Digibird.com
Primer: Seven Quick Steps to Image Processing


Click here for the Digiscoping Technique Primer.

You aren't done once you've got a memory card full of images. Now it's time to prepare them for presentation. Because birds don't consult you regarding your photographic preferences or camera settings, your images will frequently be less than perfect coming straight from the camera. One of the strengths of digital photography is the ability to process your images--correct them, if you will--on your desktop, where you can instantly see the results. Equally important, you can easily undo your processing mistakes, and thus feel free to experiment with your tools.

You will need two types of software to process your images efficiently. The first is a viewer, an application that allows you to review a large number of images quickly at full screen expansion. Such a review is known as a slide show; a web search on that term will reveal many software choices available for a nominal fee. Your camera or computer might well have been bundled with such software. (Check the manual or camera website.)

The second type is an image processor. The de facto standard here is one of the variants of Adobe Photoshop, a stripped-down version of which is often bundled with the camera or printer; although the full version is quite expensive (and worth it if you are serious about fine art digital imaging), the "LE" or "Elements" packages are available for under $100 USD. This discussion assumes use of the LE version.

Step 1: Backup!
And do it now! Save your raw camera files on a separate medium (preferably CD-R). This prevents their loss by hardware failure or accidental overwrite (the one processing step that you can't undo). Only a fool would attempt to manipulate the only copy of a precious image. You may find it helpful to save images in separate folders for the dates taken.

Step 2: Select your BEST images
Use your viewer to select the images that are worth further processing. There are two criteria for you to consider. First, is the image detail crisp, with the subject in focus and without motion-induced blur? Second, is the composition of the image pleasing, for aesthetic and/or documentary reasons? The answer to both questions should be "yes" to merit the effort necessary for further processing. You will need to be brutally objective with your images; remember that for every beautiful bird photo you've ever seen published, the photographer probably rejected 50+ others. Like you, they felt as if they were trying to choose their favorite child.

Your images will vary in the degree of detail they carry, and there isn't an objective measure of this. Even in a sequence of identically composed frames, focus and blur can change radically as the bird moves, camera shake settles, or atmospheric steadiness changes. One helpful technique for selecting those of acceptable detail is to watch a slide show of the entire day's shoot, without making any notes, to see what the range of detail is. Once you have an idea of "good" and "bad", make a careful pass through the list and reject any photo that doesn't carry an acceptable level of detail. Take your time doing this. According to tradition, the eyes of an animal should be in focus.

Once you have a list of detailed images, select those for which the composition pleases you. Generally, if you have focused on the head, the depth of field is so shallow at the focal lengths produced by the camera/scope combination that the background is beautifully blurred. This means that your subject will be isolated from distracting background (or foreground) shapes, so you can concentrate on how well the bird's beauty and character is revealed.

From this point, all subsequent image manipulation will require Photoshop. Perform the following steps in the listed order.

Step 3:Sharpen
One of the most useful tools Photoshop offers is the ability to enhance the differences in luminosity between related pixels. The Filter-Sharpen-Unsharp Mask enhances the apparent sharpness of edges between shapes in the image. This appears to sharpen the focus of the image, enhancing the level of detail by reducing blur. Every image can benefit from the gentle application of Unsharp Mask.

Judiciously applied, the unsharp mask substantially enhances the crispness of the image. The best values for the unsharp parameters are somewhat a matter of trial and error. For detailed three megapixel images where the bird occupies a large portion of the frame (and the camera's sharpening option was turned OFF), typical values are shown in the window above. Use the preview function (click the checkbox on and off) to preview the effect of your settings before clicking "OK". It's important not to overdo it. Too much sharpening will enhance the noise or graininess of the image, especially in dark areas. Further sharpening will lend an artificial look to the image. Sharpening only enhances the detail that exists in the image--it can't create detail where it didn't exist in the original.

Unsharp mask should be applied only once. If you later decide you've applied the wrong amount of sharpening, start over with the raw image.

Step 4: Levels
One of the penalties of photography through a spotting scope is the slow photographic speed of the system; combined with the desirability of fast shutter speeds (to freeze motion) and the noise penalty of high ISO (CCD gain), images tend to be light-starved.

Fortunately, Photoshop is extremely effective for adjusting image brightness and contrast. The simplest way to do this is using the Image-Adjust-Levelscontrols. Upon selecting this function you will see a histogram which graphs the intensity level of each of the 256 ( 0 through 255, black to white) gray tones that comprise the image. This is the same histogram that a Coolpix camera displays during selected playback modes. The "RGB" channel combines the individual red, green, and blue channels.

You will adjust the input levels by moving the three triangular sliders at the bottom of the histogram. This will change the tonal balance of your image. First, adjust the contrast of your image by using the shadow (leftmost) and highlight (rightmost) sliders. Move the shadow slider to the right until its level's value is non-zero, then three more units; if the first value is non-zero, move the slider to level 3. The effect will be to deepen the black tones, giving the impression of removing a light haze. Then move the highlight slider to the left until the brightest spot on your image is as bright as you want it to be while retaining detail. This sets the maximum white level.

Now, if necessary, adjust the overall brightness of the image by moving the midtone (center) slider left, if the image requires lightening, or right if the image requires darkening. This slider adjusts the center of the tone scale without changing shadow and highlight values; it allows you to lighten or darken the image without loss of detail in the shadows and highlights. The midtone value that gives a pleasing image will usually be in the range 1.25 to 0.90.

Step 5: Color Balance
If your monitor's color balance is set properly, you set your camera's color balance correctly for the light (or to "auto"), and you're lucky, the image will look perfect at this point. However, you might notice that the image has a slight green or yellow cast to it. Look carefully at foliage in your image--do the leaves look a bit sickly? It's always worthwhile to take a long eyes-only look at your subject so that you remember how the colors should look. The Coolpix 990 seems a bit strong in the green channel, so this is the channel you may need to darken using Levels.

From the "Channel" option on the Levels dialog box, select the "Green" channel. Now slide the midtone (center) slider left slightly--decreasing the green content of the image--until the image color looks correct. Check and uncheck the "Preview" box to compare the before and after image; click "OK" when you're satisfied with the result.

Step 6: Dodge and Burn
Bird images often have strong localized highlights or deep localized shadows. Once the overall tonal range of the image is set with the levels controls (step 4), it is useful to make local adjustments to these areas. Burning, represented by the thumb and finger circle (a hole to let light onto the print paper in a traditional darkroom), lets you darken an area defined by the "Brushes" palette. To burn, activate the "Tool Options" control, select "Midtones", and set an exposure of about 20-30%. Hold the mouse button down and move the brush (circle) over the area you wish to darken; choose a gradient brush (fuzzy edges) of size appropriate to this area. Dodging, represented by the lollipop-shaped paddle, operates analogously to lighten a shadow. Both effects can be overdone, so make frequent use of the "Edit-undo" option (Ctrl-Z or Cmd-Z) to see the effect on the image. (Note: undo undoes the effect of only the last mouse click in Photoshop LE.)

Step 7: Sizing and Saving
Now you need to make a decision about the presentation of your image. If you intend to archive and print this image you should crop and size it now for the intended print size. A detailed image will look reasonably good on a photo-quality inkjet printer when printed with a pixel density, measured in dots-per-inch (dpi), greater than 250. This implies a maximum print size of about 8 inches on the long edge for a 3-megapixel image. A small print with the same number of pixels will always look better than a big print because the pixel density, or resolution, is higher.

If you want to archive the image now but manipulate or print it later, save it now in either Photoshop or TIFF format; this assures that your file will retain the original image resolution and level of detail. DON'T overwrite your original image to save your work! Use File-Save As... and choose a naming convention that allows you to distinguish between original camera files and manipulated images. For example, if the original file name is DSCN0001.JPG, you might save the adjusted image as Da0001.PSD (Photoshop format) for archiving.

Use the Image-Image Size dialog box to set the size and density. Making sure the "Constrain Proportions" box is checked and the "Resample Image" box is UNchecked, set the larger of the Width or Height to 8 inches (or smaller). Note that a smaller image has a correspondingly larger resolution; if Resampling is NOT enabled, the file will retain all the original detail. "Pixel Dimensions" is the size of the image in pixels--width times height, times three (for three color channels), not the size of the file in disk space. Saved Photoshop or TIFF files will be at least the same size as the pixel dimensions.

If you intend to post your image on the Web or email it, remember that the resolution of a computer monitor is far lower than that of a printed page; a screen image holds much less detail than a same-sized print. This means your images can be reduced in both image size and file size without apparent degradation. A useful image size for screen display has a long edge of about 500 pixels (about 7 inches). Standard screen pixel density is 72 dpi, so the pixel density of the image should also be 72 dpi.

In the Image Size dialog, check the "Resample Images" checkbox--this lets you set resolution independent of image size--and set the Resolution to 72 pixels per inch. Now set the Print Size: Width or Height to the desired size and click OK. Note that the on-screen image will get smaller; use the Magnifying Glass to enlarge it to 100% size. If you need a different size, use Undo to undo the size change and try again. Keep in mind that larger images produce larger files, which are slower to download; the file size will scale with the area (width times height) of the image.

For Web or email use, save the file in JPEG format, using File-Save As.... Again: don't overwrite the original image! You might choose Dw0001.JPG as a naming convention to distinguish this file from an original camera file. JPEG compression produces much smaller image files than Photoshop or TIFF by reducing detail, so it should not be used for archiving your adjusted image (even if the original is in JPEG format!). When the File-Save As-JPEG Options dialog box pops up, choose a Quality value no greater than 7 (higher values produce virtually identical results) and choose Progressive Scans (allows your image to download in stages) if that option is available. For reference the "original" and "final" images below are 500 by 332 pixels (72 dpi), and about 35 kB each. Click on one to decide if the download speed is "friendly". It would take about 200 times longer to download this file in Photoshop or TIFF format at full resolution (~7 MB), and 25 times longer to download the original (~1 MB) camera file.

Here are several examples of an image at different levels of JPEG compression. Look carefully to see the differences, especially around the eye. Compression artifacts are most pronounced at edges and bright color changes. These are 225 by 150 pixel, 72 dpi images; uncompressed file size would be 101 kB. Highly detailed images generate larger JPEG files for a given setting.

JPG=2, 11 kB

JPG=5, 14 kB

JPG=7, 21 kB

JPG=10, 51 kB

This Primer should be considered just a very quick overview of some of the image processing capabilities of Photoshop; you should use these guidelines as a starting point for your own experiments. Practicing your Photoshop skills will improve them and give you greater confidence in shooting more challenging situations, since you will better appreciate the wonders you can produce in your "darkroom".

Many capabilities of Photoshop are beyond the scope of this Primer. For excellent discussions of more powerful Photoshop techniques, Digibird recommends:

Click here for the Digiscoping Technique Primer.


All text and images copyright Digibird.com; unauthorized use or reproduction prohibited.
Back to Digibird home