As we are using a DSLR which comes with an “only” 35mm full frame sensor, we have to do some serious post production in order to stitch together, blend and adjust the single images that make the large format final image. The amount of professional software suitable for stitching and merging of single images is large. As the demands in terms of merging are relatively moderate because of all frames laying in the same optical layer no rotations or tilting movements have to be compensated.
Therefore the Photoshop function “File – Automate – Photomerge” is just doing the job. With all single frames being loaded, just “position only” and “merge layers” have to be selected.
Photoshop is now taking over and performing the alignment and color blending operations, in order to have smooth transitions between the single frames. At this point it becomes clear again why all shots of one series have to be taken with the camera in “fully manual” mode.
Only fixed exposure values guarantee the optimal result for the merged and blended final image. An inconsistent set of images would force the algorithms to apply corrections, that will most likely reduce the overall image quality.
Taking the present example the final image is a 46 Megapixel image, consisting of 11 single vertical frames, where not even the maximum possible resolution is achieved here. Only the additionally saved .jpg images were used.
Each of these .jpg files has a resolution of 2496 x 1664 pixels, whereas the post processed RAW files have 4368 x 2912 pixels.
Here, we see the final result with only minor corrections like color adjustment, level-correction and moderate sharpening applied.