Zerene VS Photoshop - a stacking experiment

Here you can discuss topics such as focus stacking, stitching and other techniques that relate to the processing of micrographs.
Post Reply
Message
Author
User avatar
Johann
Posts: 370
Joined: Mon May 23, 2016 12:17 pm

Zerene VS Photoshop - a stacking experiment

#1 Post by Johann » Thu Jul 14, 2016 5:54 pm

Hi guys, here are my findings after doing some detail comparisons between Photoshop and Zerene stacker.

I took a series of 20 photos of the eye of a Green lacewing insect. For those who are interested you can download the 20 source images here (its the full res source images, so its a 300MB download): http://www.themicrouniverse.com/20images.zip

I think this is a very good test image since the curvature of the eye presents some alignment challenges.

Zerene stacker has got 2 staking methods, called DMap and PMax.

Image 1 below:
Zerene Stacker using DMap method
Image1: Zerene Stacker - DMap
Image1: Zerene Stacker - DMap
Stacked-Zerene-Dmap.jpg (154.85 KiB) Viewed 22326 times
- Does a nice stacking job
- Does great job at aligning all the source images - notice how the dirty spots form lines in the background as each image is re-aligned - they are out of alignment because of the focus curvature at the different focus points
- The background color blending not so good


Image 2 below:
Zerene Stacker using PMax method
Image2: Zerene Stacker - PMax
Image2: Zerene Stacker - PMax
Stacked-Zerene-Pmax.jpg (142.39 KiB) Viewed 22326 times
- Does a nice stacking job
- Does great job at aligning all the source images
- The background color blending looks better
- Color and contrast seems off

Image 3 below:
Using Photoshop - Not aligned
Image 3: Using Photoshop - Not aligned
Image 3: Using Photoshop - Not aligned
Stacked-Photoshop-not-aligned.jpg (160.71 KiB) Viewed 22326 times
- Does a nice stacking job
- No alignment of source images - see how some of the eye hexagons are miss-aligned
- The background color blending looks much better


Image 4 below:
Using Photoshop - aligned images
Image 4: Using Photoshop - aligned images
Image 4: Using Photoshop - aligned images
Stacked-Photoshop-aligned.jpg (156.96 KiB) Viewed 22326 times
- Couldn't find an option in Photoshop to align images, however Zerene Stacker allows you to save the aligned images - I used these aligned images to stack in Photoshop

Image 5 below:
Final Image - Cleaned up Image 4 Using Photoshop
Image 5: Final Image - Cleaned up Image 4 Using Photoshop
Image 5: Final Image - Cleaned up Image 4 Using Photoshop
Stacked-Photoshop-cleaned.jpg (135.9 KiB) Viewed 22326 times

Conclusion:
At first glance it would seem that Photoshop did a better job than Zerene Stacker. However, Zerene Stacker has a retouching option, where you can retouch any area on your image, telling it to use a specific source image's detail. I could've fixed the background issue very simply using this.

With this retouching ability, the auto alignment of images and its ease of use - I've convinced myself that Zerene Stacker is the better option.

I must add that I was using Photoshop CS6 and and have no idea how the latest versions of Photoshop will fare.

Hope someone finds this helpful :)
Omax M837ZL
Olympus BX53
Leica MZ6
My Facebook Page
My YouTube Channel

User avatar
McConkey
Posts: 338
Joined: Fri Feb 19, 2016 4:33 am
Location: Alberta, Canada

Re: Zerene VS Photoshop - a stacking experiment

#2 Post by McConkey » Thu Jul 14, 2016 6:43 pm

Really interesting comparison! Thanks for sharing! I found it very helpful!
I'm viewing on my phone so i haven't taken a good quality look at the images but I think your cleaned up Photoshop image 4 is hard to beat. It looks perfect.
How are you stacking in Photoshop, I use CS5 but have never stacked with it. I've been looking at getting Zerene so this is a good chunk of data for me!
How do both programs handle a 20 image stack...was one faster than the other? I know with photoshop you can allocate more processing power so it can perform tasks faster. If you had used the Zerene stacker retouching option and cleaned up the background, would you still do minor re-touching in PS?

Do you know if PS or Zerene will stack RAW images or do they have to be converted first?

Thanks!
Karl
AO21 with Canon M3

User avatar
Johann
Posts: 370
Joined: Mon May 23, 2016 12:17 pm

Re: Zerene VS Photoshop - a stacking experiment

#3 Post by Johann » Thu Jul 14, 2016 7:04 pm

Hi Karl,

I did a quick comparison, to load images and stack:
In Photoshop 1 minute 47 seconds
In Zerene 1 Minute 53 seconds

So not much of a difference.

To stack in Photoshop,
1. click on File, Scripts, Load files into stack
2. Browse and Select your images you want to stack, click OK
3. All the images will be loaded as layers in one document, make sure all the layers are selected
4. Then click Edit, Auto-Blend layers...

That's it - hope it works the same in CS5.

Cleaning up the image I will definitely always do in Photoshop.

I doubt you be able to stack Raw images - and also, I think if you sharpen the images before stacking you give the stacking algorithm something better to work with.
Omax M837ZL
Olympus BX53
Leica MZ6
My Facebook Page
My YouTube Channel

User avatar
McConkey
Posts: 338
Joined: Fri Feb 19, 2016 4:33 am
Location: Alberta, Canada

Re: Zerene VS Photoshop - a stacking experiment

#4 Post by McConkey » Thu Jul 14, 2016 7:23 pm

Johann,
Oh yeah, wow no difference in time at all! That's really good to know! Some programs will have a huge processing time difference when compared to PS but Zerene can keep up I guess!

Thanks for the run down on stacking in PS, I'll give it a try!

I always wonder about doing a bit of pre-processing before stacking and its effect on the final image. I have a few PS actions I could use to batch process before stacking. Maybe I'll give it a try!
Karl
AO21 with Canon M3

User avatar
mrsonchus
Posts: 4175
Joined: Tue Feb 03, 2015 9:42 pm
Location: Cumbria, UK

Re: Zerene VS Photoshop - a stacking experiment

#5 Post by mrsonchus » Thu Jul 14, 2016 7:54 pm

Very handy indeed, thanks for the post - the selection of source-image detail retouching facility sounds reason-enough for me to consider Zerene closely...
Great and useful post - thanks again. :)
John B

User avatar
c-krebs
Posts: 201
Joined: Sat Jun 18, 2016 7:22 pm

Re: Zerene VS Photoshop - a stacking experiment

#6 Post by c-krebs » Fri Jul 15, 2016 4:17 am

Johann,

I downloaded your source files and had a look. Some thoughts...

Try Zerene again, Dmap. Set preferences... shift X on, Shift Y on, brightness on. Be certain that brightness is checked. Don't use any "auto" value for the threshold setting in preferences. When, halfway through the stack, the program shows a slider to set the threshold use about 84% for this stack. You should get a hugely improved rendition than the one you posted. I noticed in the source files there is a significant difference in brightness with some (the last few especially are much brighter. Don't know why, the exposure times are the same). When you have a soft, out of focus background it can get pretty shabby looking when there are some significant exposure variation in the frames. In these cases be sure the brightness adjustment is checked.

You've got a fair amount of sensor dust. Not too bad if you are re-touching a single image. But when run a stack and you allow x/y shifting, or magnification changes, you will get the dust "trails" (Much less so with Helicon Focus Method "B"). In this particular image it is very easy to retouch that. In Zerene, When the stack is completed, go to "Start Retouching". Select Image #1 as the "source" and simply retouch the entire background from image #1.

(BTW... if you would like I could upload the result from Helicon Focus Method "B" for comparison)

User avatar
Johann
Posts: 370
Joined: Mon May 23, 2016 12:17 pm

Re: Zerene VS Photoshop - a stacking experiment

#7 Post by Johann » Fri Jul 15, 2016 2:07 pm

Thanks Charles, it helps getting advice from the experts ;)

I tried the settings, and yeah, it does look a lot better - thanks.

If you don't mind, it'll be great to see your results from Helicon Focus.
Omax M837ZL
Olympus BX53
Leica MZ6
My Facebook Page
My YouTube Channel

User avatar
KurtM
Posts: 1749
Joined: Tue Jun 09, 2015 12:08 am
Location: League City, Texas
Contact:

Re: Zerene VS Photoshop - a stacking experiment

#8 Post by KurtM » Fri Jul 15, 2016 3:06 pm

Please post pictures, gentlemen!! There are plenty of us who are exceedingly interested in all things stacking... :shock:
Cheers,
Kurt Maurer
League City, Texas
email: ngc704(at)gmail(dot)com
https://www.flickr.com/photos/67904872@ ... 912223623/

User avatar
Johann
Posts: 370
Joined: Mon May 23, 2016 12:17 pm

Re: Zerene VS Photoshop - a stacking experiment

#9 Post by Johann » Fri Jul 15, 2016 4:03 pm

This is what i got straight out of Zerene Stacker using Charles' advice - much better :)
Attachments
Stacked-Zerene-Dmap-new-settings.jpg
Stacked-Zerene-Dmap-new-settings.jpg (143.94 KiB) Viewed 22263 times
Omax M837ZL
Olympus BX53
Leica MZ6
My Facebook Page
My YouTube Channel

User avatar
McConkey
Posts: 338
Joined: Fri Feb 19, 2016 4:33 am
Location: Alberta, Canada

Re: Zerene VS Photoshop - a stacking experiment

#10 Post by McConkey » Fri Jul 15, 2016 5:26 pm

Hopefully you don't mind me posting these results.

I've been using Picolay since i started this hobby a few months ago with decent results but after seeing what you produced with Zerene i was interested to see what Picolay would do.

This was using regular default settings and no post-processing, just re-size for forum viewing. I think i might have to invest in Zerene.
Attachments
test stackforum.jpg
test stackforum.jpg (386.65 KiB) Viewed 22257 times
Karl
AO21 with Canon M3

User avatar
gekko
Posts: 4701
Joined: Sun Oct 12, 2014 7:38 am
Location: Durham, NC, USA.

Re: Zerene VS Photoshop - a stacking experiment

#11 Post by gekko » Fri Jul 15, 2016 6:15 pm

Many thanks to all for this very interesting and useful thread. McConkey, is that dust on your camera sensor?

User avatar
mrsonchus
Posts: 4175
Joined: Tue Feb 03, 2015 9:42 pm
Location: Cumbria, UK

Re: Zerene VS Photoshop - a stacking experiment

#12 Post by mrsonchus » Fri Jul 15, 2016 6:28 pm

Looks like it - mine was exactly like that until this morning, when I cleaned it as best I could. I think 'liveview' camera modes leave the sensor very exposed perhaps, also of course I constantly move mine about between 'scopes, forget to put the covers on etc - my sympathies old chap. When I say I've cleaned my camera's sensor - I haven't made it perfectly clean by any means, just acceptably so!
To have it cleaned at a Canon 'service-center' (i.e. shop) costs in the region of £35 (GBP)! The cleaning-kit with which to clean it yourself is of a similar price! :)

p.s. thanks also for the very useful and interesting thread - it's more-or-less made me decide to invest in Zerene stacker... A very helpful topic. :D :)
John B

User avatar
McConkey
Posts: 338
Joined: Fri Feb 19, 2016 4:33 am
Location: Alberta, Canada

Re: Zerene VS Photoshop - a stacking experiment

#13 Post by McConkey » Fri Jul 15, 2016 6:36 pm

Gekko - I believe it might be dust, however the images belong to Johann and are not mine, I just posted the results from Picolay using the 20images that were provided earlier in the thread.

John - You are right, Liveview does make the sensor a lot more exposed. I find a good item to use when cleaning a camera sensor is an air blower like this one here https://www.amazon.ca/Giottos-AA1900-Ro ... B00017LSPI

You can buy them for much cheaper and it cleans the sensor chemical free and without haven't to physically touch it.
Karl
AO21 with Canon M3

User avatar
mrsonchus
Posts: 4175
Joined: Tue Feb 03, 2015 9:42 pm
Location: Cumbria, UK

Re: Zerene VS Photoshop - a stacking experiment

#14 Post by mrsonchus » Fri Jul 15, 2016 6:50 pm

McConkey wrote:Gekko - I believe it might be dust, however the images belong to Johann and are not mine, I just posted the results from Picolay using the 20images that were provided earlier in the thread.

John - You are right, Liveview does make the sensor a lot more exposed. I find a good item to use when cleaning a camera sensor is an air blower like this one here https://www.amazon.ca/Giottos-AA1900-Ro ... B00017LSPI

You can buy them for much cheaper and it cleans the sensor chemical free and without haven't to physically touch it.
Yes I've recently bought such a blower or 'rocket' as they're known! They're very good indeed, but my sensor was past the 'blow it off' stage.. :( I Used strips of Zeiss lens-cleaning wipes the width of the sensor, held in cover-slip forceps (as they have squared ends to their jaws rather than points) and dragged (dampened with lens-cleaning spray) across the sensor - one drag per piece then a new piece - about 20 times. While not perfect it removed about 80% of the 'muck' for me - another couple of such procedures and I'll probably have it clean. But for now it's a huge improvement - thankfully... :D

It's a recurring problem I find, hence my choice not to use a service center, as I'd be back there in another 6 months!

Thanks for the advice - always very useful. :)
John B

User avatar
Johann
Posts: 370
Joined: Mon May 23, 2016 12:17 pm

Re: Zerene VS Photoshop - a stacking experiment

#15 Post by Johann » Fri Jul 15, 2016 8:01 pm

My other "hobby" is off-road motorbiking - and the camera goes along everywhere - it needs a big clean :)
Omax M837ZL
Olympus BX53
Leica MZ6
My Facebook Page
My YouTube Channel

User avatar
billben74
Posts: 1020
Joined: Sat Dec 26, 2015 10:33 pm
Location: Cambridge, UK

Re: Zerene VS Photoshop - a stacking experiment

#16 Post by billben74 » Fri Jul 15, 2016 11:07 pm

Thanks all for a really informative post.
I've stopped moving my camera. I got sick of cleaning the sensor and getting dust etc on images.

ChrisR
Posts: 138
Joined: Tue Jun 14, 2016 11:01 pm
Location: Surrey, UK

Re: Zerene VS Photoshop - a stacking experiment

#17 Post by ChrisR » Fri Jul 15, 2016 11:39 pm

Yep that's dust. It's a complete pain. The effective aperture through a 'scope is always small, which means you see dust.
As an exercise, put some tubes on your camera, & black paper over the front with a pin hole, diameter around 100th the length of the tubes. With the meter on auto, wave the camera around pointing at a white ceiling or sky, and inspect the resulting image. Expect to be horrified! More realistic is to arrange f/32 or smaller. Dust will appear as black spots and lines (hairs) with diffraction rings round them.
I wouldn't dream of trying to clean a sensor without a low power microscope to look at it with, and a bright light to get in with. I always end up using a "tacky stick" of some sort - they have a soft piece of plastic on the end which picks up bits. I've never, ever, got a sensor clean properly (for f/32 and smaller) with just a wet swab. Oil is the biggest problem I've had. Took a dozen swabs.
Be careful with Sony sensors - delicate coating apparently.

Process your RAW files before putting them into a stacker. Even the stackers which will take in Raw files, don't allow you to do the tinkering with the raw files that you'd normally do, so you're better off using your favourite RAW converter. If you controlled the lighting well enough, stacking high quality jpegs is fine, Raw may win you nothing. Zerene will stack tiffs.
Hint - start using low power (therefore very fast) flash, very very well diffused.
Note that DMap ( & the Helicon equivalent) will give you colours unaltered from your source image. PMax will usually give best detail and control with overlapping stuff, so the best result comes from a combination - you retouch one from the other. With Zerene that bit's quick, usually a minute or so. On retouching it beats Helicon usually, but Helicon has some other features.
Helicon has a (old tech) Contrast method, which works very well on some particular microscope slides. ZS doesn't have that method.

Pretty much "Everything" goes through Photoshop afterwards.

User avatar
c-krebs
Posts: 201
Joined: Sat Jun 18, 2016 7:22 pm

Re: Zerene VS Photoshop - a stacking experiment

#18 Post by c-krebs » Sat Jul 16, 2016 1:16 am

Here are the result straight out of Helicon Focus. They have three methods... A, B, and C. "C" is a pyramid contrast method and gives a result that looks just like Zerene Pmax, so I did not upload it.

"A" and "B" have two settings that can vary the results considerably. Here I used the "default" settings.

Image

Image

Heribert_Cypionka
Posts: 1
Joined: Sun Jul 17, 2016 9:34 pm

Re: Zerene VS Photoshop - a stacking experiment

#19 Post by Heribert_Cypionka » Mon Jul 18, 2016 10:14 am

Dear Stackers,

I found a poor result produced by PICOLAY in this thread, and I have to comment on that ;)

Yes, one can easily produce non-satisfying results if one ignores the powerful settings. But it takes only a few seconds to find correct parameters.

1. Download the newest version of PICOLAY, I am always trying to improve it (http://www.picolay.de).

2. Look at your stack using the slide show (F12, or Image list -> Start slide show). This will show you whether you need image alignment (which is the case here). Doing this I found another issue with the 20 images: they were not correctly named for PICOLAY, which brings them into the alphabetically order. The used file names '1.jpg', 2.jpg, ,3 ... 9,10,11 ...20.jpg will end in the list as 1.jpg, 10.jpg, 11 ... 19.jpg, 2.jpg, 20.jpg, 3.jpg ... 9.jpg. This problem can be overcome by renaming the first 9 image as '01.jpg' ... '09.jpg', which I did. (Maybe I can imply the automatic renaming in a future version of PICOLAY.)

3. If you are uncertain about the stacking parameters, check the option 'Test 4 parameter sets'. This will show you how changing parameters affect your result and help to optimise them.

4. In this case it it obvious that stacking with the pre-set parameters produces a lot of noise in the unsharp background. This can easily be suppressed by setting the minimum contrast to a value of >10. Furthermore the structures we are interested in are rather big. So one could force PICOLAY to use a larger filter by setting the filter to 7 - 9 (I left it 'smart'). Trying changed parameters (+ Go for Start stacking) does not require reading of all images again. The new result appears within a few seconds!

Here my PICOLAY result using a minimum contrast of 13 and the auto-enhance function.

Image

The image was not retouched but downsized for the forum.

Thanks for reading,

Heribert Cypionka (author of PICOLAY)

User avatar
Johann
Posts: 370
Joined: Mon May 23, 2016 12:17 pm

Re: Zerene VS Photoshop - a stacking experiment

#20 Post by Johann » Mon Jul 18, 2016 11:12 am

Thanks Heribert, welcome to the forum :) - I haven't used Picolay before, but will definitely give it a try. Will be great to get help/tricks straight from the author ;)

Must be nice to understand the underlying algorithms and mathematics involved - amazing :o ;)
Omax M837ZL
Olympus BX53
Leica MZ6
My Facebook Page
My YouTube Channel

billbillt
Posts: 2895
Joined: Mon Nov 10, 2014 10:01 pm

Re: Zerene VS Photoshop - a stacking experiment

#21 Post by billbillt » Mon Jul 18, 2016 1:19 pm

What little stacking I have done has been with Picolay.. I just found it easier for me to use than CombineZ.. I know there are others out there also..

BillT

User avatar
McConkey
Posts: 338
Joined: Fri Feb 19, 2016 4:33 am
Location: Alberta, Canada

Re: Zerene VS Photoshop - a stacking experiment

#22 Post by McConkey » Wed Jul 20, 2016 11:02 pm

Hey Heribert!
Really appreciate you jumping in and giving a detailed explanation! It made a big difference! I downloaded the latest version and went back to re-stack some of my older images. Much better results.

Thank you for giving the community a nice piece of software to use!
Karl
AO21 with Canon M3

Post Reply