1
\$\begingroup\$

I'm shooting laboratory photos of small (60mm) sized objects. It is essential for me to use the entire camera sensor to achieve adequate spatial resolution, so I am putting the camera as close as needed to have the the objects reach the sides of the photo.

With this amount of zoom, image distortion is a problem. With a 105mm focal length lens and a working distance near 30cm or so, the inside edges of objects are revealed a bit from the side, while the outside edges are visually occluded. This means the images do not reflect a top down view on the scene, as you can see here in this photo of a periodic array of vertical cylinders:

The inside faces of the cylinders are visible, the outside faces are not

This distortion is very bad for my scientific application. I guess to remove this distortion while keeping the object over the entire image sensor, I need to shoot from further away with a longer focal length. Is this correct? Are there other things I might try to avoid this distortion like shooting through some kind of lens, adding some attachment, or some more hairbrained scheme like photographing mirrors aimed at the object from different angles? I am currently using a Nikon Nikkor Z MC 105mm f/2.8 VR S.

Thank you for any help. I am not an experienced photographer, so any knowledge you could share would be useful.

\$\endgroup\$
7
  • 1
    \$\begingroup\$ Does this answer your question? Is it fundamentally possible to capture an orthographic image of the real world?. \$\endgroup\$
    – scottbb
    Jul 11, 2023 at 22:31
  • 1
    \$\begingroup\$ Also related, How can I take undistorted overhead top view pictures of a stationary gas compressor?, is again about taking an orthogonal overhead image without perspective distortion. \$\endgroup\$
    – scottbb
    Jul 11, 2023 at 22:33
  • \$\begingroup\$ Thanks @scottbb - i hadn't heard of telecentric lenses before, and the discussion in the second link especially is useful to see the relevant terminology to describe my problem. Of course, I'm aware the distortion is the result of perspective, but it's difficult for me to to predict if I could solve the problem well enough by going to a 200mm lens and a larger working distance. Is there maybe some way I can calculate the distortion I'd see with a given focal length given the constraint that the object fills the full image? \$\endgroup\$ Jul 11, 2023 at 22:57
  • \$\begingroup\$ The query you linked to is about a phenomenon called barrel distortion. This type of lens distortion causes vertical lines to appear bowed outward, like a barrel. It is most noticeable in close-up photos, and various factors, including the focal length of the lens, the aperture setting, and the distance between the camera and the subject, can cause it. You can do a few things to reduce barrel distortion in your photos. You can use a lens with a shorter focal length. This will make the image appear wider, which will help to reduce the amount of distortion. \$\endgroup\$
    – Bijutoha
    Jul 12, 2023 at 4:20
  • 1
    \$\begingroup\$ @kevinkayaks overhead perspective will cause vertical (that is, parallel to the optical axis) objects to appear to lean outwards, but that is not barrel distortion. Barrel distortion nonlinearly amplifies or magnifies the effect, but the effect will still be there even in the absence of barrel distortion, or even in the presence of pincushion distortion (the opposite of barrel distortion). \$\endgroup\$
    – scottbb
    Jul 13, 2023 at 5:33

1 Answer 1

3
\$\begingroup\$

Yes, increasing the distance is the most practical solution.

You can use your existing lens from any distance to determine at what point the perspective is suitably overhead/parallel. You would simply need to zoom in during image review. I.e. all lenses record the same perspective from the same distance; the same your eyes see. And focal length simply crops and magnifies that perspective... really no different than doing so after the image is taken (if there is enough image resolution to allow it).

Once you find the right distance, then you can use that to determine the right focal length to fill the sensor (4x dist = 4x FL). You will also want to choose a lens with minimal optical distortion (barrel/pincushion) and probably one with flat field focus (more important at shorter distances, and more common in macro lenses).

\$\endgroup\$

Not the answer you're looking for? Browse other questions tagged or ask your own question.