I have read many articles and posts on the Web on how to scan photographs. There are many opinions as to the best way to perform the scan, and I have my own. This post will explain the most common settings that I use when scanning photographs.
Pixels Per Inch (PPI)
There is some debate about how many pixels per inch to use when scanning a photograph. Some say 240ppi, others 300 ppi, and a few say more than 300ppi. My take on the situation is this: I try to aim for 300ppi. What do I mean by this? I’ll show you.
Lets say you have a 4×6 photograph that you would like scanned in, and printed at 4×6. This is rather simple to do, simply specify 300 in your scanner software where it says PPI and scan the photo. When completed, you will get a photo that is 1200×1800 (4×300, 6×300) pixels in size.
What if you want to increase the size of the photo to 8×10, and still print at 300ppi? You would increase the PPI in the scanner software to allow for a bigger print. In this case 600ppi would double the size of the print to 2400×3600 (4×600, 6×300), which allows a print of 300ppi at 8×12 inches.
Most of the time, however, I would print the same size of picture that I scan in. This means that I would keep the PPI at 300.
48bit Colour
I always scan photos in at 48bit colour. I usually need to correct colour casts in the scanned photos, and 48bits provides me that extra overhead to change the colours of the photograph without any banding. Banding can be seen when using Adobe Photoshop using the Levels tools to view the distribution of pixels in the different tonal values. If there are gaps between the lines, then it shows that those tonal values don’t have any pixels, otherwise known as banding. One of the downsides to using 48bit colour instead of 24bit colour is file size. The size of the photos is double at 48bit, meaning more memory and hard drive space is needed. With the large hard drives and 1-2GB of memory in most systems these days, most people would have the space. Another downside to 48bit is that not all photos-editing software can edit photos at that colour depth. Many of the big name editors, such as Adobe Photoshop, are providing more functionality for editing 48bit images.
Saving
When saving photos for later use, I like to use the TIFF (Tagged Image File Format) format. This file format is uncompressed, can be used on many types of systems (PC, Mac, Unix), and supports layers and 48bits. The downside is that it can be much larger than a compressed format, such as JPEG. Since it is uncompressed, there is no data loss, so you can continuously edit and save the file with any image degradation.
When sending the photo through e-mail or posting it online, I use JPEG (Joint Photographic Experts Group). This is the same format used by digital cameras, and can compress to a much smaller size than TIFF. The big downside to this format is that it uses a lossy compression, meaning every time the picture is saved, it is compressed and some information is lost. The compression can be adjusted to help balance between size and image quality.
Summary
Here is a summary of what I mentioned in the post to make things easier to remember:
- I scan the photos in at 300ppi if I will printing the same size as the scanned picture. I will increase the PPI only if I’m planning to make a larger print.
- I scan the photos in at 48bit colour depth to provide extra headroom when editing the photo, such as removing colour casts.
- When editing photos, I save the photos in the TIFF file format.
- When sending through e-mail, or posting to the Web, I will compress the photos using the JPEG format.
In later posts I’ll describe how I catalogue and archive my digital photos, both scanned and those taken with my digital camera.