Understanding the jargon

When buying, configuring and using your scanner, you'll run across some specialized words, phrases and acronyms. You don't have to speak "geek" to get the most out of your scanner, but being familiar with the following terms makes it easier to understand how to use your scanner. 

Resolution is the ability of a scanner to resolve detail. It refers to the number of pixels per inch in an image when you scan. It may also refer to the number of dpi (dots per inch) used by an output device such as a printer. The higher a scan resolution is, the better the quality of that image. However, that also means that the file is larger. Conversely, the lower the resolution is, the lower the quality of the image, and the smaller the file size.

Resolution is not really as big a deal as it's been made out to be. 200 dpi is usually sufficient for most images if you want to print them at original size. For very high quality images with a lot of detail, 300 dpi is better. But scanning at very high resolution -- say 600 to 1200 dpi -- is only useful if you intend to enlarge the image. Each time you double the resolution of a scan, you quadruple the file size. Therefore, a 1200 dpi file is 36 times larger than a 200 dpi file.

DPI is another way to refer to the resolution of an image. Essentially, it's the measurement of the number of pixels per square inch in an image. You've probably heard the term in relationship to printers: a determination of the number of dots of ink per inch that are printed. Here again the perception is that the greater the number, the higher the quality and file size. After you capture all of the detail in a scan, however, increasing the dpi only adds unnecessary bulk to the file. Though several factors determine the best dpi of an image, the overriding concern should be the final destination of your image. Images to be displayed on the Web are generally only 72 dpi, whereas images to be printed are often 300 dpi or larger.

OCR (optical character recognition) is the technology that allows you to convert scanned documents into editable text files. This allows you to convert printed material into electronic format without retyping or reentering the data.

Optical resolution is a function of scanner optics and interpolated resolution references a resolution derived by software enhancement. The process of interpolation essentially relies on the computer estimating pixel values around the known values in an image.  Most scanners deliver up to 4800 dpi optical or interpolated resolution. This process is also called upsampling or resampling in computer graphics applications, which requires adding pixels to an image that weren't there previously, and because adding information that doesn't exit in the original image, the computer does this by averaging and adding intermediary pixels. If you need to enlarge your image, you'll get better results by rescanning the original at a higher resolution. Interpolation can yield larger file sizes and can degrade image quality in some cases. When assessing your scanner's resolution capabilities, rely heavily on your optical resolution number to gauge its true abilities.

TWAIN is an agreed-upon industry standard that governs the interaction of scanners with the large numbers of software applications that support them. TWAIN compliance means you can access the scanner directly from particular applications, such as Adobe Photoshop, and eliminate the need to scan the file with one utility and edit or manipulate it with another.

TWAIN is not an acronym, although you might hear it referred to as meaning "technology without an interesting name." The name TWAIN was pulled from the famous phrase, "never the twain shall meet," and refers to two people or things being completely different or unsuitable for each other and unable to agree.

Short for picture element, a pixel is the smallest element a video screen or computer monitor can display and is the unit of measure often used for image size and resolution. The number of pixels (width and height) in an image defines its size, and the number of pixels in an inch defines the resolution of an image. For perspective, a 1" x 1" square picture at 72 dpi contains 5,184 pixels. At 300 dpi, the same image contains 90,000 pixels.

As the name suggests, bitmap refers to any image comprised of a mapped array of pixels. Bitmaps are also often referred to as rasterized images (or rasters for short). The alternative to raster is a vector image, created via illustration programs such as CorelDraw, CAD (computer-aided design), or Adobe Illustrator. Because of the way they're built (and their fixed resolution) scaling bitmap images is more difficult than scaling vector drawings, and the files are significantly larger. However, the file format allows you to do many things with bitmap images that you can't do with vector images, such as applying filters for unique effects liked blurred edges.

Resampling is the process of changing a bitmap image's size or resolution. It's also another term for interpolation.

A bit is the smallest piece of information that a computer will handle. Bit depth refers to the amount of information each pixel can carry. Higher bit depth translates into more color and more accurate color information in the digital image because more layers of information exist in the scan. As with other file information, the higher the bit depth, the larger the file. Bit depth is not just a reference for color images; it can also apply to grayscale images. The more bit depth you have, the more colors you have to work with. Just remember the increase in file size that comes with added bit depth.

Scanned files, especially when saved as images, can be very large. To help users efficiently manage and share files, programmers developed compression techniques that reduce file sizes. These compression techniques fall into two categories: lossy and lossless. JPEG and GIF are good examples of lossy compression, meaning that image quality is lost as the file size is reduced. Formats such as TIFF are lossless, which means image detail and quality is preserved but the file is compressed less.

© 2005 Hewlett-Packard Development Company, L.P.