tiny cactus experiment

[DanK]

This was the most difficult image stacking I have done, and I have done a lot. This is a very small cactus. The flowers are approximately 9 mm across when fully opened. The bane of image stacking is depth, particularly large distances between edges and the surface behind them--precisely like this. This is a stack of 20 images, done with Zerene PMax, which has less problem with halos than the DMax method I usually use with flowers. I removed some halos by retouching from individual images within Zerene, but that left me with a lot of ugly, large halos in numerous places. I removed much more using the clone tool in photoshop.

C&C welcome, as always.

[Shadowman]

Very nice.

[xpatUSA]

Most impressive. It is often instructive to compare the result with one of the originals.

I imagine that the spikes provide lots of control points for the stack alignment phase.

[Jim A]

Amazing and beautiful, Dan!

[DanK]

Thanks. Ted, as long as things aren't moving around, alignment isn't generally a problem for good stacking software. I did have one slight alignment problem in this case because I didn't let things settle enough between shots (I work on a wooden floor), but it wasn't serious.

The main difficulty is parallax If there is a large distance (front to back) between an edge and the surface behind it, there is too great a difference in apparent size between the slice that has the edge in focus and the slice that has the background in focus. Zerene allows you to paint from any image onto the composite to correct errors, but when the distance is too great, there are no images that are useful for this. You can also paint from a PMax composite, which generally has less severe halos, onto a DMax composite, which is my normal procedure. In this case, the halos were sufficiently severe that I used the PMax composite as my base. However, I had to do most of the touch-up the clone tool, which as a real pain given all the nooks and crannies caused by the thorns in this image.

[Tri Danimal]

holy cripes. Man thats a lovely image. Im continually impressed by your work. Keep up the insperational work.

[wilgk]

That is wonderful, patience & effort well rewarded.

[DanK]

Thanks for the kind comments. This is a genre I had put aside for a while, and it is usually something I do in the winter, when the outside is pretty barren here, but perhaps it's time to do a little more.

[Wavelength]

O, this is so close to viewer's heart

[dem]

That's sharp!

[TonyW]

Yes, that is sharp. Did you change focus by moving the camera or changing the lens focus?

I tried something like this using photoshop but it was a dismal failure. It seems that the small scale depth changes male it difficult. This inspires me to make a proper attempt.

[DanK]

Tony,

no, I didn't move the camera; I changed focus using the lens barrel. There was some discussion of this on this forum some time ago, sparked in part by people saying--incorrectly--that one needs a rail to do this sort of work. The bottom line is that for things of this size, a rail will produce inferior results to those obtained using the focusing ring. A generalization offered by the author of Zerene, the stacking software I use, is that you are better off using the focusing ring for any subject larger than a raisin.

The key is learning how much to turn the barrel to avoid unfocused areas between images.

You can also buy software that will automate this: you indicate the nearest and most distant point of focus, and it moves the focusing ring in appropriate increments. I have one such piece of software (Helicon Remote), but I never use it.