Generic selectors
Exact matches only
Search in title
Search in content
Search in posts
Search in pages
Home > Featured > The Desktop Camera: Using a Scanner to Make Photos

The Desktop Camera: Using a Scanner to Make Photos



I’ve owned or borrowed a few cameras in my time, but there’s one ‘camera’ I’ve owned for longer than I can remember, and would be very reluctant to get rid of: my flatbed scanner. Okay, as a camera it’s a bit… sub-optimal. It’s not easy to take outdoors for thing, but at this present moment in time that’s really not a huge problem. And, admittedly there are a few restrictions on the types of photos I can ‘shoot’. But isn’t overcoming the various quirks of a camera part of what makes photography such a fun thing to do?

First though, a brief description of how a flatbed scanner works…

When you want to scan something –  a document say – you place it onto the scanner’s glass plate. Below the plate is an imaging head, that uses a CCD array to capture light to produce an image. (The light provided by a lamp inside the scanner.) Press the scan button and the imaging head is moved along a stabilising bar across the document, repeating this action as the stabilising bar and imaging head are moved incrementally along the document. This all takes time, far longer than it takes to shoot a typical exposure with a camera. It means that scanners can’t be used to shoot anything that moves, though their slowness can be used to create a nifty effect which we’ll come back to later.

If you want to shoot something that could potentially move – your own hand for instance – you’ll need to ensure that it stays completely still until the scan is complete.

Shooting documents and other flat subjects is boring though. More fun can be had scanning three-dimensional things. However, doing so reveals another restriction: limited depth of field, with no way to increase it.  Scanners weren’t designed to shoot three-dimensional things. Essentially the only bit of a scanned three-dimensional object that will be sharp will be the bit that rested on the glass. This is a drawback if you want pin-shape images, but sometimes a bit of softness is appealing.

It’s easy to get flat three-dimesional objects – if that’s not a contradiction – nice and sharp, particularly if they’re fairly geometric like this pair of pliers.

Scanners come in a variety of shapes and sizes, but the most common is probably the A4 scanner, which has a glass plate approximately 22 x 32cm. This is another restriction that needs to be overcome: you can’t shoot huge objects with a scanner. (Though you could scan overlapping areas of a large object and – as long as you’ve scanned everything you need – ‘stitch’ the scans together later, rather like creating a panoramic photo.)

Practically, you also have to think about the weight of what’s placed on the scanner too. The glass plate is fairly tough, but it’s really not a good idea to put anything on it that’s too heavy; you really, really don’t want the glass to crack. It’s also a good idea to avoid scanning anything that could scratch the glass too. Before you scan it’s a good idea to clean any dust off the glass plate. This will save cleaning the image up in post-production later.

I cleaned all the grit and sand from these pebbles before putting them very, very carefully on to the glass plate of the scanner.

You’ll be presented with a number of choices when you run your scanner’s software. One is resolution. Typically I use 300 DPI for larger objects, or 600 DPI or greater for smaller objects (the smaller the object, the higher the resolution I use.). The key is to use only a scanner’s optical resolution. Some scanners boast insanely high digital resolution. However, this is a bit of a cheat as the scanner uses interpolation to achieve this resolution. (Essentially scanning at the highest optical resolution and then ‘blowing up’ the image to the required resolution in software afterwards.) Using a scanner’s optical resolution will always give sharper results. If you need to increase the resolution of your photo later you’ll get far better results using Resize in your editing software.

Another choice is colour depth. Which you choose will depend on how much editing of colour and contrast you think will be required later. If it’s a lot, choose 48-bit (or 64-bit) colour and save the final scan as a TIFF. The downsides to scanning in 48/64-bit colour is that the time taken to make the scan increases, and the resulting file will take up more room on your hard-drive. If you don’t think you’ll want to edit the photo scan using 24-bit colour and save as a high quality JPEG.

This image of rose petals required quite a bit of post-production. All the colours, including the black background, were inverted, and then the petals given a pale pink wash.

As mentioned above, a scanner creates its own illumination. However, scan a three-dimensional object and you won’t be able to close the scanner lid. This means that ambient light will leak into to the scanner, reducing the quality of the final image. One simple solution is to shoot at night, when you can scan in the dark. Or, put a box over the scanner as you scan. Use a box with a plain interior, black or white work well. (For a more interesting effect, try using aluminium foil as a lining for the box.)

Now, back to the fact that scans take time to complete. If you move your subject during the scanning process, you can change the shape of the subject. Move it in the opposite direction to the scanning head then you’ll compress its shape, or stretch it if you move it in the same direction and at the same speed as the scanning head.  The key is to experiment and don’t be put off if the experiments don’t succeed first time.

To create this image I placed an old print onto the scanner. Then, as the scan was being made, I gently moved the print from side to side.

So, why don’t you make it a ‘resolution’ to try using your scanner to make images? Stay safe.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.