Merge to High Dynamic Range (HDR) and Merge to Panorama in Lightroom and Camera Raw
In the old days (before Lightroom CC/6 and ACR 9), if you wanted to do an HDR or pano merge, you had to first take your images into Photoshop. Now you can do that directly in Lightroom and Camera Raw.
Both Lightroom and Camera Raw use the same functionality and UI. The manner of accessing those features is different. However, both Preview windows behave the same way with the same options and limitations. There are some limitations: the preview is limited to 2048 pixels in the long dimension if you’re using a Retina-type, high-resolution display or 1024 pixels when using a normal display. At this point the engineers have said this is due to performance issues, and you can’t zoom into the preview. Hopefully, that will change in the future. It should also be noted that the UI functionality offers very little control compared to third-party applications dedicated to HDR and pano merging. However, the results are actually very good.
HDR Merging in Camera Raw
To do an HDR merge in Camera Raw, you must first select multiple images in Bridge and open the images in the Filmstrip mode in Camera Raw. Once in Camera Raw, select the images you want to merge and use the flyout menu at the top of the filmstrip, as shown in Figure 3.81. You’ll see that the “most selected image” is at the bottom. The merge function produces a new DNG based on the most selected image’s filename and appends a –HDR to the filename. If you do multiple HDR merges, the later merges receive a number to avoid overwriting the previous –HDR file. The most selected image is also the one used to propagate EXIF metadata.
FIGURE 3.81 Flyout menu in Camera Raw.
Once you select the option, a preview window comes up allowing a selection of various functions. Figure 3.82 shows the HDR Merge Preview window.
FIGURE 3.82 The HDR Merge Preview window for Camera Raw.
There are various options to choose from:
- Auto Align attempts to align images prior to merging. I use this all the time, even if I’ve shot the bracket on a tripod. If you handhold the bracket, you really should use this option.
- Auto Tone applies the default auto-exposure corrections for the image. This can be useful, but since the settings are parametric, you can always change them later.
- Deghost Amount uses an algorithm to determine which of the bracketed shots to use if there is any movement (ghosts) between exposures. In this image of redwood trees in a forest, the breeze moved some of the branches. There was also a problem with a slow shutter speed, which introduced some motion blur in the longest exposure.
- Show Deghost Overlay shows a preview of where the deghosting will occur.
Figure 3.83 shows the various deghosting options and the Deghost Overlay for each option. In the first figure, I’ve selected the option to use Auto Tone to lighten the image for better previewing.
FIGURE 3.83 The Deghosting Amount options and overlay previews.
How well did Camera Raw do with the deghosting? I’ll show you in a moment, but first I wanted to cover some of the details of using Camera Raw’s DNG HDR. First, the resulting file is a Linear DNG, which means it’s not a completely raw file; it’s what I call a half-baked raw file. The image has been demosaiced, but it’s still a linear gamma (1.0). Next, the resulting DNG file is stored as a 16-bit floating-point image, but the processing applied is done in 32-bit floating-point. Don’t confuse 16-bit floating-point to 16-bit integer images; it’s still a floating-point image. You can use all the processing tools in Camera Raw to adjust the image. However, one thing to note is instead of the normal Exposure range of +– 4 stops, the HDR range is expanded to +– 10 stops. The other adjustments remain the same.
Also, when doing an HDR merge, Camera Raw ignores all the tone adjustments you may have already made, as well as any local adjustments, including spot healing. So, you may as well save your time by working on the HDR image after merging and not bother working on your original raw files. The adjustments you can (and should) make are lens corrections, image sharpening, and noise reduction. Applying these before the merge will aid in the production of an optimal merge result.
Okay, back to the results. I’ve zoomed way into the four HDR DNG files I produced. They show extreme detail of an area in the upper right of the trees. Figure 3.84 shows the results.
FIGURE 3.84 Comparing the Deghosting Amount options.
As you can see, merging images that contain movement can lead to anomalies in the merged result. In this case, the only option that did a decent, but not 100 percent perfect job, was the High option. There is still a tiny hint of the anomalies, but depending on the size you need to use the image, they would likely be invisible. Remember, we’re pixel-peeping here. If Camera Raw’s deghosting isn’t up to your standards, you’ll need to resort to a special purpose application, such as Photomatix Pro (www.hdrsoft.com), which includes a Lightroom plug-in for improved workflow. You still need to render and export the images in the plug-in and return them to Lightroom as 32-bit TIFF files, but working on them in Lightroom works well.
Pano Merging in Lightroom
The Lightroom HDR and Panorama functions are found in either the Library or Develop modules in the Photo menu under Photo Merge. Select multiple images in the Library or the filmstrip in Develop and choose Panorama. Figure 3.85 shows the images selected and the flyout menu from Photo Merge. I used eight shots of the city center in Florence, Italy, which I took with my Sony DSC RX 100M2 22 MP camera. I brought that point-and-shoot camera with me when I went to Italy with my wife so I could stay married (she doesn’t like it when I drag lots of cameras and lenses along with me).
FIGURE 3.85 Selecting images for merging.
The Panorama Merge Preview window offers various options, including the option to Auto Select Projection, three projection options, and an Auto Crop checkbox. I tend not to use the Auto Select Projection because I prefer to see the results and choose for myself. The three projections are
- Spherical: This projection treats the pano merge like it was projected on the inside of a sphere. It does not do any corrections to perspective.
- Cylindrical: Instead of a sphere, the projection is based on a cylinder. This projection also does not do perspective correction and usually produces a taller merge than Spherical.
- Perspective: This projection uses the center image to determine the correct perspective correction and merges the other images to match the perspective.
Which projection you use is really dictated by the subject and what you are trying to accomplish. I generally use Cylindrical for landscape pano merges because I don’t care about trying to correct for perspectives out in nature. If you have buildings in your images, Perspective is a better projection choice. You do tend to lose a lot of the image because of the way the Perspective projection accomplishes the corrections for the merge. Figure 3.86 on the previous page shows the three projection options.
FIGURE 3.86 Comparing the Projection options.
You can see that the Spherical and Cylindrical projections are very similar. They both exhibit problems with distortions, but the merge looks okay. The Perspective Projection has the typical “bow tie” effect, but the buildings are corrected. Figure 3.87 shows the result of selecting the Auto Crop option. Remember that this is a parametric adjustment so you can always change the crop in the Develop module.
FIGURE 3.87 Auto Crop turned on.
The merge functions in Lightroom have the same preview limitations as Camera Raw: 2048 pixels for Retina displays and 1024 pixels for normal displays. The Merge to Panorama does honor and use all global adjustments but not local adjustments, including spot healing, which is a bummer. However, I got an indication from an engineer that not keeping spot healing spots was an oversight, not intentional. So there’s hope that in the future spot healing will be honored (yippee). With the HDR merge I recommended applying lens correction before the merge. With Merge to Panorama there’s no need because the lens corrections will automatically be applied as long as you have them. If you don’t have a lens profile, you can apply manual corrections.
There are some other limitations with Merge to Panorama, primarily relating to the size of the resulting merged image. The maximum pixel dimension on the longest side is 64,000 pixels and a maximum file size of 512 MP. If you are making really long pano merges, you may hit the limit.
There’s another limitation, generally only encountered when using the Perspective projection. You may get an error Unable to merge the photos. Please try a different Projection option. When this happens, you’ll usually be able to use either the Spherical or Cylindrical projections. However, there can be times when none of the projections will work. This is usually caused by not leaving enough image overlap between images. I generally overlap between to ½ to ensure enough image area for the algorithms to work with. You also cannot manually arrange the order of the images. If you do a lot of high-end and resolution panoramic imaging, you might want to look into a third-party application called PTGui Pro (www.ptgui.com). I have a copy and resort to using it if I’m having problems in Lightroom/ACR or Photoshop.
Can you shoot multi-row panoramic images and merge them? You bet, but the criteria for providing enough image overlap becomes even more important. What about merging HDR into a panorama? Yep, that works. You first need to build the HDR DNGs, and then use Merge to Panorama to merge the HDR images. Can you do a focus merge? Not yet, but the engineers have said it’s on their to-do list.
Adding both Merge to HDR and Merge to Panorama are arguably the stars of the new features of Lightroom CC/6 and ACR 9. Being able to maintain parametric editing on semi-raw files is a benefit.