How to shoot sequential images and then merge them together in the computer to create a panoramic view of a scene
Nigel recently gave his first online talk of 2022, and on this occasion he gave the audience a guide to panoramic photography. This is the technique associated with the creation of a large rectangular image from several standard images, often with a wider view than is possible in any single image.
He gave an overview of both the in-camera photographic techniques, and the post-photography in-computer skills needed to create beautiful panoramic images. This talk is now available any time on You Tube, and can be watched here. Just click on the link below.
Step 1: Creating a sequence of images
As a guide to panoramic photography the first step towards success is to shoot a series of overlapping image that pan across a scene. To do this follow this list of do’s and don’ts:
- Shoot with the camera in the vertical/portrait orientation. This gives a wider (ie from top to bottom) image that can help if any later cropping is needed.
- Mount the camera on a tripod. Although the sequence of shots can be done with the camera hand-held, you are much more likely to be able to maintain a level horizon throughout the sequence if the camera is on the tripod.
- Make sure both tripod and camera head are completely level. If they are not, then as the camera rotates to shoot the sequence it will start to tilt, putting the horizon out of line.
- Shoot the sequence with a lens no wider than 35mm in focal length (on a full-frame camera). Using a lens with a shorter focal length than this can result in distortion in the image corners, making it difficult for the panoramic software to subsequently merge the images.
- If shooting a view where there is no foreground with elements less than about 30 metres from the camera, it is fine to rotate around the camera body.
- However, if shooting a view with a foreground and nearby elements, to avoid any parallax changes as the camera rotates it is better to rotate the camera around the lens’s nodal point, not the camera body. The nodal point is the lens’s optical centre, not necessarily the physical centre.
- To achieve this, have the camera mounted on a sliding focussing mount, coupled with a vertical locking plate that allows the camera to sit above the focussing rail in the vertical position.
- Slide the camera backwards on the rail until there are no parallax changes as the camera rotates: the relative positions of all foreground elements remain the same.
- Once the camera is set up, focus on the scene and then turn the autofocus off, ensuring that the focus will not change as the camera goes through its sequence.
- Meter the exposure in aperture priority, pointing the camera at something that is towards the middle of the view. Then switch to manual exposure and set the exposue manually. This will ensure that the exposure will be the same for all images in the sequence.
- If shooting with a slow shutter speed make sure to use either a remote shutter trigger or a two-second time delay, and (if using a DSLR) switch the mirror-up facility on. These steps ensure there will be no vibration to blur the images.
- Go through a couple of practice sweeps to make sure everything lines up nicely, and that there are no nasty surprises creeping into the view.
- Then go through the shooting sequence from left to right or right to left. Make sure to leave about 30% overlap in the view between successive images. This ensures that there is plenty of duplicate data that the panoramic software can use to accurately sequence and line up the images.
- Most sequences shot with the camera in the vertical/portrait orientation will need 7-8 images to make a complete view, but fewer or more may be needed for some views.
And that’s the in-camera work done. Now we move on to the in-computer work.
Step 2: Merging the images in-computer
Continuing with our guide to panoramic photography, having shot the sequence of images we now need to merge them together in-computer to create the final panoramic.
If you shot the images as Raw files you’ll first need to convert them to the Tiff format. In doing this, any changes you make to colour temperature, saturation, contrast etc will need to be applied to all the images equally.
Having created the Tiff files you can then load them into the panoramic software. In this article, I’m assuming the use of Adobe Photoshop. Follow these steps:
- Access the panoramic tool via the File>Automate>Photomerge commands.
- In the dialogue box that opens load the files using the Browse option.
- Choose the type of image merge method you need from the list on the left. For panoramic photography it is as well to stick with ‘Auto’, the default option.
- Of the four boxes at the bottom, tick those whose actions you need. ‘Blend Images Together’ must be ticked to ensure the images are correctly blended to together. Tick ‘Vignette Removal’ if there is any shading in the corners of your images. I generally leave ‘Geometric Distortion Correction’ unticked as it often results in a curving horizon rather than the nice straight one I have in my original images. ‘Content Aware Fill Transparent Areas’ is a useful tool that you might to tick. Usually, the merged images don’t marry up to create a perfectly rectangular panoramic image. Instead, the outer edges curve or slope, leaving blank/transparent areas around the edges. Ticking the option to have these areas filled can be highly effective. Leaving it unticked will make it necessary to crop the final panoramic image to create the correct rectangle.
- Once ready, hit ‘OK’ and sit back.
Depending on how much computing power you have, it may take some time to process the images to produce the panoramic. When finished, you’ll have a panoramic image with all its component images arranged in separate layers. If you’re happy with the result then flatten the layers into a single layer by going to Layer>Flatten Image.
You may then need to crop the panoramic to remove blank/transparent outer areas. Finally, check the whole panoramic for any bits of sensor dust that might be present and repeated across the panoramic. Also look closely for any artefacts due to imperfect image merging.
Finally, name and save your file. You now have your completed panoramic image.
Hopefully, you’ve enjoyed this guide to panoramic photography.
Our next online talk will be about underwater photography, on 22nd June. All our talks are free to attend – just register to be sent the Zoom link below.
You can also watch past talks. Just click on the link below.