TechTalk

WHAT IS A BITMAPPED GRAPHIC?
Bitmapped, or raster files are used for digital photos and scans. Unlike vector graphics which are composed of resolution-independent lines (see Part II of this series, July 2009 S&DG), bitmapped files are composed of microscopic picture elements (pixels) that are defined (“mapped”) by their vertical and horizontal position. The “bit” in bitmapped refers to the depth of the pixels, in gray or color levels.

Bitmapped images are purpose-specific, meaning they must be created for the final output size. For example, a digital photo must be taken at a specific resolution, in pixels per inch, or megapixels, whether it’s used for a photo album or a poster.

When printed, bitmapped and vector images are both converted to a device-specific bitmap commonly called a “raster” file. “Raster” refers to the imaging head’s or laser’s back-and-forth scanning pattern over the media or image carrier.

Figure 1: Bitmapped images are “purpose-specific,” meaning the image’s resolution has to be determined in advance based on the output size and print resolution. To illustrate, a 300-ppi test photo (left) was downsampled to an unrealistically low 18 ppi (right) to show the prominent “stair-stepping” or “jaggies” of too-low resolution bitmapped images.

SIZE DOES MATTER
When it comes to bitmapped images, size does matter. Bitmapped images must be sized specifically to the output, including the printer’s resolution and the viewing distance. If too small in size, the normally microscopic pixels will be visible, and the image will have a “stair-stepped” or “jaggies” effect (see Figure 1 at right ). If too large, quality won’t improve, but file will be much larger than necessary and output will be slower.

Size is measured by the number of pixels per linear inch (ppi), or less correctly, as dots per inch (dpi). For example, inkjet printing at normal viewing distance typically requires 200 ppi, while lithographic printing requires 300 ppi.

Today’s most common source of bitmapped images is digital cameras, whose resolution is measured in millions of pixels (megapixels, or MP). The megapixel measurement includes the horizontal and vertical dimensions. When choosing a digital camera to capture images for print, it’s helpful to calculate the number of megapixels you’ll need for a certain sized print. For example, a 10-MP digital camera can produce a 16" x 16" image at 200 ppi.

Size for Lithography — Lithography, used for printing large quantities of brochures, magazines, and catalogs — is done with microscopic halftone dots measured in lines per inch (lpi). The lpi will vary with paper quality, and may include 85–100 lpi for newspapers, 133–150 lpi for magazines, and 150–175 lpi for commercial printing.

Size for Digital Printing — Inkjet and toner-based printers, along with newer forms of lithographic plate-making, use random-dot patterns known as stochastic screening. The fixed size of stochastic dots generally requires 200 ppi images.

Size for Viewing Distance — The guidelines in the accompanying chart apply to normal viewing distances (generally 14" for printed material). Many inkjet products, such as posters or banners, are viewed from further away, so they don’t require such high resolution. Technically, the viewing distance is considered to be the diagonal of the product. So a 3' x 5' poster would be about 6' (see Table).

FILE FORMATS
Bitmapped images can be rather large, and so can be compressed for easier handling. Compressed files include “lossless” formats such as ZIP that preserve all the original data, and “lossy” formats such as JPEG (Joint Photographic Experts Group) that discard the least-noticeable picture data.

The highest-quality file format is TIFF (Tagged Image File Format), originally developed by Aldus Corp. (now part of Adobe) for its PageMaker page-layout program (now succeeded by InDesign). However, JPEG can reduce file sizes to one-tenth the original size with no noticeable loss in quality.

WORKING WITH CUSTOMER IMAGES
More often than not, customers will ask you to resize an image for printing. You have two options: scale or resample. Here are the differences.

Scaling is simply adjusting the size at which the image will print. Scaling up increases the space between the pixels to make the image larger, resulting in decreased resolution. Scaling down decreases the image size by squeezing the pixels together, therefore increasing resolution. Neither scenario changes the number of pixels or the file size.

On the other hand, resampling involves changing the image’s pixel dimensions. Since the number of pixels changes, so does the file size. It is called resampling because the graphics program must take samples of the tones and colors in an attempt to maintain the image’s appearance.

Figure 2: When opening customer files of unknown resolution, you can use Photoshop’s Image > Image Size dialog box as a size calculator. Uncheck “Resample,” enter the desired output resolution (such as 200 ppi for inkjet at reading distance), and note the image’s dimensions. If a larger-size images is needed, the file would need to be upsampled using interpolation, but the quality would not be the same as if the original photo was taken at the correct resolution.

To reduce file size, you must “res down” (also called downsampling). If you res down an image you can speed up file transmission and RIP time, but you may also soften sharp edges within the image. This occurs because pixel pairs along changes in color must be combined into one mixed-color pixel. For example, an image of a yellow sun in a blue sky has yellow and blue pixels next to each other. If you res down, these two bordering pixels may get merged into one green pixel. This will make the transition between the sun and sky appear less sharp.

If you “res up” a file you will increase the resolution of the image but you will not change its dimensions. Also called upsampling, this process adds more pixels per inch, but does not necessarily increase sharpness or detail. When you res up, the graphics application creates new pixels based on the color values of the existing ones. It’s only a guess. Generally, you can res up by about 1.5 percent and still retain quality. If you need to res up higher, use fractal imaging software instead.

Photoshop is an excellent tool for scaling and resampling images because it has a built-in algorithm. Select Image > Image Size to bring up a dialog box that will guide you through the process of resampling to increase or decrease an image’s resolution. This dialog box can also be used as a tool for calculating the percent or size to which you can enlarge a customer file and still retain the required resolution (see Figure 2).

Figure 3: Photoshop’s Curves dialog box enables you to set the image’s highlight and shadow to the printing system’s smallest and largest printable dots, respectively, using the white and black eyedropper tools (double-click them to set dot size). A printed image’s maximum contrast is actually determined by the print media and not the image, so you have to work backward from the media to optimize the file for printing.

IMAGE QUALITY
Images attract and inform. Viewers like images with good contrast and detail. Unfortunately, ink and paper, or computer monitors, can’t usually duplicate the contrast range of the original scene. “Tone compression” is inevitable; and there may be a tradeoff between contrast and detail.

After electronic scanning greatly accelerated the making of color separations from hours to minutes, technologists developed an entire scheme of ensuring maximum image quality.

The essence of printed image quality is to recognize that each image has a lightest and darkest point, called the highlight and shadow, respectively. These points, measured in density (on a transparency or print) or in RGB levels (in a digital file), should be set to reproduce using the printing system’s smallest and largest printable dots. A printed image’s maximum contrast is actually determined by the print media. By setting the highlight and shadow, you are actually working backwards from the print media to the image file. These settings can be made with most scanning programs and with Photoshop’s Image > Adjust > Curves dialog box (see Figure 3).

The “curve” shows the relationship between light and dark areas of the original file and its reproduction. The middle point, or midtone, can be used to lighten or darken the image and to trade off contrast between highlights and shadows. “Three-point control” refers to these settings for highlight, shadow, and midtone.

Click here to read Part 4 of this article series.