Couple of extreme closeups of a leaf and a petal

Images taken in a controlled environment or with a posed subject. All subject types.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

lauriek
Posts: 2402
Joined: Sun Nov 25, 2007 6:57 am
Location: South East UK
Contact:

Couple of extreme closeups of a leaf and a petal

Post by lauriek »

Firstly a close up on a bamboo leaf shot with the Olympus 20mm 3.5 bellows lens on OM bellows at maximum extension...

Image

This was backlit from one of the flash heads on my macro twinflash. I generally find I have to increase the contrast by adjusting the levels quite harshly in RAW processing when I shoot with this kind of lighting).

Secondly an extreme closeup of a Daffodil petal, shot with an extremely cheap 40x microscope objective (came as part of my second hand Brunel compound scope, grand total cost of whole scope inc p&p around $80, so I figure this objective owes me about $5!). Working distance - literally nil. Pure backlighting.

Image

Again, pretty agressive levels tweaks in RAW processing to enhance contrast and bring out a little detail. This is a single shot, I've had no luck stacking these shots so far, I think I will have to try a completely manual stack but not enough time for now!!

Any idea what we are looking at here, are these cells in the plant? I've noticed with the 20mm lens, when looking at pretty much any plant matter at decent magnification, that it always appears to be made up of these little blobs, but I don't know what they are, I do know they seem to cause major problems to both Helicon and CombineZM...

Edited to add:

First shot is a stack of around 6 shots.

With both these subjects I mounted them onto microscope slides, but I have no cover slips so for the first shot I mounted the subject between two slides.. For the second shot the working distance of that objective is less than the thickness of a slide, I could not focus on the subject with a slide above it, so I left the subject mounted on a slide with no cover (sort of clipped in place).

rjlittlefield
Site Admin
Posts: 23608
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Re: Couple of extreme closeups of a leaf and a petal

Post by rjlittlefield »

lauriek wrote:Any idea what we are looking at here, are these cells in the plant? I've noticed with the 20mm lens, when looking at pretty much any plant matter at decent magnification, that it always appears to be made up of these little blobs, but I don't know what they are, I do know they seem to cause major problems to both Helicon and CombineZM...
Yes, cells. Your 20 mm lens on maximum extension is quoted as 13.5X magnification, so with your Olympus DSLR, you have a field width of only about 1.33 mm, about what you'd see through a 100X microscope.

At these magnifications, you have very little depth of field and your focus step has to be very small or frame-to-frame registration will get messed up. (With a 40X objective, your nominal DOF is less than 2 microns -- 0.002 mm!)

Also whenever you're backlighting, you'll end up with detail at the same place in multiple frames, so you have to use a stacking method that tolerates that. Helicon's Method A should work OK, or CombineZM's Do Weighted Average.

What sort of difficulties are you running into?

--Rik

lauriek
Posts: 2402
Joined: Sun Nov 25, 2007 6:57 am
Location: South East UK
Contact:

Post by lauriek »

Thanks Rik - I've only just started trying to stack purely backlit subjects, and was not aware that was what weighted average was for (My demo of Helicon has expired now and to be honest I don't have the cash available for it at the moment so I'll be making do with CZM for a month or two until I have the cash available for HF...

I will have another go tonight with one of these using Weighted average. Of course it's entirely possible my setup does not allow fine enough focus steps through the subject so I might still run into problems.

I'd worked out the problem was overlapping detail due to focussing 'inside' the subject but not how to get around it! Certainly CZM barfs badly if you try a normal stack with this kind of subject!! ;)

I wondered what WA was for when I started using CZM, tried it on a normal (non transparent, frontlit) subject, and of course it came out rubbish...

BTW the first shot above was hand stacked as the stack was so small...

rjlittlefield
Site Admin
Posts: 23608
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

lauriek wrote:I will have another go tonight with one of these using Weighted average. Of course it's entirely possible my setup does not allow fine enough focus steps through the subject so I might still run into problems.

I'd worked out the problem was overlapping detail due to focussing 'inside' the subject but not how to get around it! Certainly CZM barfs badly if you try a normal stack with this kind of subject!! ;)

I wondered what WA was for when I started using CZM, tried it on a normal (non transparent, frontlit) subject, and of course it came out rubbish...
Be sure to download a fresh copy of CZM, version "15th of March 2008" or later.

I discovered a couple of weeks ago that the auto-alignment function of CZM had broken sometime in the last year. It was adjusting x and y but not scale. That significantly degraded the quality of some images, particularly at moderate magnifications with short lenses where there would be lots of scale change between successive frames. Alan was able to fix the problem very quickly after I reported it, but I know that bad versions were posted for at least 6 months so no doubt quite a few bad copies got downloaded.

I suspect that effects of the alignment problem could have been a lot worse with Do Weighted Average than with Do Stack, since Do Weighted Average is specifically in the business of averaging content across possibly numerous frames while Do Stack only averages between adjacent frames. (At least, that's my understanding.)

Anyway, Do Weighted Average is significantly better than Do Stack even for some normal (non transparent, frontlit) subjects, notably when there are strong overlapping fine features such as bristles. Take a look at this comparison of methods for an illustration. CZM's Do Weighted Average and Helicon's Method A (which used to be its only method) give similar results.

The other thing to notice is that CZM now has a new macro named Do Soft Stack, which is supposed to work better with images that don't really have sharp detail. I have no experience with it, however.

--Rik

lauriek
Posts: 2402
Joined: Sun Nov 25, 2007 6:57 am
Location: South East UK
Contact:

Post by lauriek »

Aha, I will go download the latest version asap, I'm currently running on a release from December, that might explain why I've been having so many problems!! (although I appreciate what stacking sw is doing is so dang complex it's likely never going to be /perfect/!)

rjlittlefield
Site Admin
Posts: 23608
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

Laurie, I am sorry to report that I was wrong about the version ID of the corrected CombineZM.

The March 15 version was still buggy.

It is the "24th of March 2008" version that is corrected. And it has not yet been placed in the usual place for distribution. Instead it is posted as an interim release at http://hadleyweb.pwp.blueyonder.co.uk/CombineZM.exe .

See http://tech.groups.yahoo.com/group/combinez/message/652 for Alan's note.

--Rik

Post Reply Previous topicNext topic