Cloning from one image to another
Posted: October 6th, 2019, 6:46 pm
Not providing a tip but asking for input on a better method.
I do macro photography, with a 150mm 1:1 macro lens. In an effort to keep extremely narrow depth of focus overall in the image, I typically shoot at the maximum aperture (in this case, although it is a 2.8 lens, at 1:1 the max is 5.6). Currently I am doing a bunch of flowers with a single water drop.
Because the dof is so narrow, the entire depth of the drop cannot be captured sharp in a single shot, and usually requires at least two, if not more, depending on the intended final image.
As it is, I shoot one with the edge of the drop sharp, one with the window/light source on the drop sharp, and often a third with the background, which is shown in the drop, sharp. I have a homemade stand/support/table with an attached tripod head (no tripod) and flexible arm and clamp for the subject. I hand-focus by rotating the focus collar on the lens and do not use a stacking program (either in-camera or in post). The result is three or more, differently focused, images composited/cloned together.
The post processing method I use is as follows:
- determine my base image - usually the image with the edges of the drop in focus.
- use Composite, with a mask around the drop in the #2 image (clone source).
- use 1 point (and zoomed in considerably) and 50% opacity, I move the two images so the drops line up in each image.
- once satisfied, I change to 100% opacity and "Ok" and then save the resultant image. Usually the border around the masked area (full mask no feathering) is obvious but that gets fixed in the next step.
- Using "Clone" I then clone what I need from that resultant image, into the base image, using a high transparency and softness to blend.
- that result is saved, then the process is repeated (using the newest image as the base) for each subsequent blend.
I should note, the reason I have to initially Composite (with 1 point) is when each image is made with different focus planes, the drop actually mis-aligns slightly from one image to another, so the objects in a straight clone would be out of alignment slightly.
So, my question is more about the Composite aligning process. At 50% opacity, the ghost of both images can be seen, however, because one image is blurred around the edges of the drop and the other is blurred in the centre of the drop, it is difficult to exactly align one over the other. I make multiple movements and between each, I flip the opacity back and forth between 0% and 100% to see if they line up. As I have multiple compositions and each has at least three images to composite/clone, it gets to be quite long and tedious.
Can anyone suggest a more efficient/accurate method to accomplish what I am doing, either just on the composite/lining up, or anywhere in the process for that matter.
I am attaching a finished image (a four-image composite) to give you an idea of what I mean.
I do macro photography, with a 150mm 1:1 macro lens. In an effort to keep extremely narrow depth of focus overall in the image, I typically shoot at the maximum aperture (in this case, although it is a 2.8 lens, at 1:1 the max is 5.6). Currently I am doing a bunch of flowers with a single water drop.
Because the dof is so narrow, the entire depth of the drop cannot be captured sharp in a single shot, and usually requires at least two, if not more, depending on the intended final image.
As it is, I shoot one with the edge of the drop sharp, one with the window/light source on the drop sharp, and often a third with the background, which is shown in the drop, sharp. I have a homemade stand/support/table with an attached tripod head (no tripod) and flexible arm and clamp for the subject. I hand-focus by rotating the focus collar on the lens and do not use a stacking program (either in-camera or in post). The result is three or more, differently focused, images composited/cloned together.
The post processing method I use is as follows:
- determine my base image - usually the image with the edges of the drop in focus.
- use Composite, with a mask around the drop in the #2 image (clone source).
- use 1 point (and zoomed in considerably) and 50% opacity, I move the two images so the drops line up in each image.
- once satisfied, I change to 100% opacity and "Ok" and then save the resultant image. Usually the border around the masked area (full mask no feathering) is obvious but that gets fixed in the next step.
- Using "Clone" I then clone what I need from that resultant image, into the base image, using a high transparency and softness to blend.
- that result is saved, then the process is repeated (using the newest image as the base) for each subsequent blend.
I should note, the reason I have to initially Composite (with 1 point) is when each image is made with different focus planes, the drop actually mis-aligns slightly from one image to another, so the objects in a straight clone would be out of alignment slightly.
So, my question is more about the Composite aligning process. At 50% opacity, the ghost of both images can be seen, however, because one image is blurred around the edges of the drop and the other is blurred in the centre of the drop, it is difficult to exactly align one over the other. I make multiple movements and between each, I flip the opacity back and forth between 0% and 100% to see if they line up. As I have multiple compositions and each has at least three images to composite/clone, it gets to be quite long and tedious.
Can anyone suggest a more efficient/accurate method to accomplish what I am doing, either just on the composite/lining up, or anywhere in the process for that matter.
I am attaching a finished image (a four-image composite) to give you an idea of what I mean.