First attempt at stacking 1:2 close-up photos taken at 4 ft

A forum to ask questions, post setups, and generally discuss anything having to do with photomacrography and photomicroscopy.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

LVF
Posts: 66
Joined: Sun Apr 23, 2017 4:17 pm
Location: Sequim, Washington

First attempt at stacking 1:2 close-up photos taken at 4 ft

Post by LVF »

I have posted in April, in this forum, several tests I performed in my search to be able to take close-up photos without getting close to the subject. I learned from these tests, that I can take 1:2 close-up photos at a 48 inch working distance. My Nikkor 105mm f/2.8G lens and my Sigma 180mm APO MACRO lens require a 10 inch and a 16 inch working distance, respectively, for a 1:2 photograph. So 48 inches working distance for a 1:2 photo would really be great. Note - by working distance I mean the distance from in-front of the lens to the subject being photographed.

However, the penalty for getting 1:2 photos at 48 inches is a very shallow depth of field. So, I have to do photo stacking of the subject to get a deeper DOF.

However, I have never stacked photos, so I do not know that this can be done; until now. I have Photoshop CS6 which can do stacking of photos, so I decided to learn photo stacking.

The following post shows my successful photo stacking of a subject 48 inches from the front of my lens.

In my past tests posted here, I found that using the Kenko 68mm extension tubes attached to the Nikkor 300mm f/4E PF lens produces 1:2 photos with the front of the lens 48 inches from the subject.

So I decided to use the 68mm extension tubes with the 300mm lens to stack photos of a subject to see how much I can increase the DOF.

My very first photo stacking test used the following equipment setup.

I attached the Nikkor 300mm f/4E PF lens with the Kenko 68mm extension tubes to my Nikon D500 camera. I mounted the camera on a Really Right Stuff Macro Focus Rail and mounted this on a tripod. I used live view to initially manually focus the lens. I used the Nikon MC-36 cable release to fire the shutter. I set the camera at ISO 100, and the lens at f/8 (effective f/10 because of extension tubes). I set the front of the lens 48 inches from the subject to be photographed.

The test subject was the Lens Align DOF Ruler mounted on its Base Assembly. The Ruler was illuminated with a Ott Lamp which gave a white
balance around 4900 degrees kelvin. Here is a photo of the DOF Ruler:

Image

I took 8 photos by moving the RRS Focusing Rail in small increments. I post processed the photos using Photoshop CS6. The final stacked photo
was a complete mess. The little squares were twitched in different directions, and the numbers were rotated in different directions. I examined the 8 photos in Photoshop and found that each photo was at a different angle to the other photos. It turns out that the Focusing Rail produced to much wobble as I turn the knob. This was due to the 8.5 inches of extended mass hanging off in front of the camera (lens + ext. tubes). Photoshop just could not make since out of the 8 photos.

So I removed the RRS Focusing Rail and focused the lens by manually rotating the lens focusing ring. I also decided that the DOF Ruler was to complicated. So I masked the Ruler so that only the second column of numbers on the right side of the ruler were showing. Here is a photo of the masked ruler:

Image

Here is the first photo which shows that I focused on the zero mark on the ruler:

Image

This photo shows that the 0 and 1 marks are in sharp focus. That gives a DOF length of 0.25 inch at 48 inches.

Here is the resulting photo stack of 6 photos taken by rotating the lens focus ring clockwise in small increments:

Image

This stacked photo shows that the 0 to 16 marks are in focus. That gives a DOF length of 2.25 inches.

Since I was able to get a good stacked photo by rotating the lens focus ring, I decided to try photographing the entire Lens Align DOF Ruler
again to see if I could get a good stacked photo.

Here is the first photo of the whole Ruler which shows that I focused on the zero mark in the first column to the right:

Image

This photo shows that the 0 mark in the first column on the right is in sharp focus. That gives a DOF length of 0.25 inch.

Here is the resulting photo stack of 6 photos of the whole Ruler taken by rotating the lens focus ring clockwise in small increments:

Image

This stacked photo shows that the 0 to 18 marks in the first column on the right are in focus. That gives a DOF length of 2.50 inches. Notice that the whole DOF Ruler above the zero mark is in focus.

Next I decided I would try to stack the full 10.5 inch Lens Align ruler. I had to lay the ruler at a very shallow angle to get the full ruler in the field of view which is slightly less than 2 inches.

I took 21 photos of the whole ruler at f/8 and stacked them in Photoshop. Here is the resulting stack of 21 photos:

Image

You can see the numbers for the full length but there are obvious problems. Having never stacked photos before, I thought maybe I did not take photos at small enough distances. So here is a stack of 61 photos where I turned the focusing ring at very small increments:

Image

I am getting there but there are problems around the center zero marker.

Next I thought that I should increase the depth of field, so I increased the f number of the lens from f/8 to f/22 (eff. f/27). I took 8 photos of the ruler. Here is the first photo taken at f/22:

Image

Here is the final stack of 8 photos taken at f/22:

Image

Bingo! This stacked photo shows the full 10.5 inch ruler without any distortion! However, it looks like using f/22, which is effectively f/27 because of using extension tubes, caused diffraction, so the numbers are not very sharp.

So I decided to lower the f number to f/14 (eff. f/17) to see if I could get sharper numbers. Here is the first photo taken at f/14:

Image

Here is a final stack of 16 photos taken at f/14:

Image

The numbers are some what sharper but there is a problem around the center zero mark on the left side of the ruler.

I concluded that the problem is me. It is very difficult to accurately turn the focusing ring in the same amount of very small increments. So I must have messed up in turning the focusing ring when the lens was focused around the area of the center mark on the ruler.

However, I just showed that I can get a stacked photo of a 10.5 inch ruler from one end to the other end of the ruler with 1:2 close-up photos taken at 48 inches from the ruler.

Obviously I need more practice, but I am very satisfied with my first try at photo stacking. It is amazing that I can get a depth of field 10.5 inches in length with close-up photos taken at 48 inches.

I hope members will find this information useful.

Leon

rjlittlefield
Site Admin
Posts: 23564
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Re: First attempt at stacking 1:2 close-up photos taken at 4

Post by rjlittlefield »

LVF wrote:The numbers are some what sharper but there is a problem around the center zero mark on the left side of the ruler.

I concluded that the problem is me. It is very difficult to accurately turn the focusing ring in the same amount of very small increments. So I must have messed up in turning the focusing ring when the lens was focused around the area of the center mark on the ruler.
My guess is different. I think that Photoshop messed up its feature matching while trying to align the images. This sort of highly repetitive pattern is hard for image alignment algorithms to handle. It's sort of like trying to make a stitched panorama of a brick wall -- there's no compelling reason why the software should match brick A with brick A' instead of brick B. This is especially true for Photoshop's method of image alignment, which relies on individual small features rather than gestalt pattern patching.

To continue your experiments I suggest two revisions:

1. Try using Zerene Stacker instead of Photoshop. It uses a completely different method of image alignment that is more tolerant of this sort of target. If nothing else, I can tell you better what's going wrong because I wrote that software.

2. Read http://www.zerenesystems.com/cms/stacke ... versusrail to learn something about why focusing by ring is better than focusing by rail in this situation. (It is because using the focus ring introduces less change in perspective from one image to the next.)

--Rik

ChrisR
Site Admin
Posts: 8668
Joined: Sat Mar 14, 2009 3:58 am
Location: Near London, UK

Post by ChrisR »

Fwiw, I tried comparisons of a converter, tubes and close-up lenses on a good 300mm lens. I don't remember the exact mag ratio, but the subject was a 35mm film box. Not sure if I posted here or not. Best was with Nikon T5 and T6 dioptres , easily.
Tubes were the worst.

YMMV!
Chris R

LVF
Posts: 66
Joined: Sun Apr 23, 2017 4:17 pm
Location: Sequim, Washington

rjlittlefield and ChrisR

Post by LVF »

Today I did a stack of the Lens Align Ruler using the Nikon 17E II teleconverter instead of the Kenko 68mm extension tubes. I took 32 photos of 1:3 close-up photos of the ruler. I had to move the camera back from 48 inches to 72 inches to get the full 10.5 inch ruler in full view in the camera.

Photoshop CS6 did a final stack of the 10.5 inch ruler with no distortion and very sharp numbers and the sharp black squares.

I will present the findings tomorrow, Thursday, May 4th.

Leon

Post Reply Previous topicNext topic