Photography: Appreciation through Understanding

The vanishing DoF at large magnifications – CombineZM to the rescue

Comments[0] 05:14PM Jul 15, 2007 by Charles in General

When the magnification is greater than 1X, the depth-of-field (DoF) becomes paper thin; and 3D objects cannot be imaged very well.  For subjects that are completely still, one can take a series of photos with different focus setting and then attempt to combine them.  By this I mean the construction of a composite image by using the sharp parts of the component images.  If you have tried this, you know that there are problems.  First of all, our camera lenses are not telecentric, and the magnification changes with the focus setting.  Therefore, the first step is to resize the images before trying to bring them into register.  With more than a couple of images, this gets to be a lot of work.

As you probably know, there are computer programs for image reconstruction and DoF expansion.  One review of such programs can be found at:  http://www.crystalcanyons.net/pages/TechNotes/3DMicroMacro.shtm

Here Prof. Hart compares various programs and concludes that the freeware program CombineZX does rather well.  Information about Combine ZM can be found at:

http://www.hadleyweb.pwp.blueyonder.co.uk/CZM/Manual/combinezm_help.htm

 
A couple of weeks ago I downloaded CombineZN and finally got around to testing it.  I have only used the automatic mode, but I am impressed.  Here I show a few images to serve as an illustration.  The shots were taken with a Sigma 150mm lens at close to 1:1 focus (1X magnification) with 68mm of extension tubes added to bring the magnification to 1.68.  In all cases I used f/5.6.  I took seven shots with a change in focus the same direction each time.  These are photos of a fly about 2mm in length that has been trapped by a Cape Sundew plant.  In the first image (148 kb, jpg)  I attempted to focus on the closer eye, and the problem with DoF is obvious.  The second image (209 kb, jpg) was constructed by CombineZM from a stack of seven jpg images.  This image was satisfactory from the start, but I did improve a few places with the clone tool.  I have a lot more to learn about CombineZX, but I thought that you might enjoy playing with it also.

By the way, the plant is small.  This photo (228 kb, jpg) puts it all into context.   

Diffraction Problems in Macro Photography

Comments[0] 04:11PM Jul 15, 2007 by Charles in General

Avoiding diffraction problems in macro-photography:  We all know about the loss of resolution that occurs with small apertures (large f#’s), but it is easy to forget that the effective f# is not equal to the one shown in the camera.  The f# we need to use is: f#(effective) = f# times (1 + m).  Here m is the magnification.  My Sigma 150mm macro lens will focus to 1X magnification.  At that point all the f#’s shown should be multiplified by two in order to estimate the DoF and the effect of diffraction.  It gets worse with extension tubes.  I add 68mm of additional extension to get 1.68X and the f#  shown needs to be multiplied by 2.68!  When I forgot this factor and set the camera for f# = 22, I got unsatisfactory images.  As an example take a look at this figure:

Sundew plant (100% crop of 1X macro photos) (635 kb, jpg)

The setting of f/4.5 is probably the best and f/8 is satisfactory, but f/22 is fuzzy.  You can find useful information in John Shaw's book, "Closeups in Nature."