_______________________________________________________________________

Developing the digital film

What matters in this section is colour rendition, rather than masses of fine detail. Although JPEG compression is still used, the photos have been sized-down to 480 x 640 pixels, allowing a small file size with less compression. These are photo-processing examples, not demonstration pictures. One microscope photograph will be followed through as the 'digital film' is developed.

Because the original 'hot mirror' coating on the back of the manufacturer's lens has gone, along with the lens itself, the colour rendition has been changed. This filter had a strong green cast, and the 'auto white balance' of the camera was set to compensate for this. The camera settings are left on auto white balance, simply because that is the default. You may use another setting. The result, combined with the low light levels of a photomicrograph, and the subject matter, is a dull red picture. Don't panic.

As long as all that needs to be done to a colour image is adjust the red/green/blue levels, set the brightness, and alter the contrast, this can be done in a few minutes. Terrible photos, which need the 'curve responses' to be altered until things look better, can take half an hour or so, and mercifully are not produced by the ledlyt/TDC 1b combination. Any photo-manipulation package (Adobe Photoshop is the most famous) will allow these basic alterations to be carried out, along with several hundred more options.

I use Ulead Photoimpact 4 (Win 98SE) which was given away (by Ulead) a few years ago as a slightly outdated, but fully operational package. Virtue is more than its own reward. The package was so solid that Ulead, previously restricted to supplying scanner makers with add-on packages, was able to sell full packages successfully on the open market. Good on them. Photoimpact 4 provides a 'quick keyboard' to be mouse-clicked to the right of the screen, providing incredibly fast incremental manipulation of red, green, blue, brightness and contrast. Any other package has this facility somewhere, but as well as being slow, it is accompanied by a little check box saying 'maintain brightness' or similar. Uncheck this box. Facilities are often provided for working on the dark, medium and light areas of the photo separately. For a 'quick and dirty' correction, which is all you need here, such facilities, if compulsory, just get in the way.

To recap, we have a dark red picture, as they all are with my settings. We need green, lots of it. I keep clicking the green button to progressively raise the green level until the picture is just too green. It has got much lighter in the process. You may be shifting a slider with the mouse to achieve the same effect.

We need blue, but not nearly as much. I click the blue button until the picture is just too blue. We are starting to get somewhere.

Before the contrast is raised, the picture has to be set to a brightness such that the background darkens or lightens very little as the contrast is raised. If there is any skill to the process, it is the setting of the brightness at this stage. In this case, if the brightness is raised, the background brightens too much when the contrast is raised. The best brightness happens to be just where it is. The picture is expanded to full screen, and the contrast raised until the best picture results. You will know when you have gone too far.

Appreciable detail has developed in the spherical cells. The process is quite legitimate since you have made what was already in the image visible. Painting detail in with the artwork facilities is another matter!. The contrast enhancement has left the background too blue, so the final step is to use slight red/green shifts to reset it to something closer to that seen down the microscope. The 'correct' background colour is as close as you can easily get to a light, neutral grey, but that is a matter of taste.

That concludes basic digital film development. I get this far, using Photoimpact 4, in less than two minutes. To remove the obvious dust specks, use the 'clone' tool and read the 'Help' on the subject first.

To my eyes, the detail is fine, but the colour has ended up a little on the gross side. If I were presenting this as one of the major pictures, I would cut the saturation down by about six per cent, which would move the background more towards grey, and cut the lurid colours down to size. Since a saturation decrease boosts monochrome at the expense of colour, the line detail inside the cells sharpens a fraction more. A very small amount of 'unsharp mask' could be used to add a bit more 'bite' but the technique is not entirely legitimate.

Two colonies of the green algae Sphaerocystis, with a Coelastrum in the top background. Out of focus is a Staurastrum and several strands of the diatom Aulacoseira. Zeiss Winkel 25X, Olympus GB40, ledlyt illumination

What we have left, believe it or not, is more or less what the eye sees looking down a binocular microscope at a ledlyt-illuminated image.