PRESERVING HISTORY

How to Digitally Archive and Share Historical Photographs, Documents, and Audio Recordings

Chapter 3. Setting Tone and Color in Master Images

Key Points.

·        If the master images are adjusted for fading and exposure, the basic steps for tone and color adjustments for a photograph are:

o    Set the black and white points on the histogram to be on or just outside the lightest and darkest tones in the image. For typical color images, set the black and white points for each color channel as well as for the overall tone channel.

o    Adjust the mid-tone slider on the histogram if needed.

o    For color images, explore color adjustments with the gray eyedropper applied to different gray areas in the image. If needed, make manual adjustments to color with the histogram mid-tone sliders for each color channel.

o   Increase or decrease contrast with the curves or Shadows/Highlights adjustments if needed.

·        When scanning documents or photographs with a limited range of tone and color the tone and color settings are manually adjusted for the particular item rather than assuming the full range of tone.

·        If the master images represent the tone and color of the current condition of the items without adjustment for fading or exposure:

o   Tone and color in the master image are set using the white, black, and gray eyedroppers and mid-tone slider applied to a reference target included in the image.

o   Slides or paper originals without a reference target will have at most minor adjustments with the mid-tones slider and/or curves to visually match the original item. Color management, including a scanner profile, would be optimal.

o   Scanned images from negative film cannot be directly compared to the original negative and are usually processed to adjust for fading and exposure.

·        Tone and color adjustments for master images can be made during scanning or in a later step with image-editing software.

·        Tone and color adjustments with image-editing software such as Photoshop Elements and Photoshop CS can be done more quickly and reliably using layers.

·        The Shadows/Highlights adjustment in Photoshop Elements and Photoshop CS is often beneficial for photographs by amateur photographers and is a better alternative than curves for reducing contrast and bringing out detail in dark and light areas.

 

 

As noted in Chapter 2, historical master images typically require adjustments to tone and color. The tone at a point in an image is the overall darkness or lightness. Grayscale images for black and white photographs have tone without color. Color images have both tone and color. Adjustments to tone and color are needed whether the master images represent the current condition of an item or are adjusted for fading and exposure.

Tone and color adjustments for master images can be made either during the initial scanning or during a later step using image-editing software. Three basic strategies for workflow are:

A.     The scanning software is used to create output files with 16 bits per channel and minimal adjustments for tone and color. Image-editing software is used to make adjustments and to output the final master images, which may be either 16 or 8 bits per channel. Professional level image-editing software is optimal for this workflow, but some consumer level software can be used with certain limitations.

B.     The scanning software is used to make the major adjustments to tone and color with 16 bits per channel and to create output files with 8 bits per channel. Image-editing software is used to make refinements and to output the final master images. This workflow is often optimal with consumer level image-editing software.

C.     The scanning software is used to make the major adjustments and to output the master images without the use of image-editing software. In practice the master images are similar to option B above, but without the final refinement step. With current scanning software this results in some significant compromises for the master images, but may be appropriate when time or image editing capabilities are limited.

Appendix B provides an overview of several commonly used scanning software programs and which workflow strategies are appropriate for each program. The Guidance on Using Scanning Software at http://archivehistory.jeksite.com/chapters/scan_part1.htm describes the specific processes and settings for using each program for archival scanning. If the initial digital image is made with a camera, more extensive use of image-editing software is needed and the first option is preferable.

This chapter first introduces the basic methods for adjusting tone and color whether applied with scanning software or with image-editing software. Adjustments for master images with reference targets are discussed separately because different methods are used. After these basic methods have been described, some general information about using Photoshop Elements and Photoshop CS is provided. Then the specific techniques for color and tone adjustments with Photoshop Elements and Photoshop CS are described for master images that are adjusted for fading and exposure and separately for master images that represent the current condition of the item without adjusting fading and exposure.

3.1 Basic Methods to Adjust for Fading and Exposure

Histogram Black and White Points for Tone and Color

The histogram is the starting point for adjusting tone and color. A histogram for grayscale images lists horizontally each possible grayscale tone from black to white, and for each tone a vertical line indicates the number of pixels that have that tone in the image—or in the area of the image selected for scanning. No visible vertical line appears on the histogram for tones that do not appear in the image. Color photographs have four separate histograms, one for overall tone equivalent to a grayscale and separate histograms for red, blue, and green colors. The horizontal axis for each color histogram ranges from the maximum to minimum possible value for the color. A pixel that has the maximum value for all three colors is white and a pixel that has zero for all three colors is black. Most pictures have some areas that are white and some that are black so the histograms normally cover the full range.

Fading appears as a narrow range of tone or colors in the middle of the histogram, with significant areas on the dark and light ends having no pixels. White and black areas have faded into shades of gray. The histograms for red, blue, and green each have similar fading. The left histogram in Figure 3.1 shows a typical histogram for a faded black and white photograph.

Histogram adjustment controls include black point and white point sliders at each end of the horizontal axis that set which tones will be pure black and pure white in the output image. The left slider is usually the black end and the right slider the white end. When these sliders remain at the very ends of the histogram as shown on the left in Figure 3.1 there is no change in the output file. When the sliders are moved toward the middle, tones that are not pure black or white in the image will be converted to pure black and white in the output image. The tones at the location of the sliders and all tones from the sliders to the corresponding end of the histogram are set to pure black or pure white in the output image.

 

 

 

Histogram tone adjustments for a faded historical photograph

Figure 3.1. These histograms are for the faded black and white military photograph in Figure 2.2 taken about 1944. The horizontal axis represents possible color tones in the image from black on the left to white on the right. The height of the vertical line or bar indicates the number of pixels in the image that have the tone indicated on the horizontal axis. For this photograph the dark and light tones on the ends of the horizontal axis have no pixels (vertical lines) because the photograph has faded to shades of gray.

The histogram settings on the left are the default settings that cause no changes. The histogram settings on the right have the black and white point sliders set to adjust for fading. The small triangles below the histogram are the sliders for setting the black point on the left and white point on the right. The sliders are set to the darkest and lightest tones that occur in the image. The middle slider adjusts the mid-tones in the image without changing the black and white points. The Output Levels slider sets the exact tones that are used for the black and white points in the output file and shifts other tones to fit within this range.

These histograms are from Photoshop CS.

 

 

 

The goal for faded images is to move the black and white point sliders in to or near the darkest and lightest tones that actually occur in the image. The histogram on the right in Figure 3.1 shows the black and white point sliders set for the faded military photograph in Figure 2.2 in Chapter 2. These settings convert the darkest and lightest tones to pure black and pure white, which is the expected range for a typical photograph. The various shades of gray between the black and white points are spread out to fill the histogram. When done properly, this spreading of tone corrects fading and increases the contrast in the image in a way that enhances details. For color images, color shifts due to fading are often corrected when the histogram black and white points are set individually for the red, green, and blue color channels.

Clipping

All histogram adjustments must be done carefully to avoid clipping, which is moving the black and white points so far that areas that should be dark gray or off white become pure black or white. Clipping occurs anytime the black and white point sliders are set inside the darkest and lightest tones in the image. Histogram displays are often small, particularly in scanning software, and the vertical lines may be visible only for tones with many pixels in the image. This can easily result in unintended clipping. Most image-processing software has options for automatically setting the black and white points to the darkest and lightest tones in the image with no clipping.

Documents and Photographs with Limited Colors

When making a digital image of a document or photograph with limited tone and color, the white and black point settings on the histogram are adjusted for the particular document rather than assuming the full range of tone. In addition to documents, this applies to photographs taken in blowing snow or in dim light in the evening and to sepia photographs scanned in color. In these cases the black and white points need to be set based on the content of the image rather than on the assumption that the image has the full range of tone. Manual adjustments of individual colors are generally needed. Useful preset values can be developed as a starting point.

When the writing on the backside of a page shows through when scanning, placing black paper behind the page is usually beneficial. It is much easier to adjust for the uniform darker color than to remove visible writing.

Manually Minimize Distortion from Borders and Defects

Dust, damage, borders, rounded corners, and other artificial black and white areas in an image can interfere with setting the appropriate black and white points. In these cases, the darkest and/or lightest tones in the histogram are often artificial rather than actual content of the image. Dust, damage, and borders are black in images from slides and white in images from negatives. For images scanned from paper prints, borders are often white, dust on the paper is black, and damage can be almost any color. The basic histogram does not indicate which pixels are from artificial areas and which are useful content of the image. The degree of interference depends on the amount of fading and other characteristics of the image. In some cases the effects of the artificial areas will be only a slight softening of contrast. In other cases the effects can be more significant.

One of the best methods for handling these problems is to display the actual pixels that are being clipped for a particular setting of the black and white points. A display of clipped pixels on the image can be used to verify that clipping only occurs in the artificial areas. As described in Appendix B, the VueScan scanning software has an option to display clipped pixels and similar options are described for Photoshop CS and Photoshop Elements later in this Chapter.

Another strategy for reducing the problems from artificial black and white areas is to exclude them from the area that is being used to determine the black and white points. The area selected for scanning can be reduced or a mask can be applied with image-editing software. After the black and white points have been determined, the area that will be adjusted with the histogram can be expanded back to the original size before the adjustment is applied. Also, applying infrared dust and scratch corrections before the histogram adjustments reduces the adverse impacts of defects when setting the black and white points. As discussed in Appendix B, some scanning software can apply infrared dust and scratch correction before histogram adjustments. Of course, adjustments with image-editing software will be after infrared dust and scratch corrections.

A third method that is sometimes recommended for setting the black and white points is to use the black and white eyedroppers. Most histogram adjustment controls include black and white eyedropper tools that can be used to manually select points on the image that will be set as the black and white points. Every tone that is darker than the selected black tone or lighter than the selected white tone is clipped. I find it very difficult to visually identify the actual lightest and darkest points on an image. These eyedroppers are much more likely to produce unwanted clipping than the methods described above. As discussed in a later section, the black and white eyedroppers are used when a reference target is included in the image.

Avoid Extreme Black and White in the Output Image

As noted in Chapter 2, the preferred practice is to set the darkest and lightest tones in the output image to not be pure black and white. This result can be achieved in two ways.

One simple method for making the darkest and lightest tones not pure black and white is to set the black and white sliders a short distance away from the darkest and lightest tones in the image. This method is usually optimal with software that does not display or otherwise precisely track clipped pixels. Leaving space at the ends of the histogram minimizes the possibility that some pixels in the image are being clipped but are not visible on the histogram. This method is particularly appropriate for histograms with limited resolution.

Alternatively, the histograms in most scanning and image-editing software have additional output sliders that can set the darkest and lightest tones in the output image to be different than pure black and white. These sliders compress the range of tones in the output image. With this method the black and white point sliders are typically set at the darkest and lightest tones in the image, which is appropriate when clipped pixels can be displayed or are otherwise precisely tracked. These output sliders can be set to exact numeric values. The axis for a histogram typically ranges from 0 for pure black to 255 for pure white. The histogram on the right in Figure 3.1 includes an output slider set to 5 for the darkest tone in the output image and 250 for the lightest tone. As noted in Chapter 2, values of 8 and 247 are considered best practice.

When scanning software or image-editing software has options to automatically and precisely set the black and white points to have no clipping, the output sliders can be used to efficiently set the dark and light tones in the output image. For Figure 3.1 the output slider settings combined with the black and white point settings quickly make the output image have the desired range of tones.

Input and Output Histograms

The histogram displayed for adjustments is usually from the input image before the adjustments are applied. The histogram for the output image after adjustments have been applied is also useful. Most scanning and image-editing software have options to display both the input and output histograms.

The criteria for evaluating the adjustments are different for input and output histograms. The key criteria for an input histogram is that the sliders for the black and white points are set at or just outside the darkest and lightest tones that occur in the image as shown in Figure 3.1. The key criterion for an output histogram is that the tones in the image are spread over almost the entire width of the histogram. Ideally the lightest and darkest tones in the output histogram will be near but a short distance away from the ends of the histogram axis. Figure 3.2 shows the output histogram for the input histogram adjustments shown on the right in Figure 3.1.

 

 

 

 

 

Output histogram for tone adjustments of a faded historical photograph and for comparing 16 and 8 bits per channel

Figure 3.2. The output histogram on the left has been spread out to almost cover the full range of tone. This is the output histogram for the adjustments in Figure 3.1.

The output histogram on the right shows systematic gaps from spreading out the histogram with only 8 bits per channel. This applies the same adjustments as the output histogram on the left, except with 8 bits per channel rather than 16.

These output histograms are from Photoshop CS.

 

 

 

Clipping is indicated on an output histogram by the tones in the image extending completely to the ends of the histogram axis and an apparent cutoff at the end. A vertical spike at the very end of the histogram axis is a definite indication of clipping. Clipping is indicated on an input histogram by tones in the image occurring outside the black or white point sliders.

An output histogram can also show the posterization banding that occurs when the bit depth is not adequate for the adjustments to an image. The histogram on the right in Figure 3.2 shows the output histogram for the same adjustments as on the left, except the adjustments are applied to a copy of the image with 8 bits per channel rather than 16 bits. The systematic gaps in the histogram result from spreading the histogram with the limited shades of tone with 8 bits per channel. The gaps indicate discrete artificial bands of tone rather than continuous tones.

Note that gaps in the histogram occur even with the relatively mild fading in this photograph (Figure 2.2). It is also important to note that the bands are not noticeable in the output image. In fact, bands that are visible in the histogram often are not noticeable in the output image and many historical images could be adjusted with 8 bits per channel. However, the artificial bands are noticeable in some images and always indicate a loss of information that could easily be avoided. The standard practice of making the major tone and color adjustments with 16 bits per channel avoids the risks from information loss due to artificial gaps in the histogram. As noted in Chapter 2, once the major adjustments have been made, further refinements can be done safely with 8 bits per channel if needed.

Mid-tone Adjustment of Tone

After the black and white points have been set, the mid-tone or gamma slider on a histogram can be used to adjust the brightness of the image. The histograms on most scanning and image-editing software have a middle slider that adjusts the tones in the middle of the histogram without changing the black and white points. The middle slider is shown set to 1.10 in Figure 3.1. This slider often provides useful adjustments. The term gamma refers to mathematical equations that enhance the brightness of mid-tones similar to the way human vision handles mid-tones. If the black and white points have been set to clip some pixels, the mid-tone adjustment may alter which pixels are clipped.

Gray Eyedropper and Color Adjustments

Histograms usually have a gray eyedropper tool that is used to make color adjustments. This eyedropper is used only with color images. After the gray eyedropper is activated the mouse pointer is moved over a spot on the image that should be a shade of gray. A white wall in the shade, a weathered gray board, or a section of concrete is a good choice for gray color adjustments. When the mouse is clicked on the selected spot the colors are adjusted to make equal amounts of red, green, and blue at the spot. This produces gray, but is done in a way that does not change the overall tone at the spot. For most software the gray eyedropper operates by automatically adjusting the mid-tone sliders for each color without altering the black and white points for the histogram. The technique works with a wide range of shades of gray. Different points can be tried until a good adjustment is found. This method can be remarkably effective. Minor revision of the mid-tone slider for overall tone may be beneficial after the gray eyedropper has been applied to adjust color.

Note that the gray eyedropper operates differently than the black and white eyedroppers even though they are usually located together on the histogram controls. The gray eyedropper removes color casts rather than sets a specific shade of gray tone. On the other hand, the black and white eyedroppers are primarily intended to set specific black and white tones and are applied only to spots that should be set to pure black or pure white. The black and white eyedroppers sometimes modify color casts, but that is not their primary purpose. The black and white eyedroppers can cause substantial clipping while the gray eyedropper generally does not.

If additional manual adjustments to color are needed, the mid-tone sliders can be adjusted for the individual color channels on the histogram. In addition, the curves adjustment described below can be used to modify colors.

Curves Adjustment of Contrast

The curves adjustment can be used to increase or decrease contrast in an image without changing the black and white points. The Curves adjustment is a line or graph that shows how each possible tone on the input image is modified for the output image. The tones of the input image are displayed across the bottom, typically with black on the left and white on the right. The corresponding tones of the output image are displayed vertically. If the graph is a straight diagonal line, the output image is the same as the input image for every tone.

To increase contrast without shifting the black and white points, the lower left part of the curve is shifted down to make the dark areas darker and the upper right is shifted up to make the light areas lighter. This produces a curve that is vaguely S shaped as shown in Figure 3.3. Contrast can be reduced in an image by making the opposite shaped curve, increasing the dark areas and decreasing the light areas. Also, tone can be adjusted for only part of the tonal range. The curve on the right in Figure 3.3 makes the dark areas darker without changing the light areas of the image.

 

 

 

Figure 3.3. The curves adjustment on the left has a S shape to increase contrast in an image. The horizontal axis represents color tones in the input image from black on the left to white on the right. The vertical axis represents color tones in the output image from black on the bottom to white on the top. A straight diagonal line would indicate the output image matches the input image. Setting the left side of the curve lower makes the dark tones darker in the output image. Similarly, setting the right side of the curve higher makes the light tones lighter in the output image. The end result is increased contrast. Small changes in the shape of the curve produce significant effects.

The curves adjustment on the right makes the shadows darker but does not alter the highlights of the image. Smaller modifications than shown here will often be applied.

These curves are from Photoshop CS. The histogram is displayed in the background of the curve.

 

 

Curves Adjustment of Color

As with the histogram, there are 4 channels for curves, one for overall tone and others for red, green, and blue colors. Contrast adjustments are made with the overall tone curve and color adjustments can be made by shifting the curves for the individual colors. This is useful when the colors need to be adjusted for certain ranges of tone, such as a color cast for shades of dark blue.

Other Methods for Adjusting Tone and Color

Other options for making tone and color adjustments are described for image-editing software later in this chapter. When an image requires major reductions in contrast the Shadows/Highlights adjustments in Photoshop CS and Photoshop Elements described below are significantly more sophisticated and less prone to color shifts than a basic curves adjustment in scanning software. Making such adjustments in Photoshop is often the best strategy. The Guidance on Using Scanning Software at http://archivehistory.jeksite.com/chapters/scan_part1.htm describes the options and processes for making tone and color adjustments with specific scanning programs.

Note that simple brightness and contrast adjustments done without using the histogram or curves usually shift the black and white points in ways that cause significant unwanted clipping. These methods should be avoided when working with archival images.

3.2 Basic Methods with Reference Targets

When a reference target is included in a master image, tone and color adjustments are typically set using the black, white, and gray eyedropper tools and the mid-tone slider on the histogram. The black point eyedropper tool is used to set the black point by placing the mouse pointer over the black color patch in the reference target and clicking. The black point is set to the tone at that spot. A corresponding step is done to set the white point. The gray eyedropper is used to remove color casts by clicking on a gray patch in the reference target. The output black and white tone values are set with the histogram sliders described earlier for avoiding extreme black and white tones in the output image.

The typical sequence is to set the black point, set the white point, apply the gray eyedropper to adjust color, set the output black and white maximum values, and finally use the mid-tone slider to set the correct tone for the middle gray patch on the reference target. The specific values for each of these steps were presented in Chapter 2.

3.3 Using Photoshop Elements and Photoshop CS

Photoshop CS (Creative Suite) ($620-$700) is the professional standard for image editing. Photoshop Elements is a widely used and much less expensive ($80-$100) consumer grade application. This chapter discusses the use of both applications for preparing master historical images. Most functions needed for processing historical images are available with Photoshop Elements as well as with Photoshop CS. The specific versions used in preparing this chapter are Photoshop Elements 9 and Photoshop CS5. The features described here will almost certainly be present in future versions of these programs. Almost all of the features described here were present in previous versions of Photoshop CS, but that is not true for Photoshop Elements. Photoshop Elements 9 was a significant enhancement over previous versions.

One limitation of Photoshop Elements is that it does not fully handle images with a bit depth of 16 bits per channel. As discussed later, certain basic adjustments to tone and color can be done with 16 bits per channel in Photoshop Elements, but many useful capabilities are not available. The limited capabilities can be used for the initial adjustment of tone and color in a way that works but is not optimal. Once the tone and color of an image are set reasonably well, processing with 8 bits per channel will normally be acceptable. The image can be converted to a bit depth of 8 bits with the dropdown menu Image> Mode> 8 Bits/Channel. After this conversion, all editing functions work. The title line of the image indicates the bit depth (e.g., “gray/16”) and the menu items that do not work are set to gray, but there often are no warning messages as to why most functions do not operate when the image has 16 bits per channel.

The basic workspace layout for both Photoshop Elements and Photoshop CS is a large image display area with vertical tool bar on the left and a series of panels on the right that display adjustment tools and layers. The usual main dropdown menu bar is across the top. Controls for the active tools or adjustments are on a bar below the main dropdown menu bar. Keyboard shortcuts are a key factor in using Photoshop programs efficiently. A cheat-sheet with the most commonly used shortcuts can save a lot of time and effort, particularly for those who do not use Photoshop daily. The notes I use are available at http://archivehistory.jeksite.com/download/download.htm.

For many of the techniques described here, it is very valuable to be able to quickly enlarge an image by zooming in and then move around the zoomed image. The Hand Tool moves an image by clicking and dragging, and is a good default state for the mouse pointer because the image cannot be unintentionally modified with it. The Hand Tool is activated by tapping the H key or by clicking the Hand Tool icon on the left toolbar. When any other tool or adjustment is active, holding down the spacebar converts the mouse pointer to the Hand Tool and allows movement of the image. The mouse pointer reverts back to the other tool or adjustment when the spacebar is released. Setting the mouse scroll wheel to zoom is very handy and is done with the dropdown menu Edit> Preferences> General> Zoom with Scroll Wheel. The scroll wheel zoom combined with the spacebar hand tool can make navigation on an image very quick and apply with both Photoshop Elements and Photoshop CS.

Working with Layers

Modifications to an image can be made most effectively using layers that overlay the main image but can be controlled and modified separately from the main image. The advantages of layers include that the modification can be easily compared with the previous state, can be easily canceled, can be easily duplicated, can be partially implemented, can be implemented to different degrees in different parts of the image, and often can be revised later. Another key point is that layers also protect the initial image from degradation due to repeated manipulations, which is important when exploring effects. Layers can be combined or blended in ways that produce innovative and very useful results. After the modifications are completed, the layers can be flattened or combined to produce a final image. Images with layers can also be saved to a file if further adjustments may be needed later. However, images with layers produce large files that are less reliable over the long-term than flat images.

In Photoshop Elements and Photoshop CS, different layers are indicated by rows in the Layers panel on the right of the screen. Clicking the eye icon on a row toggles turning the effects of the layer on or off. This toggle can be used to easily examine the effects of the layer. After a layer is selected by clicking on its row in the panel, the opacity for the layer can be reduced from the default of 100%. A lower opacity partially implements or fades the effects of the layer. In the Layers panel, the blend mode field is to the left of the opacity field. Blend mode determines how the layers combine and is set to “Normal” by default. Alternative blend modes will be described later when relevant. Layers are flattened using the dropdown menu Layer> Flatten Image or by right-clicking on the Layers panel and selecting Flatten Image.

In Photoshop Elements 9, layers cannot be used with images with 16 bits per channel. As discussed earlier one option for working with this limitation is to do the processing that requires 16 bits per channel in the scanning software and make refinements with 8 bits per channel in Photoshop Elements. Another option is to make adjustments to tone and color directly to the image rather than with layers. As discussed below Photoshop Elements can make adjustments for tone and color with 16 bits per channel if the adjustments are made directly to the image. This sacrifices the many advantages of working with layers, but it will work.

When making adjustments directly to an image the Preview checkbox should be used to verify the results before closing an adjustment dialog box. Once an adjustment has been applied directly to the image the adjustment can be revised only by either adding other adjustments or by undoing the previous adjustment and reapplying the adjustment. In general, when making changes directly to an image undoing an adjustment that does not work as hoped maintains the integrity of the image better than adding other adjustments to compensate. The dropdown menu Edit> Undo displays the type of adjustment that will be undone.

A layer mask is a more advanced technique that can be used to make the effects of the layer apply partially or not at all in different parts of the image. Layer masks may occasionally be used in preparing master images, but are more commonly applied when preparing images for display.  They are described in Chapter 5.

Practical Optimization of Layers

After an adjustment has been initially implemented on a layer, in some cases it is possible to return to that layer and revise the adjustment. In other cases, the adjustment on a layer cannot be revised. A related point is that certain adjustments and filters currently require that all lower layers be merged to a new layer before the adjustment or filter can be applied. After one of these merged layers has been created, revisions to lower layers are not propagated up to the merged layer. For example, a layer could be used to adjust color. A layer on top of that could be used to apply the Shadows/Highlights adjustment. The layer for Shadows/Highlights requires that the lower layers be merged and therefore any revisions to the lower color adjustment layer will not show on the layer for Shadows/Highlights. Putting the merged layers first (bottom) or last (top) in the stack of layers provides the greatest flexibility.

Certain adjustments are available with built-in layers and these have important advantages. The adjustments with built-in layers can be revised and any revisions to lower layers propagate up through adjustments with built-in layers. These built-in layers also have masks ready for use.

A layer needs to be created for an adjustment or filter that does not have a built-in layer. A new layer can be created is several ways.

·        A blank layer can be created with Alt-Ctrl-Shift-N or with the dropdown menu Layer> New> Layer. A blank layer is appropriate for certain adjustments as described below.

·        A new layer that merges all lower layers is created with Alt-Ctrl-Shift-E. Normally this is done when the top layer is active, which can be set by clicking the top layer in the layers panel, or by tapping “Alt-.”.

·        A layer can be duplicated with Ctrl-J or by right-clicking a layer in the Layers panel and then clicking “Duplicate Layer”. This duplicates one layer and does not merge layers. It can be used to make a layer when there is only the background image, or to duplicate a layer to increase an effect.

Alt-Ctrl-Shift-E does not work if the background layer is present with no other layers added. Ctrl-J can be used to create a new layer in this situation or Alt-Ctrl-Shift-N followed by Alt-Ctrl-Shift-E can be applied.

In Photoshop CS smart objects or smart filters allow adjustments to be revised after the initial implementation. This is useful for certain adjustments that are described below. Also smart objects have a mask for the layer. After a layer has been created, right-click the layer in the Layers panel and then click on “Convert to Smart Object.” Alternatively, tap Ctrl-Shift-S or use the dropdown menu Filter> Convert for Smart Filter. Smart objects are not available with Photoshop Elements.

3.4 Using Photoshop Elements and CS to Adjust for Fading and Exposure

Histogram Black and White Points for Tone and Color

Histogram adjustments are called Levels adjustments in Photoshop and the operation is basically the same in both Photoshop Elements and Photoshop CS. Histogram or levels adjustment layers can be initiated with the icons in the Adjustments panel on the right of the screen, or with the main dropdown menu Layer> New Adjustment Layer> Levels. However, if an image has 16 bits per channel, layers do not work in Photoshop Elements. In these cases, the Levels adjustment can be done directly to the image without using layers, or alternatively, the image can be converted to 8 bits per channel if major adjustments are not needed. Tap Ctrl-L or use the dropdown menu Enhance> Adjust Lighting> Levels to initiate a levels adjustment directly to the image. Figure 3.1 discussed earlier shows the histogram adjustment for Photoshop CS. The histogram for Photoshop Elements is very similar.

The first step is to set the default settings for auto histogram adjustments. Open a Levels adjustment layer as described above. Then hold down the Alt key and click the Auto button on the Levels adjustment panel, as can be seen if Figure 3.1. This brings up the dialog box for the settings for the auto adjustment. Click the button for Enhance per Channel Contrast. This instructs the auto levels adjustment to set the black and white points in the histogram separately for each color. Then set both the Shadows Clip field and the Highlights Clip field to zero, which will set the black and white point to the darkest and lightest tones in an image without clipping. Make the checkbox for Snap Neutral Midtones checked. With this setting, the auto color processing will find an area that appears to be gray and make color adjustment based that area. I recommend leaving the Target Colors for Shadows and Highlights to the default values (0 and 255) because there are more obvious methods for adjusting those. Finally, check the box at the bottom for “Save as defaults” and click Ok to close the dialog box. In Photoshop Elements, the default settings apparently must be set with a Levels adjustment layer. Once set, the default settings are used with direct adjustment to the image, but apparently cannot be set with direct adjustment.

After the default settings have been specified, clicking the Auto button on the Levels panel automatically sets the black and white points to the darkest and lightest tones in the image. For color images the black and white points are set to the darkest and lightest tones for each color channel. In a few cases, the auto setting for Find Dark & Light Colors produces better results than Enhance per Channel Contrast. Open the Auto Color Corrections dialog box by Alt-clicking the Auto button and click on the button for Find Dark & Light Colors. When the button is clicked, the adjustments are immediately displayed on the main image. Clicking the button for Find Dark & Light Colors will toggle back to that adjustment. If the Find Dark & Light Colors is preferred for an image, click Ok when that button is selected. The default settings remain as set previously and will be used when the Auto button is applied for other images.

As noted earlier, these auto settings will generally not be optimal for items with limited ranges of color, such as documents, sepia photographs, and photographs taken in snow or at sunset.

The histogram for the output image after adjustments have been applied can be displayed by clicking the Histogram tab toward the top of the panels on the right. If the Histogram tab is not shown, select the dropdown menu item Window> Histogram. The components of the histogram that are displayed can be selected from the Channel: pick-list right above the histogram chart.

For both the main histogram and the output histogram, a small warning icon with an exclamation point may display to indicate that the histogram is not fully displaying some changes. Click the warning icon to update the display.

Manually Minimize Distortion from Borders and Defects

Examining clipped pixels while manually adjusting the white and dark point sliders is an easy, effective method for handling images that have borders or defects. This can be very useful when borders or defects influence the results of the auto adjustment. Manual refinement of the black and white point settings may be needed in these cases. The best practice is to know exactly which pixels are clipped.

Any pixels that are clipped with a histogram adjustment can be seen by holding down the Alt key while clicking or dragging the black or white point slider. When the black point slider is clicked while the Alt key is held down any colors except white indicate clipped pixels. Different colors indicate different color channels are clipped. Each color channel can be examined separately. Black indicates all color channels are clipped. When white point slider is clicked with the Alt key held down, any colors except black indicate clipped pixels.

Another strategy is to exclude borders and defects when the auto settings are being determined, and then apply the settings to the full image. Layer masks mentioned earlier and described in Chapter 5 can be used to exclude borders and damage. The masks generally do not need to be precisely aligned with the borders and damage. An easy way to develop a mask for an image with borders and edge defects is to select a rectangular area before creating the Levels adjustment layer. The selected area will automatically become a mask when the new layer is created.

After the mask has been set and the Auto button clicked to determine the adjustment settings, the mask can be disabled by putting the mouse pointer over the mask icon in the Layers panel and holding down the Shift key while clicking the mouse. Alternatively, the right mouse button can be clicked and “Disable Layer Mask” selected. With the mask disabled, the auto settings apply to the full image. If the Auto button is clicked again while the mask is disabled, the settings will continue to be based on the masked area.

Avoid Extreme Black and White in Output Image

The Levels adjustments in both Photoshop Elements and Photoshop CS include an additional slider to set the tones for the black and white points in the output image. As shown in Figure 3.1, these sliders are labeled “Output Levels” and appear below the histogram chart. The left slider sets the specific numeric tone value for the darkest possible tone in the output image and the right slider sets the value for the lightest possible tone. These sliders compress the range of tones to make the darkest and lightest tones have certain values. They do not introduce additional clipping.

Mid-tone Adjustment of Tone

After the black and white points have been set, the mid-tone or gamma slider on the histogram can be used to adjust the brightness of the image without changing the black and white points.

Gray Eyedropper and Color Adjustments

As noted earlier, the gray eyedropper is a very useful method to make color corrections. This adjustment is most effectively and safely explored with a new Levels layer, which can be initiated with the dropdown menu Layer> New Adjustment Layer> Levels. As shown in Figure 3.1, the Levels adjustment has icons for three eyedroppers. Click the middle icon for the gray eyedropper. This converts the mouse pointer to an eyedropper. Move the tip of the mouse pointer over a spot on the image that should be a shade of gray and click the mouse. The colors on the image will be adjusted to make the clicked point a shade of gray without changing the overall tone. A white wall in the shade, a weathered gray board, or a section of concrete is a good choice for gray color adjustments. The gray eyedropper remains active until the icon is clicked again or until another tool is selected, such as by tapping the H key. Clicking a different spot makes the new spot the active adjustment point. Different points can be tried until a good adjustment is found. This method can be very useful in some cases that are difficult with other methods. The layer can be deleted if improved results are not found.

Additional color corrections can be made manually if needed. The histogram can be used to adjust the red, green, and blue colors individually as well as the overall tone. The channel pick-list to the left of the Auto button is set to RGB by default, which is overall tone. Click on the pick-list and select the red, green, or blue channel to display the histogram for one color channel. The mid-tone slider can be adjusted for an individual color. In addition, the white and black points can be set manually for a channel by dragging the end sliders.

If color adjustments cause unwanted shifts in overall tone for the image, the layer blending mode can be set to Color. This will reduce shifts in tone for the color adjustments.

Shadows/Highlights Adjustment of Contrast and Tone

The Shadows/Highlights adjustment brings out detail in areas of an image that are too dark or too light and is one of the most useful features in Photoshop Elements and Photoshop CS. It is effective with photographs that are over exposed, are under exposed, have backlighting, or have excess contrast in bright sun. With a faded photograph, excess contrast may not be apparent until after a histogram adjustment. The Shadows/Highlights adjustment in Photoshop is a sophisticated algorithm that produces results with minimal impacts on other areas of the image.

At least minor application of the Shadows/Highlights adjustment is beneficial for the majority of photographs taken by amateur photographers. It is not available with built-in layers and is appropriate for use as a smart object. Appendix E describes how a hotkey can be configured in Photoshop CS to implement a series of steps such as creating a smart object layer for the Shadows/Highlights adjustment.

After creating a new merged layer with Ctrl-J if this is the first layer or Alt-Ctrl-Shift-E if there are other layers, initiate Shadows/Highlights in Photoshop Elements with the dropdown menu Enhance> Adjust Lighting> Shadows/Highlights. This brings up three sliders. The top slider adjusts shadows and the middle slider adjusts highlights. I start by setting both all the way to the left, and then increase the highlights slide until the optimal has been produced. Then I increase the shadows slider. The additional slider for Midtone Contrast allows compensation for effects that sometimes occur with shadow and highlight adjustments, and a slight increase is often beneficial.

As with the Levels adjustment, the Shadows/Highlights adjustment in Photoshop Elements can be used directly on an image with 16 bits per channel, but it cannot be used with layers for images with 16 bits per channel. Initiating the Shadows/Highlights adjustment without creating a new layer makes the adjustment operate directly on an image with 16 bits per channel.

Photoshop CS has more options for controlling the Shadows/Highlights adjustment. Create a new merged layer as described above, make the layer a smart object with Ctrl-Shift-S, and then initiate Shadows/Highlights with the dropdown menu Image> Adjustments> Shadows/Highlights. This brings up a series of sliders. I usually do the highlights first. Set the Amount slider all the way to the left and then slide it to the right until the optimal degree of change has occurred. The associated Tonal Width slider can be adjusted to make the changes occur over a larger or smaller tonal range and are used to refine the effects. I usually start with 20% or 30% for highlights. The Radius slider controls how many adjacent pixels are used in making the adjustment and provide a final refinement of the effect, but usually has little effect for highlights. Repeat this process for the shadows sliders. I start with Tonal Width set to 40% for shadows. Reducing the radius slider from the default of 30 pixels can lighten smaller dark areas. The additional sliders for Color Correction for color images, Brightness for grayscale images, and Midtone Contrast allow compensation for effects that sometimes occur with shadow and highlight adjustments. Increasing the Midtone Contrast slightly is often beneficial. I have my preferred settings for all the sliders saved as the default by clicking the button below the sliders.

If undesired color shifts occur with a Shadows/Highlights adjustment, they can be reduced by setting the blending mode to Luminosity rather than Normal. This is a safe standard practice.

Curves Adjustment of Contrast

The Curves adjustment is primarily used to increase contrast in certain tonal ranges. The Shadows/Highlights adjustment is generally preferable for reducing contrast. The contrast of an image is increased by making the darker tones darker and the lighter tones lighter. This produces an S shaped curve for the RGB curve channel as shown for the left curve in Figure 3.3. An important feature of curves is that the certain tones can be adjusted, such as enhancing the dark tones, without altering the light tones, as shown with the right curve in Figure 3.3. Also, contrast can be changed with curves without changing the black and white points.

The Curves adjustment in Photoshop Elements has sliders for modifying different parts of the curve. There are also styles that can be used to produce common effects. The Curves adjustment in Photoshop Elements does not have a built-in layer. After creating a new merged layer with Ctrl-J if this is the first layer or Alt-Ctrl-Shift-E if there are other layers, the Curves adjustment is initiated with the dropdown menu Enhance> Adjust Color> Adjust Color Curves. The Curves adjustment in Photoshop Elements 9 does not work with images with 16 bits per channel.

The Curves adjustment is available with a built-in layer in Photoshop CS and is initiated with the Curves icon on the Adjustment panel or with the dropdown menu Layers> New Adjustment Layer> Curves. The list menu in the upper right corner of the Curves adjustment panel includes an option for “Expanded view” that increases the size of the display. The channel for tone or color is set with a pick-list above the curve graph. The histogram is displayed on the curve by default. The Curves adjustment has black and white sliders on the bottom that can be moved along the histogram and produce the same result as a Levels histogram adjustment. These can be set for the individual color channels as well as for overall tone. As with the Levels adjustment, holding down the Alt key while moving the black and white sliders shows the pixels that are clipped. Note that the slope of the line increases when the black and white sliders are moved and this indicates an increase in contrast. When the histogram is modified, the warning icon on the lower left for “Calculate a more accurate histogram” may appear. Clicking on this icon will update the histogram.

In Photoshop CS, clicking on a point in the curve creates a set point that can be dragged to modify the shape of the curve. Moving a point on the curve up or down increases or decreases the output tone or color. Other set points may be needed to prevent a section of the curve from moving. If the mouse pointer is moved over the main image, the cursor becomes an eyedropper and a circle appears on the curve indicating the tone at the location of the mouse pointer. If the mouse is clicked while it is over the image, a set point on the curve is created for that tone and the point on the curve can be adjusted my dragging the mouse over the image. This feature is toggled on and off by clicking the icon for “Click and drag in image to modify the curve” in the upper left of the adjustment panel.

With both Photoshop Elements and Photoshop CS, undesired color shifts sometimes occur when significant modifications are made with a Curves adjustment. The color shifts can be reduced by setting the blending mode for the curves layer to Luminosity rather than Normal. This is a good standard practice when a curves adjustment is intended for only changes in tone.

Curves Adjustment of Color

The Curves adjustment in Photoshop CS allows adjustments to be made to each color channel separately. This can be used when different color adjustments are needed for light or dark areas of the image or for different colors. The Curves adjustment in Photoshop Elements does not allow adjusting individual color channels. The Levels adjustment with layer masks is an alternative in those situations.

3.5 Using Photoshop Elements and CS with Reference Targets

The Levels adjustments in Photoshop Elements and Photoshop CS have black and white eyedroppers that can be used to set the black and white points when a reference target is included in the image. Clicking on an eyedropper icon activates the eyedropper. The mouse pointer becomes an eyedropper, which is placed over the appropriate color patch on the reference target and then clicked. The specific tone values for black and white in the output image can be set with the sliders as described earlier for avoiding extreme black and white in the output image. The gray eyedropper can be applied to a gray patch on the reference target to remove color casts. After the eyedroppers have been applied, the mid-tone slider is adjusted to set the needed tone value for the mid-gray patch on the reference target.

The numeric color values can be checked using the Info panel near the top of the panels on the right of the screen. Click the Info tab. If the Info tab is not displayed, select the dropdown menu item Window> Info. The Info tab shows that actual numeric values for the red, green, and blue color channels at the location of the mouse pointer on the image.

3.6 Other Adjustments

In addition to adjusting tone and color, other modifications to the master images may sometimes be needed. Chapter 5 describes how to make other modifications using Photoshop Elements and Photoshop CS, such as changing bit depth, changing working color space, or applying grain reduction.

Resources

Andrews, Philip, 2011, Adobe Photoshop Elements 9 for Photographers. Published by Focal Press in Burlington, MA.

Evening, Martin, 2010, Adobe Photoshop CS5 for Photographers. Published by Focal Press in Burlington, MA.

 

[Version of 8/12/2015]


Historical photos and documents

Contents of Chapter

Chapter 3. Setting Tone and Color in Master Images

3.1 Basic Methods to Adjust for Fading and Exposure

3.2 Basic Methods with Reference Targets

3.3 Using Photoshop Elements and Photoshop CS

3.4 Using Photoshop Elements and CS to Adjust for Fading and Exposure

3.5 Using Photoshop Elements and CS with Reference Targets

3.6 Other Adjustments

Resources

Chapters in the Book

Introduction: Background, Purpose, and Example Archive Projects

1. Basic Principles of Archiving Photographs and Documents

2. Good and Best Practices for Making Digital Images

3. Setting Tone and Color in Master Images

4. Documentation for Historical Images

5. Preparation and Display of Historical Photographs and Documents

6. Good and Best Practices for Making Audio Recordings

7. Preparation and Presentation of Historical Audio Recordings

8. Historical Slide Shows

9. Copyrights and Other Legal Rights

Appendices

A. Practical Color Management

B. Overview of Scanning Software

C. Effective Resolution of Scanners

D. Using Copy Stands with Cameras

E. Batch Processing of Images

Other Resources

Guidance on Using Scanning Software

Guidance on Using Slide Show Software

Downloads and Links

This website and book were developed by Jim Kennedy.

Email: archivehistory@jeksite.org

© 2012, 2013, 2014, 2015 James E. Kennedy