Red (Command+1). Use the input sliders on the left and right sides to readjust the tonal scale so that the shadows and highlights begin with image information.

5. Use the pull-down Channel menu to select Green (Command+2). Use the input sliders on the left and right sides to move the edges of the endpoints of the line graph to the point where image information exists.

6. Use the pull-down Channel menu to select Blue (Command+3). Use the input sliders on the left and right sides to move the edges of the endpoints of the line graph to the point where image information exists. Click OK.

Tip: Adjusting the Curves (or Levels, either panel could have been used for this last exercise) manually for each color channel produces a better result than making one adjustment in the composite channel.

7. Look again at the histogram. It should show a graph with information that spans from the left side of the x-axis (shadows) to the right side (highlights).

The image on the left is before the curves were altered and the image on the right after they have been modified. Information spans the entire x-axis on the histogram. Notice that the contrast is slightly modified, but the overall change to the image is slight. Be careful about pushing the sliders too far. The modifications should be minimal.

[edit] Exercise 5: Targeting saturation levels

Image > Adjustments > Hue/Saturation can be used to increase or decrease the saturation of specific hues within the image. This panel is often used to make a dominant color appear more vibrant in an image, but it is hard to notice if the image is not being viewed at 100 percent. Even then, sometimes it is easier to see the results of this image adjustment in the final print. If you are using the file included on the wiki, the following details are the adjustments that we made to demonstrate this concept.

1. Click Image > Adjustments > Hue/Saturation. Use the pull-down menu on the word "Master" to choose Green. You will work specifically on the green areas of the image.

2. Use the Saturation and Lightness sliders to modify the image. The image below demonstrates our settings, but remember that our monitors may be calibrated differently. It is best to eyeball these numbers, rather than follow our specific settings. Remember to be sure the image is showing at 100 percent (use the Zoom tool to zoom in or out) before making any adjustments.

[edit] Exercise 6: Sharpening the image

Whenever an image is scanned or captured digitally, the process of transforming the continuous tone of reality or analog photography into pixels on a grid results in a loss of crispness. Unsharp Mask is a filter that is commonly used to compensate for this loss. This filter finds edges by looking for contrast and increases the contrast of those pixels, while leaving the flat areas untouched. The resulting image looks sharper, without introducing noise into the image.

1. Click Filters > Unsharp Mask.

2. Be sure that the Preview button in the Unsharp Mask dialog box is checked. Look at the image while clicking on the Preview button. Un-checking it displays the “before state” and checking it shows what the image will look like after the filter is applied.

3. There are no set rules, but the guiding relationship is between the settings in this dialog box and file size. The larger the file size, the larger you will set the threshold, radius and amount. With smaller file sizes (anything less than 30 megabytes) you will probably leave the threshold at 0, the radius lower than 1.0, and adjust the percentage by eye between 20 and 250 percent.

4. You will know when you’ve gone too far. Too much contrast will create visible halos at every edge. Applying this filter should produce a minor modification. In general, if the change is obvious, your settings were probably too high.