Kodak 9500 manual Evaluating scanned images, Source image types, Scanned images, Printed Images

Evaluating scanned images

This section describes methods for evaluating digitally scanned

images. Evaluating scanned images requires an understanding of

how scanned images are created and the types of source

documents used to create these images. Image creation and

source image types are also discussed in this chapter, as well as

various evaluation criteria using a standard target as an example.

Scanned images

A scanned image is a digitized representation of a source image.

An image is digitized by viewing the source image and

determining a numeric value for each finite area (picture element

or pixel) of that image.

The size of a pixel is described in dots-per-inch (dpi). As dpi

increases, the ability to retain fine "spatial" detail from the source

document is increased.

The numeric value assigned to a pixel is based on the number of

bits-per-pixel. This value is assigned based on the tone of the

source document. A black pixel will have a different value than a

white pixel. When the number of gray shades increase, the

scanner’s ability to retain tonal information increases. For

example, 8 bits-per-pixel allows 256 shades (levels) of gray.

By increasing dpi and bits-per-pixel, image quality improves, but

the digitized image file size increases. In fact, most scanned

images are stored as a bi-tonal (single bit-per-pixel) images. A

balance must be struck between image quality and file size.

Source image types

There are many types of source documents. Three common

types of documents are described in this section. An individual

source document can contain more than one source image type.

For example, a printed business form may contain computer-

generated printing.

Printed Images

Printed images are those created by a printing press, ink-transfer

process. These images contain discrete levels of tone and space.

Discrete in tone means that there are a limited number of tone

levels (shades of color) in the images. Discrete in space means

that image elements are printed by placing ink at specific points.

These may be random or patterned locations. Examples of

printed images include business forms, newspapers and printed

manuals.

Font size and line width largely control image quality. The use of

halftone printing screens, which are printed using a series of

small dots to create lighter colors, can result in scanned image

degradation due to aliasing and moiré patterns. (The “Evaluation

criteria” section describes aliasing and moiré).

A-61122/A-61124 March 1999

E-21