Sticky William (handheld field stack)

Just bought that first macro lens? Post here to get helpful feedback and answers to any questions you might have.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

ChrisLilley
Posts: 674
Joined: Sat May 01, 2010 6:12 am
Location: Nice, France (I'm British)

Sticky William (handheld field stack)

Post by ChrisLilley »

Image

Short field stack of 8 handheld images, taken in 'macro wind' in the Var estuary, France. Plant is "Sticky Willy" (Galium aparine). Result is not at all good, but I felt I should post something.

I was using a Nikon D90, Micro-Nikkor 105mm f/4 AIS plus a Marumi +5 diopter achromat close-up lens. By itself the Micro-Nikkor gets to 1:2, with the diopter it gets a bit better than 1:1 but with a shorter working distance.

Raw NEF images developed in Nikon Capture NX2. Stacked in Zerene Stacker, PMax method, exported as 16-bit TIFF. Some fairly obvious artifacts due to insufficient shots to cover the depth of the subject. ZS did well considering the plant was moving in the wind, 5% or more of the total field of view between shots. The final image is cropped to about 80% to get rid of border artifacts and finished with a slight levels tweak, reduced to 2k resolution and an unsharp mask (on the L channel, in Lab) then converted to sRGB and saved as JPEG.
Last edited by ChrisLilley on Fri Jun 04, 2010 10:12 am, edited 1 time in total.

rjlittlefield
Site Admin
Posts: 23562
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

Chris, this is looking good. I see a couple of foreground leaves OOF, and some motion echos at lower right, but the bulk of the image appears clean. Are you seeing some other artifacts at higher resolution that I am missing here on the web? About the motion echos, I suggest retouching from one or two of the source frames. A little bit of blur on the transitions would be less noticeable than the echos.

--Rik

ChrisLilley
Posts: 674
Joined: Sat May 01, 2010 6:12 am
Location: Nice, France (I'm British)

Post by ChrisLilley »

rjlittlefield wrote:Chris, this is looking good. I see a couple of foreground leaves OOF, and some motion echos at lower right, but the bulk of the image appears clean. Are you seeing some other artifacts at higher resolution that I am missing here on the web?
The image is posted here at 1k, but links to the flickr page where the 'original' size (in their terminology) is a 2k image.

Motion echoes are the parallel wavy lines to the upper right of one of the spines on the step, lower right of the picture?

Given how the plant was blowing about in the wind I was surprised the stack came out as well as it did. I guess the rock background is only really in focus in one shot, which probably helps.

On the larger image, looking at the leaf which points directly out at the viewer, there is a horizontal line artifact which I assume is due to the DOF of two adjacent slices not actually overlapping.
rjlittlefield wrote:About the motion echos, I suggest retouching from one or two of the source frames. A little bit of blur on the transitions would be less noticeable than the echos.
--Rik
Thanks, I will try that. I have not investigated the retouching feature yet.

rjlittlefield
Site Admin
Posts: 23562
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

ChrisLilley wrote:The image is posted here at 1k, but links to the flickr page where the 'original' size (in their terminology) is a 2k image.
Thanks -- on first reading I missed that the picture was also a link. My browser (Firefox) gives no visual indication when an image is linked, and in my sleepy state I didn't think to try hovering the cursor.
Motion echoes are the parallel wavy lines to the upper right of one of the spines on the step, lower right of the picture?
I guess you're talking about what appears to be the attached fiber at x=1739,y=1153. I think that's an example too, but what I was really looking at were lots of features on the leaf around x=1862,y=1034. Right at those coordinates there is a bright streak at the edge of some crystal. That edge appears duplicated. So do all the edges of the crystal next to it, and indeed all other features on that same leaf including the big spine at the end that hits the edge of the frame.

It's possible that the duplication occurs entirely in one source frame, but for stacked images it's more likely that the leaf is about equally clear in two frames, just in different positions. When that happens, stacking algorithms tend to preserve both sets of details, leading to what I'm calling "motion echos". I don't think that's a standard term, just the best short description I could think of.
Given how the plant was blowing about in the wind I was surprised the stack came out as well as it did. I guess the rock background is only really in focus in one shot, which probably helps.
It does help, but mostly I think you got lucky about how the image composition interacted with the alignment algorithm. ZS does not rely on sharply focused features for alignment. Aligning based on discrete features works well for pano stitching, but not so well for focus stacking where by design each frame contains different sharply focused features. Instead, ZS uses a different approach based on correlation of intensities for all pixels. This gives more weight to places where there is high contrast. You have the good fortune to have a couple of large dark shapes with high contrast internal detail superimposed on a pretty vague light background. This probably caused the alignment algorithm to pay lots of attention to what we consider the subject. Unfortunately the alignment algorithm has no clue about what's important to us humans. If the rock had been strongly patterned, then it would have aligned the rock and rendered the subject as a collection of ghosts -- one for each different position it blew into.
On the larger image, looking at the leaf which points directly out at the viewer, there is a horizontal line artifact which I assume is due to the DOF of two adjacent slices not actually overlapping.
From this one image I can't be sure what's going on. It could be a transition between depths if this image was rendered by DMap. If it's PMax, which I'm inclined to think because I can't see any halos, then probably something else is going on because PMax rarely makes a sharp transition between depths.

--Rik

PaulFurman
Posts: 595
Joined: Sat Oct 24, 2009 3:14 pm
Location: SF, CA, USA
Contact:

Post by PaulFurman »

Fascinating seed head - that would be worth revisiting, with a soft out of focus, dark, complimentary color background. I'll have to hunt down a local Gallium species & take a closer look, thanks for finding this. Soft diffused lighting would give a very different look to this shiny glistening subject.


What is that on the left - soil glued on it with snail poop? :-)

ChrisLilley
Posts: 674
Joined: Sat May 01, 2010 6:12 am
Location: Nice, France (I'm British)

Post by ChrisLilley »

Bright Mediterranean sun does give a harsh look, I agree.

No idea what the glued-together collection of small stones is. Presumably some sort of insect made it?

Craig Gerard
Posts: 2877
Joined: Sat May 01, 2010 1:51 am
Location: Australia

Post by Craig Gerard »

ChrisLilley wrote:
Presumably some sort of insect made it?
Possibly some type of debris-carrying Larvae.

http://bugguide.net/node/view/72302/bgimage

Craig
To use a classic quote from 'Antz' - "I almost know exactly what I'm doing!"

ChrisLilley
Posts: 674
Joined: Sat May 01, 2010 6:12 am
Location: Nice, France (I'm British)

Post by ChrisLilley »

Craig - I agree. I was vaguely thinking of Caddis-fly larvae which I used to look for as a small boy (they are under water, but same idea of glued-together bits of debris).

Post Reply Previous topicNext topic