I have a lot of infrared aerial photos printed onto transparent film with dimensions 24cm x 26cm. They should really be scanned with a scanner incorporating a light source, but we unfortunately are not able to outsource that task. I want to use my Olympus E-420 DSLR to capture the images in the best resolution possible whilst maintaining all the colour information. I have a light table, tripod, remote control, and the DSLR.

Which and focal setting would be best to avoid distortion as these photos will be referenced in a geographic information system to be used as basemaps for environmental agencies.

I have done a few tests and found that only the middle of the photos are really sharp and the outer edges are blurred. Is there any way to reduce this perpective-blurring? I really need to retain the colours as well, as these colours give meaning the photos regarding vegetation density and type.

  • \$\begingroup\$ I'd do a mosaic of telephoto shots with extension tubes so you can focus close. \$\endgroup\$ Nov 30, 2015 at 20:42
  • \$\begingroup\$ Unfortunately I don´t have a telephoto lens. I have a Pancake and 14--42mm \$\endgroup\$ Nov 30, 2015 at 21:01
  • \$\begingroup\$ Well if that's the case I would suggest 42mm (unless the pancake lens is longer) and get as close as you can and still focus then putting the camera in aperture priority and seeing how clear the edges are when you increase the f/stop. You may need to just crop the centers and use those if that if the best you can do. (Either that to take the photos to the closest photography/office supply shop and see if they can scan them.) \$\endgroup\$ Nov 30, 2015 at 21:14

2 Answers 2


Ideally for this you want a macro lens, as they tend to have flat imaging fields which are ideal for this sort of copying work. But we can only work with what you've got :-)

So -

  • 14-42 will be your best lens for this assuming you're talking one of Olympus' plastic pancake lenses with manual focus.
  • Longer focal lengths in general help, but in this case will increase the risk of barrel distortion distorting your originals. I'd shoot in the middle of the range at 28mm. On a tripod if at all possible, you want this as flat and straight as you can get this.
  • All lenses are sharper in the centre of the field than the edges to some degree. Stopping them down helps, but if you do it too much then diffraction starts limiting your resolution. I'd start with about f/8.
  • Even lighting will make a major difference. A guy I used to know had done something similar for art students many years ago, using white foam board or polystyrene on either side and (matching, to avoid colour temperature issues) desk lamps, pointed at the foam. It gave even, soft lighting across the whole image.

I'd still definitely try to find a way to get them commercially scanned if at all possible - for the usage you want to put them to that'd give much higher quality, but with the kit you've described that's how I'd do it.

Good luck!


Use the longest focal length lens you can, and stop it down at least to f8. This will give you the most depth of field in the "sweet spot" of most lenses. You don't say if your transparencies are perfectly flat on the light table or not. Depending on the type of film it might have some curve to it pushing it up off the table and causing it to go out of focus. You can actually achieve pretty good "scans" using a light source and a camera. Since you are shooting a still setup you can just adjust the shutter speed to meet your needs at that f-stop. Also make sure your light table and camera are level in both directions so you don't get any keystone distortion. Set your camera to the lowest ASA setting possible to help reduce noise.

  • \$\begingroup\$ Thank you for the comments. I have indeed ,as you mentioned, found that the transparencies do bend slightly on the flat surface of the light table. I suppose I could tape the corners down,, but maybe you have a better solution. Laying a clear pane of glass or perspex over the image would only distort it more i can imagine. \$\endgroup\$ Dec 3, 2015 at 10:00
  • 1
    \$\begingroup\$ Depth of field decreases in proportion to focal length increasing. I don't think depth of field will be the limiting factor here, but that's not the best way to increase it. \$\endgroup\$
    – eftpotrm
    Dec 3, 2015 at 14:03
  • \$\begingroup\$ I understand that. but he's basically approaching macro photography. Using a wider angle lens will give plenty of depth of field but introduce more distortion. I'm not recommending he go out and get a 500mm lens. His longest lens is his 42mm. I certainly wouldn't shoot any with anything shorter. Depth of field will become much shallower as he will have to shoot as close as the lens will let him or more likely get an extension tube. Shooting wide open will leave him with paper thin depth of field. \$\endgroup\$
    – Beartech
    Dec 3, 2015 at 21:21
  • \$\begingroup\$ Robert, you can put something over the image if it's a really good piece of glass. I would try it to see the results. Normally the problem with laying glass over something is worse when you are dealing with a photo rather than a transparency. Since you will ONLY be lighting from behind, it is less of a problem. Personally I would get a macro bellows unit like this: amazon.com/Neewer-Focusing-Bellows-Cameras-Digital/dp/… \$\endgroup\$
    – Beartech
    Dec 3, 2015 at 21:30
  • \$\begingroup\$ Put a Nikon to M4/3 adapter on the camera side, and a 50mm used nikon lens on the other end. You'd be able to fill the frame with the transparency and get perfect focus. \$\endgroup\$
    – Beartech
    Dec 3, 2015 at 21:31

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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