Digital Glossary

by Joe Farace

Digital imaging is the next, logical extension of the same type of camera and darkroom techniques that photographers have been using for 150 years. The tools used for digital imaging may appear to be different, but they are no more of a radical change than what late 19th century daguerreotypists encountered when they switched from silver plated copper sheets to glass plate negatives.

Like other disciplines, photography has its share of jargon. As photographers we "shoot" brides and kids, "bracket" these shots, and apply "Zone IV" interpretations to an image on "sheet" film that's processed using "N+1" development. Part of the problem of understanding digital imaging concepts--like much of computing--boils down to semantics. Digital imaging uses a combination of buzzwords that have been borrowed from the printing, design, and photographic fields, liberally blended with a sprinkling of computer jargon.

Joe Farace's book, The Digital Imaging Dictionary, is published by Allworth Press. It is available in bookstores or order directly from the publisher. You can contact them at (800) 247-6553 or their web site at: www.allworth.com

Here's an abbreviated look at some common digital imaging terms.

Access Time: The performance of fixed or removable drives is measured by seek time or the amount of time required for the arm of a direct access storage device to position itself over the track where the required information is stored on the disc.

ADB: Apple Desktop Bus. The Mac OS communications port that is used for keyboards, mice, trackballs, graphics tablets, joysticks, and other input devices.

Alpha Channel: This is a separate image channel that is added to the traditional red, green, and blue channels and contains a value that indicates the amount of transparency for each pixel.

Analog: Information presented in continuous form, corresponding to a representation of the "real world." A traditional photographic print is an analog form, but when this same image is scanned and converted into digital form, it is made up of bits.

Archive: To copy any kind of data from the media it is currently stored on (typically a hard disk) onto a removable media cartridge or tape for back-up purposes. Archive and back-up software often compress the data to maximize the capacity of that storage media.

ASCII: American Standard Code for Information Interchange. ASCII is a standardized computer code for representing text data. The code has 96 displayed characters (characters you see on the screen) and 32 non-displayed characters (some of which you can see, others that you can't).

Binary: A mathematical system based on the numbers one and zero. This is ideal for computers, because electrical signals can be represented by electrical current being positive and negative, on and off.

Bit: Binary digit. The smallest unit of information with which a computer can work. Computers are digital devices because they represent all data--including photographs--using numbers, or digits, that are measured in bits.

Bit Depth: This refers to the number of bits that are assigned to each pixel in an image. The more bits you have, the more photo-realistic the screen image will be. Let's look at the typical choices in bit depth for computer screens:

1-bit: If a computer has the ability to display 1-bit per pixel, each pixel can either be black or white.

4-bit: Some computers, especially laptops, offer 4-bit video capability, which translates into 16 shades of gray or color.

8-bit: With an 8-bit color depth, you can see 256 colors or levels of gray. An 8-bit system can work well for black and white photographs, but is just barely adequate for critical evaluation of a color photograph.

16-bit: This bit depth has the potential to display 32,000 different colors. At 16-bits and above, the video signal must split into thirds, providing one each for the red, blue, and green channels. Your computer devotes 15 bits to color (5 bits per color channel) and the one remaining bit is used to overlay all these colors.

24-bit: Each pixel on a screen can handle up to 256 colors, which lets systems display 16.7 million colors. A 24-bit model provides true photographic quality.

32-bit: Often when someone is talking about 32-bit color, they really mean 24-bit. Only a few computers offer 32-bit capabilities.

Since 256 levels of gray are displayed on an 8-bit system, that's all you really need when working with black and white digital photographs, but if you plan on working with color images, you should use a computer that has a 24-bit display.

Bitmap: There are three classes of graphic files: bitmap, metafile, and vector. A bitmap (sometimes known as "raster") is any graphic image composed of a collection of tiny individual dots or pixels--one for every point or dot on a computer screen.

BMP: Short for bitmap. A Win-dows based digital image format, BMP--often pronounced "bump"-- is really a file extension for a specific kind of bitmapped graphics file.

Bundle: Part of a software or hardware package, which may or may not be a good deal. When you purchase a product that's packaged with a product from another company at no additional cost, the extra product is considered to be "bundled" with the main product.

Byte: Each electronic signal is one bit, but to represent more complex numbers or images, computers combine these signals into larger 8-bit groups called bytes. When 1024 (not 1000) bytes are combined, you get a kilobyte, often called K.

Calibration: A term used in Color Management Systems (CMS). Calibration stabilizes the inevitable variables in the way devices reproduces color. To produce optimum results, all color-reproducing devices must maintain a consistent, calibrated state.

CCD: A Charge-Coupled Device is the same kind of light gathering device used in flat-bed scanners, digital cameras, and even video camcorders to convert the light passing through the lens into an electronic equivalent of the original image. These images become digitized by the CCD device.

CD-ROM: Compact Disc Read Only Memory. This is a disc that resembles a musical compact disk but can contains all kinds of data--including photographs.

CD-R: Compact Disc Recordable

CD-RW: Compact Disk Re-Writable

CEPS: Color Electronic Publishing Systems

Characterization: A Color Man-agement System (CMS) term that establishes the relationship of your calibrated device to what is referred to as a device independent Reference Color Space or RCS.

Clipping Path: A path is a series of line segments connected by endpoints. Paths are created by the pen tool and can be reshaped and moved--even exported to other files or programs. Clipping paths silhouette an area to mask out the background, so only that part of the images within the path appears when the image is placed in another program or combined with another image file.

CLUT: Color Look Up Table. A table--it can be in hardware or software form--that contains information on the mixing of red, green and blue color intensity in a palette.

Compression: Compression is a method of removing unneeded data to make a file smaller without losing any data, or in the case of a photographic file, image quality. There are many techniques and technologies for compressing graphics and how well each works depends on what is more important to you: file size or image quality.

CPU: Central Processing Unit. There are two basic families of CPUs: Intel and Motorola. Intel and similar chips are used in IBM-compatible PCs, while Motorola makes the chips for Mac OS machines. How well a chip processes data is determined by how many bits of information it can process at one time. The larger the number of bits a chip can process simultaneously, the faster it can process.

DCS: Desktop Color Separation. The QuarkXPress format for defining color separated output using a personal computer.

DCT: Discrete Cosine Transform. This is an algorithm that converts data--including pixels--into sets of frequencies. The first frequencies appearing in the set are the most meaningful; the latter, are the least. For compression purposes, such as JPEG and MPEG, the latter frequencies are removed on the basis on allowable resolution loss.

Despeckle: A feature of image-editing programs, and some plug-ins (e.g. Extensis Intellihance) that detects the edges of an image--the place where significant color changes occur--and blurs a selection except the edges. This has the effect of removing noise while preserving detail.

Device Resolution: Refers to the number of dots per inch (dpi) that any given device, such as a monitor or printer, can produce. Screen resolution for computer monitors varies from 60-120dpi. Don't confuse this with line screen, which refers to the number of lines per inch (lpi) in the screen used by printers to reproduce a photograph.

Dither: A graphics display or printing process that uses a combination of dots or textures to create the impression of a continuous tone gray scale or color image.

DOS: Disc Operating System. The official name of the Microsoft Disk Operating System is MS-DOS but is often referred to as simply DOS. The IBM-PC version is called PC-DOS, or just DOS.

DRAM: Dynamic Random Access Memory

Dynamic Range: One key feature to look for in a scanner or digital camera is its dynamic range. A scanner's dynamic range depends on the maximum optical density that can be achieved and the number of bits captured. In simple terms, the greater the density range the better the scanner.

DVD: Digital Versatile Disc, sometimes referred to as Digital Video Disc.

EDO DRAM: Extended Data Out-put DRAM is popular in Windows based computers. EDO DRAM sends our data--even if a controller loads more data in vacant addresses.

EMF: Electromagnetic Field. If you spend more than a few hours at your computer a day you should be aware of the potential problems caused by the electromagnetic fields that computer monitors produce. All monitors emit some kind of Very Low Fre-quency (VLF) and Extremely Low Frequency (ELF) radiation, and color monitors emit more than monochromatic ones.

FAQ: Frequently Asked Questions. A term, often found on web home pages, that will lead you to a page containing the most frequently asked questions visitors to the web site may have.

FPM DRAM: Fast Page Mode DRAM is the most common type that is used in Mac OS systems because it speeds the overall readout of data but does introduce a wait state.

Gamma: All photographs have a characteristic called gamma. The amount of gamma present in an image is measured as the contrast that affects the mid-level grays (the mid tones) of an image. The good news for digital imagers is that this gamma is adjustable by most image-enhancement programs, and you aren't stuck with the gamma that is present in the original negative or print.

Gamut: Every output device (e.g., a printer or monitor) has a range of colors that it can accurately reproduce. This range is called the gamut of the device. Every device from every manufacturer, whether it is a monitor or printer, has a unique gamut.

Gaussian Blur: Photoshop's blurring filter gets its name from the fact that it maps revised pixel color values according to a Gaussian curve. A Gaussian curve is typically used to represent a normal or statistically probable outcome for a random distribution of events and is often shown as a bell shaped curve.

GIF: (Pronounced like the peanut butter.) Graphic Interchange Format developed by CompuServe is completely platform independent: the same bitmapped file created on a Mac OS computer is readable by a Windows graphics program. A 256 color GIF file is automatically compressed making it ideal for use on the World Wide Web.

Gigabyte: A billion bytes or (more correctly) 1024 megabytes

Gray Scale: Refers to a series of gray tones ranging from white to pure black. The more shades or levels of gray, the more accurately an image will look like a full-toned black and white photograph. Most scanners will scan from 16-256 gray levels. A gray scale image file is typically 1/3 the size of a color one.

GUI: Graphic User Interface

HTML: HyperText Markup Lan-guage, a format used on World Wide Web home pages on the Internet that uses multimedia techniques to make the web easy to browse.

Image Resolution: Refers to the amount of information stored in a photograph and is typically expressed in pixels or dots per inch (dpi). The image resolution of a photograph determines how big the file is. The important thing to remember is that the higher the image resolution, the more disk space it takes and the longer it will take to print or image.

Ink Jet: In an ink jet printer, a print head sprays one or more colors of ink onto paper to produce output, and the type of methods used to accomplish this has an effect on output quality.

Interlaced GIF: (See GIF.) The interlaced version stores rows of bitmapped data out of order: All of the even numbered rows are stored first and all of the odd rows are stored last. Interlaced GIF uses a four pass interlacing. The first pass starts at row zero and reads every eighth row of bitmap information. The second pass starts on the fourth row and reads every fourth row. The third pass starts on the second row and reads every second row while the last pass starts with the first row and reads every second row.

ISP: Internet Service Providers. Companies that offer direct access to the Internet through local or toll free telephone connections. This includes companies such as CompuServe, America Online, as well as a plethora of local operations that often provide service at a flat ("all you can eat") rate.

JPEG: An acronym for a compressed graphics format created by the Joint Photographic Experts Group, within the International Standards Organization. JPEG achieves compression by breaking an image into discrete blocks of pixels, which are then divided in half until a compression ratio of 10:1-100:1 is achieved. The greater the compression ratio, the greater loss of sharpness you can expect.

K: In the computer world, K stands for two to the 10th power, or 1024. A Kilobyte (or KB) is, therefore, not 1000 bytes but is 1024 bytes.

Layer: In image enhancement programs like Adobe Photoshop, layers are any one of several on-screen independent levels for creating separate, but cumulative, effects for an individual photograph. Layers can be manipulated independently and the sum of all the individual effects on each layer make up what you see as the final image.

LUT: Color Look Up Table. A table--it can be in hardware or software form--that contains information on the mixing of red, green, and blue color intensity in a palette. A LUT is a two-dimensional table with input values (X axis) and output values (Y axis.) By default, LUT are neutral, which means that the overall input values must equal the overall output values.

Magneto-Optical: This class of removable drives uses the ability of lasers to heat material and thus change reflectivity to produce media that can be erased and reused. One of the negative aspects of optical drives is that writing data to optical media requires three spins of the disc. The first erases existing data, the second writes new data, and the third verifies that the data is there.

Mask: Many image-enhancement programs have the ability to create masks--or stencils--that are placed over the original image to protect parts of it and allow other sections to be edited or enhanced. Cutouts or openings in the mask make the unmasked portions of the image accessible for manipulation while the mask protects the rest.

Megabyte: When you combine 1024 kilobytes, you have a megabyte (MB) or "meg."

Moiré: (pronounced "mwah-RAY") Moiré patterns are an optical illusion caused by a conflict with the way the dots in an image are scanned and then printed. When scanning an original photograph or artwork, a single pass scanner is all most people require, but when scanning material that has previously been printed, a three pass scan (one each for red, green, and blue) will almost always remove the inevitable moiré or dot pattern.

Photo CD: Kodak's proprietary process that places digitized files of photographs onto a CD-ROM disc. Photo CD facilities can digitize images from color slides and black and white or color negatives. A Photo CD transfer station converts your analog images into digital form by using a high-resolution film scanner, a computer, image processing software, a disc writer, and a color thermal printer. Each image is adjusted for color and density, compressed to 4.5MB using Kodak's Photo YCC format image, and written to a CD-ROM. A thermal printer creates an index sheet showing the images transferred to disc and is inserted into the cover of the CD's jewel case.

A subset of this disc type is the Pro Photo CD Master Disc which accept images from 120 and 70mm rolls and from 4x5 sheet film. The Pro disc includes built-in copyright protection and offers a sixth, Base*64 resolution image that yields a size of 6144x4096 pixels.

PICT: An acronym, without a strict definition, for a metafile file format. PICT files contain bitmapped or object oriented information. Some people love PICT files because they are excellent for importing and printing black and white graphics, like logos. Others hate them because they don't always retain all of the information in the original image.

Pixel: Pixel is an acronym for picture element. A computer's screen is made up of thousands of these colored dots of light that, when combined, can produce a photographic image. A digital photograph's resolution, or visual quality, is measured by the width and height of the image measured in pixels.

PostScript: A programming language created by Adobe Systems that defines all of the shapes in a file as outlines, and describes these outlines using mathematical formulae called Bezier curves. Any PostScript compatible output device uses those definitions to reproduce the image on your computer screen.

PMT: Photomultiplier Tubes are a type of sensing technology used in drum scanners.

PNG: Portable Network Graphics (pronounced "ping"). When Compu-Serve and UNISYS started charging royalty fees for the compression algorithm in the formerly free GIF file format, a group of independent graphics developers formed a coalition to create an improved royalty free file format. PNG is a flexible and open format for storing bitmapped graphics images and is the result of their efforts.

PPI: Pixels Per Inch (see Device Resolution)

Q-Table: Quantization table. The JPEG format is a "lossy" one, in which some image information is discarded. To determine the appropriate amount of information to discard, the compression algorithm divides the Discrete Cosine Transform (See DCT) by a quantization coefficient and rounds the results to an integer. The larger the coefficient, the more data is lost.

Resolution: A digital photograph's resolution, or image quality, is measured by the image's width and height in pixels. When a slide or negative is converted from silver grain into pixels, the resulting digital image can be made at different resolutions. The higher the resolution of an image--the more pixels it has--the better the visual quality. An image with a resolution of 3072x2048 pixels has better resolution and more photographic quality than the same image digitized at 192x128 pixels.

RGB: Red, Green, Blue. Color monitors use red, blue, and green light to produce all of the colors that you see on the screen. The concept is built around the way these three colors of light blend together to produce all visible colors.

SDRAM: Synchronous DRAM. Used as a substitute for Video RAM (VRAM.)

Search Engine: Since the actual number of web sites on the World Wide Web is big and getting bigger every day, finding the Exacta Collectors home page might be impossible without a way to search for the word "Exacta."

Seek Time: The performance of fixed or removable drives is measured by seek time, or the amount of time required for the arm of a direct access storage device to be positioned over the appropriate track.

SRAM: Static Random Access Memory typically is used in caches.

TARGA: Raster Graphics Adapter, a raster graphics file format developed by TrueVision, Inc. for its line of (originally) PC based video graphics boards used for high-resolution digital imaging. Targa files are identified by a .TGA file extension and handle 16, 24, and 32-bit color information.

Thermal Dye Transfer: Often called dye sublimation. A printer type utilizing a head that heats a dye ribbon to create a gas that hardens into a deposit on the special paper used by the printer.

TIF, TIFF: Tagged Image File Format is a bitmapped file format developed by Microsoft and Aldus. A TIFF file (.TIF is the extension used in Windows) can be any resolution from black and white up to 24-bit color. TIFFs are supposed to be platform independent files, so files created on your Macintosh can (almost) always be read by a Windows graphics program.

Transfer Rate: A measurement of the average number of bytes per unit of time passing between disk storage and the computer's memory.

TrueType: A successful attempt by Apple and Microsoft to introduce scalable font technology into the Mac OS and Windows environments. TrueType fonts look good on-screen and when output from a PostScript or non-PostScript printer. Unlike PostScript fonts, TrueType fonts have only one component and a slightly different icon so you won't get them confused with bitmapped fonts.

TWAIN: According to "Digital Dave" in ComputorEdge magazine, TWAIN means Technology Without An Interesting Name. TWAIN is a hardware/software standard that allows users to access scanners or other image-capture devices, such as a video frame grabber, from inside Windows applications.

UC: Upload Control. Part of the FilmWorksNet process for uploading image to a Personal Home Page. This file manages the interconnections, associations, and protocols between your computer and Seattle Film-Works and is what makes the FilmWorksNet uploading process work correctly.

Unsharp Mask: In Photoshop and other image-editing programs, this is a digital implementation of a traditional darkroom and prepress technique in which a blurred film negative is combined with the original to highlight the photograph's edges. In digital form, it's a more controllable method for sharpening an image than standard sharpening commands.

URL: Universal Resource Locator. The technical term for the location of a home page on the World Wide Web. For Example, the URL for the Joe Farace's web site is: http://www. hyperzine.com/writers/joef.html.

Virtual Memory: Sometimes called (by Adobe in Photoshop) a scratch disk. When not enough "real" memory (RAM) is available, this process borrows a chunk of your hard disk to store data and perform imaging calculations.

WMF: Windows Metafile Format. A vector graphics format designed to be portable from one PC-based program to another.

WORM: Write Once Read Many times

World Wide Web: One of the most popular aspects of the Internet, the World Wide Web has a set of defined conventions for publishing and accessing information using Hyper-Text and multimedia.

WYSIWYG: (Pronounced "wissy-wig.") What You See Is What You Get. This term refers to the ability to view text and graphics on-screen in the same manner as they will appear when printed.

YCC: The color model used by Kodak in its Photo CD process. This involves the translation of data that was originally in RGB form into another system using what scientists call "luminance" but the rest of us call brightness (the Y component) and two parts (the CC) of color and hue.

ZIP: Hardware: Iomega's Zip removable media drive uses a combination of conventional hard disk read/write heads with flexible disks that produces an average seek time of 29ms. Pocketable Zip cartridges, scarcely larger than a floppy disk, are less expensive than other removable media.