The 32-bit/channel file can be saved if you like, but must first be downsampled to a 16-bit/channel file in order to use most of the Photoshop tools. To downsample the 32-bit/channel file into a fully editable 16-bit/channel file, from the Image Menu select Mode and choose 16 Bits/Channel. An HDR Conversion dialogue box will open and allow you to choose one of four methods to convert to a 16-bit/channel file. Most often, the default method, Exposure and Gamma, works best, but you should experiment with the other three methods. In the Exposure and Gamma method, make changes to the Exposure and Gamma sliders; in the Local Adaptation method make changes in the Radius and Threshold sliders. Observe the display as you change methods and settings; use the result you prefer. Click OK and Photoshop will downsample to 16 bits/channel. Save the resulting 16-bit file and then edit it as you would any other image.
Figure 11-17. Church interior, Reutte, Austria, exposed to record shadow detail. Note the loss of tonality in the windows and all bright architectural details.
Figure 11-18. Church interior, exposed to record midtone values. Note the loss of tonality on the windows and on the darkest interior areas.
Figure 11-19. Church interior, exposed to record detail in the extreme highlights. Only the clear bubbled glass windows, and their muntin separators, are visible.
To illustrate this approach, and demonstrate the ability of multiple exposures to capture high dynamic range scenes, the three captures shown in Figure 11-17, Figure 11-18, and Figure 11-19 were merged into the final HDR composite shown in Figure 11-20. The brightness range of the church interior was approximately 15 stops; far greater than the dynamic range of the sensor. Therefore, it was not possible to record shadow, midtone, and highlight information in a single capture. The capture in Figure 11-17 was exposed to record substantial shadow detail, particularly the marquetry on the back of pews; the capture in Figure 11-18 was exposed to record the predominate midtone values in the church; and the capture in Figure 11-19. was exposed to record information in the extreme highlights, including the lead muntins in the otherwise plain glass windows. The resulting HDR composite image is shown in Figure 11-20.
In Photoshop CS5, the integration of multiple exposures into a HDR 16-bit file is accomplished with the new Merge to HDR Pro function. In the File Menu, select Automate and choose Merge to HDR Pro. In the dialogue box, you can select one of four modes, selectable from the drop-down menu located at the top of the panel. The parameters applicable to the selected mode will appear in the panel. Local Adaptation offers the most control settings the greatest number of options and therefore is most likely to give you the desired rendering. But you should try each of the modes and select the one that gives you the result you prefer. In Local Adaptation mode, you can adjust overall contrast and set the black and white points by selecting the Curves tab located in the lower right corner of the dialogue box and then working with the curve.
Figure 11-20. Church interior—high dynamic range composite. Combining the three exposures as described yields an image containing all desired detail, and projecting the magical brilliance of the ornate interior.
Be aware of the fact that there are no rules here; the correct settings are those that give you the image you want. Don’t expect those settings to be the same for each image. Each one must be determined based on its own merits, which should be no surprise to you. Finally, a couple of technical points: If your capture was hand held, select the Remove Ghosts check box to compensate from any possible camera motion, and set the Mode to 16 bits to get a 16 bits/Channel file.
2. Using Two Captures and a Luminosity Selection In Photoshop
In many situations, the entire brightness range of an HDR scene can be recorded in two captures taken two or three stops apart. If the histograms of two captures fully contain detail throughout the entire brightness range (fully contained on the left; no clipping on the right), let the computer make a selection in proportion to luminosity and use the selection to paint on a layer mask. When you give the command to make a luminosity selection, Photoshop will select pixels in proportion to luminosity with white being 100 % selected, perceived middle gray being 50 % selected, and black not selected.
Prepare the two images in the RAW converter: refine the good highlight exposure to properly render the lighter values, and save it with a name that includes the term “Highlights” (Figure 11-21). Refine the darker values of the other capture and save it with a file name that includes the term “Shadows” (Figure 11-22).
Open the two images in Photoshop and arrange the two images so that both are visible, and be sure the Layers Palette is visible (if not, press F7). Next, make a copy of the Background in each of the images; do so by activating one of the two images and in the Layer Menu select Duplicate Layer, and then repeat the process for the other image. Now, rename the new Background Copy layers with the names “Highlights” and “Shadows”, respectively.
Figure 11-21. A highlight exposure of a Pacific northwest conifer forest on a sunny day. Made from a plateau in my own backyard, the highlight exposure retains the brightest highlight detail at the expense of dark shadow detail.
It’s time to move the Highlights layer to the Shadows image. To do so, select the Move Tool (v), activate the Highlights image, using the Move Tool shift-drag the Highlights image over the Shadow image, and release. The Highlights image will appear as the top layer in the Shadow image. If for some reason the Shadow appears above the Highlights layer, drag the Shadow layer down in the Layers Palette. Now that we have the two images as layers in the same photograph, save the composite as a new file with a file name that includes the word “Composite” in the title and close the Shadow file. If the captures were hand held, auto-align the layers, select all of the layers by doing a shift-click on every layer, and then from the Edit Menu select Auto-Align Layers. Photoshop will move the layers into alignment.
Now that you have both images in a single Photoshop file, the objective is to create a composite image containing shadow information from the Shadow layer and highlight information from the Highlight layer. In the steps set out in the next paragraph, you will make a luminosity selection of the Shadow layer (that is, select the brightest, and therefore overexposed, portions of the Shadow layer) and use the selection to create a mask on the Highlight layer. The luminosity mask will mask the darker (underexposed) portions of the Highlight layer, thereby allowing the good shadow detail from the Shadow layer to contribute to the composite. If this sounds confusing, try the steps below on a few images and the logic will become apparent.
Figure 11-22. The shadow exposure of the same scene, with all highlight detail clipped, but obtaining full detail in the shadows.
Here are the steps. In the Layers Palette make the Highlights layer not visible by toggling off the eye symbol. Next, make the Shadows layer active and visible, and do a luminosity selection by pressing Command-Option-2(Mac)/Control-Alt-2(PC). Photoshop will select pixels of the Shadows layer in proportion to their luminosity; that is, Photoshop will select the overexposed and brighter portions of the Shadow layer. The marching ants on the display will outline the areas of the Shadows layer having a luminosity of 50 % or more. The last step is to make the layer mask for the Highlight layer, which reveals the luminosity selection, that is, reveals the brighter portions of the Highlight layer and masks its darker portions. To make the layer mask, activate the Highlight layer and make it visible; in the Layer Menu select Layer Mask and choose Reveal Selection. Photoshop will build a layer mask that reveals the lighter portions of the Highlight layer and conceals the remainder of the layer.