Question on pixel trails

canonian · Post by **canonian** » Sun Jul 01, 2012 1:01 am

In a topic on testing a microscope objective I spotted some pixeltrails in one of the images, send in on reply.
It made me think that the setup used to shoot the frames is very solid:

Detail of a Cream-Spot Ladybird (Calvia quatuordecimguttata) by Laurie Knight

Mine seems to wander off, I guess I might look into the alignment of the parts or the stability of my vertical setup, but I'm not sure:

Zerene has to calculate a lot of alignment and I wonder if that's influences the quality of the final rendered image.
Does short or long pixeltrails stand in relation to setup stability or are there other factors involved? Are pixeltrails ''bad news" for the setup or maybe for the sensor?

rjlittlefield · Post by **rjlittlefield** » Sun Jul 01, 2012 2:26 am

There are a lot of factors involved. Especially around the edges of the frame, trails can be influenced by perspective of the lens and by the way things "bloom" as they go in and out of focus.

In the case that you're showing, some of that vertical drift could be an artifact caused by having lots of nice stable black background on top versus a lighter detailed subject that will tend to bloom at the bottom. On the other hand, considering the magnitude of the steps, I'm inclined to think that the drift reflects real misalignment in the setup, maybe something as simple as the lens is pointed a bit downward with respect to the direction of focus stepping. The only reliable way I know to separate the factors is to shoot some nice uniform subject like a chunk of construction paper or leaf or moth wing that fills the whole frame

You mentioned about "quality of the final rendered image". Personally I've not found output quality to be much affected by how the trails move around. Perhaps other people have had other experiences, and if so I'd be interested to hear about them.

--Rik

BugEZ · Post by **BugEZ** » Sun Jul 01, 2012 2:03 pm

I use pixel trails to help align my rig. If the trails point to the middle I am usually well aligned. See here...

http://www.photomacrography.net/forum/v ... 124#111124

In the instance shown above, the nice tight group near the center of Laurie's photo suggests to me good alignment.

Kind regards,

Keith

rjlittlefield · Post by **rjlittlefield** » Sun Jul 01, 2012 2:24 pm

BugEZ wrote:I use pixel trails to help align my rig. If the trails point to the middle I am usually well aligned. See here...

http://www.photomacrography.net/forum/v ... 124#111124

In the instance shown above, the nice tight group near the center of Laurie's photo suggests to me good alignment.

I agree, but I notice also that this particular group is near the center of Laurie's crop, which is actually in the upper right corner of the full frame. The fact that the group is tight in the corner means that not only is the apparatus well aligned, but also that scale has not changed throughout the stack. That is, Laurie's trails in this case apparently don't point anywhere -- they're just tight random clusters. This too can happen for at least a couple of different reasons. It could be that the objective is naturally telecentric, or it could be that Laurie turned off Scale in the Alignment options when he was processing the stack. It's often a good idea to turn off Scale adjustment at high magnification because it can do more harm than good. See HERE for some further discussion of that.

BTW, this thread seems misplaced in the Technical and Studio image gallery. I will move it to Technical Discussions.

--Rik

BugEZ · Post by **BugEZ** » Sun Jul 01, 2012 3:21 pm

Rik wrote:

It's often a good idea to turn off Scale adjustment at high magnification because it can do more harm than good. See HERE for some further discussion of that.

Very neat and something else to try! Thanks for the suggestion. I have previously just used the defaults. I know just the image (a relatively deep 10X stack) to try with the scale adjustment turned off...

Keith

canonian · Post by **canonian** » Mon Jul 02, 2012 3:19 am

Thank you Rik that's good advice, I left all settings on default too.
Yesterday I tried to do some partial stacking in the problem area.
What surprised me was that the trails were mostly in the off-focus frames of the stack and not in the 20 or so frames that were in proximity of focus around that area.
Ticking and unticking rotation and scaling settings gave me a different scattered pattern in the full stack, but only in horizontal way.
Vertically the trail lenght was the same.
So now to find out what al this means in relation to (mis)aligning the parts.
I 'm using an Ihagee stand which is very flexible in positioning but also leave plenty room for mis-alignment.
In my previous post I was wondering if misalignment could cause troubles, but never experience degradation in the final rendered results.