Panasonic Lumix cameras & Post Focus
Moderators: rjlittlefield, ChrisR, Chris S., Pau
Re: Panasonic Lumix cameras & Post Focus
I have chosen post focus because it allows me to achieve good shots, quickly in the field. It does come at a cost; the quality is not as good as using a macro rail. Post processing is even more important to maximise image quality.
Depending on the subject, I use Helicon Focus and Zerene to image stack. I still occasionally have softness in a stacked image.
I have also noticed that post focus does not always record the video from front to back, which may also impact image quality. I cannot change that because the process is automated.
Recently I have used the Stablise mode in Topaz Sharpen AI with good results. Using this method I can sharpen the image and improve IQ in one step. It has been effective is removing some softness.
Depending on the subject, I use Helicon Focus and Zerene to image stack. I still occasionally have softness in a stacked image.
I have also noticed that post focus does not always record the video from front to back, which may also impact image quality. I cannot change that because the process is automated.
Recently I have used the Stablise mode in Topaz Sharpen AI with good results. Using this method I can sharpen the image and improve IQ in one step. It has been effective is removing some softness.
- rjlittlefield
- Site Admin
- Posts: 23596
- Joined: Tue Aug 01, 2006 8:34 am
- Location: Richland, Washington State, USA
- Contact:
Re: Panasonic Lumix cameras & Post Focus
When you're using Zerene Stacker, there's a fairly convenient trick that often works around this issue.
The trick is this: after you've processed the stack at least once, so that all the images have been aligned against each other, do a File > "Re-order input files" > "Sort by Scale". If the files were originally not in strict focus order, chances are they will be in strict focus order after the sort is done. Then you can do another Stack operation to take advantage of the improved ordering. This will not matter with PMax, but it can make big improvements with DMap.
To understand how the trick works, first note that with most optics, the image changes size along with focus. Depending on the lens, the image might get smaller or larger as focus moves back, but it's probably only going to do one of those, and it will do it consistently and by a large enough amount to be reliably determined by the computational alignment process. Fortunately, the alignment process can usually determine scale accurately even when the images are initially out of order, and then sorting by the recovered scale puts them back into order again.
I see this is not addressed by the FAQs at zerenesystems.com, so I'll go fix that oversight.
The nice thing about "Sort by Scale" is that it doesn't require much thinking or mouse-clicking. The alternative is to figure out by eye what the focus order was, then manually add files into the stacking software one at a time or in appropriate groups to get the correct ordering. I've also seen people rename their image files so as to be in focus order.
--Rik
Re: Panasonic Lumix cameras & Post Focus
Thanks for the tip Rik. I will bear that in mind next time I see it happen.
Re: Panasonic Lumix cameras & Post Focus
Short story
My Post Focus workflow has been revised. I now use Method A in Helicon Focus to stack the frames extracted from the 6K video, then I put the stacked image through Topaz Sharpen AI, giving even better results than before.
Long story
When version 2 of the Panasonic G9 firmware was released, the Post Focus feature did not take the 6K video at focus points from front to back. The focus points were mixed up, so when the frames were extracted they were not necessarily taken from from to back. Helicon Focus would protest at the frames being out of order.
Changing from Method C to Method A in Helicon helped greatly, but using Method A gave a softer stacked image. However, around that time Topaz Sharpened AI was improved and running the stacked image through Sharpen AI with images stacked using Method A gave excellent results. Better than the earlier methodology.
As my images are taken in field, even a gentle breeze can impact the frames in the video. So Sharpen AI has helped greatly.
Helicon Focus use to extract 8 bit JPG frames from the 6K video, but in version 7.7.0 the program now extracts 10 or 12 bit TIFs to stack. This has improved the quality even further.
You will need a fast graphic card so the PC is not slowed to a crawl. My RTX4000 workstation card works really well with Helicon and Sharpen AI, using OpenCL.
Zerene Stacker runs faster as well (I use both Zerene and Helicon).
My Post Focus workflow has been revised. I now use Method A in Helicon Focus to stack the frames extracted from the 6K video, then I put the stacked image through Topaz Sharpen AI, giving even better results than before.
Long story
When version 2 of the Panasonic G9 firmware was released, the Post Focus feature did not take the 6K video at focus points from front to back. The focus points were mixed up, so when the frames were extracted they were not necessarily taken from from to back. Helicon Focus would protest at the frames being out of order.
Changing from Method C to Method A in Helicon helped greatly, but using Method A gave a softer stacked image. However, around that time Topaz Sharpened AI was improved and running the stacked image through Sharpen AI with images stacked using Method A gave excellent results. Better than the earlier methodology.
As my images are taken in field, even a gentle breeze can impact the frames in the video. So Sharpen AI has helped greatly.
Helicon Focus use to extract 8 bit JPG frames from the 6K video, but in version 7.7.0 the program now extracts 10 or 12 bit TIFs to stack. This has improved the quality even further.
You will need a fast graphic card so the PC is not slowed to a crawl. My RTX4000 workstation card works really well with Helicon and Sharpen AI, using OpenCL.
Zerene Stacker runs faster as well (I use both Zerene and Helicon).
Re: Panasonic Lumix cameras & Post Focus
It looks like the new Panasonic GH6 does not have the Post Focus feature of the GH5 and G9 cameras.
I am seeking confirmation, but the dial on the top left of camera no longer has the macro/post focus image on it, it has been replaced by a high resolution mode.
I am seeking confirmation, but the dial on the top left of camera no longer has the macro/post focus image on it, it has been replaced by a high resolution mode.
Re: Panasonic Lumix cameras & Post Focus
it is confirmed, the GH6 does NOT have Post Focus.
Re:
Not that anybody likely cares, but this one is Pyrausta aurata ... also known as Mint Moth.Cactusdave wrote: ↑Thu Aug 30, 2018 7:51 am
Details: Small moth, F5.0 1/320th exposure. Showing off its proboscis very nicely.
Last edited by JKT on Tue Mar 08, 2022 2:46 pm, edited 1 time in total.
Re: Panasonic Lumix cameras & Post Focus
That is one colourful moth.
Re: Panasonic Lumix cameras & Post Focus
In the starting message of this topic you wrote that you expect to receive 200 Mpix images stacking in Helicon from high-res sources.
(I assume meg are megapixels)
Please, could you clarify how is that possible, if the source high-res images of G9 are 80 Mpix?
Is there a function of increasing the resolution in Helicon that I don't know about?
Re: Panasonic Lumix cameras & Post Focus
I have not used the G9's high resolution mode, because when I focus stack from a 6K post focus video file, the final stack gives me a tif file normally over 100 megs.
Initially Helicon extracted jpgs from the post focus video file, but more recent versions extract tif files (you can now choose jpg or tif and I use tif). From memory the stacked jpgs gave a final tif file around 90 megs, when stacking using extracted tif files the resulting stack is around 100-110 megs. The exact file size depends on how many tif files that are stacked.
If I created a stacks using the high megapixels mode, each file would be around 80 megs, stacking multiple 80 megs files would give a me a larger stacked file. For some reason the final Helicon stack always seems larger than a single source file.
Initially Helicon extracted jpgs from the post focus video file, but more recent versions extract tif files (you can now choose jpg or tif and I use tif). From memory the stacked jpgs gave a final tif file around 90 megs, when stacking using extracted tif files the resulting stack is around 100-110 megs. The exact file size depends on how many tif files that are stacked.
If I created a stacks using the high megapixels mode, each file would be around 80 megs, stacking multiple 80 megs files would give a me a larger stacked file. For some reason the final Helicon stack always seems larger than a single source file.