PostScript Imaging of Bar Codes

PostScript is a vector based image description language. PostScript images are described by mathematical co-ordinates. When you save an EPS file, it uses the PostScript language.

The PostScript description of a bar code defines bars and spaces in very precise terms. No imaging device available today can approach the accuracy of this PostScript description.

Every print device is limited in the quality it can produce by its resolution. Virtually all computer printers produce images composed of dots. These dots form a grid known as a raster. Each dot (or cell) within the grid may be ON (printing) or OFF (not printing).

Text and images are constructed from a regular pattern of dots.

Obviously the higher the resolution, the better the image quality available. At present, the resolutions available on PostScript printers varies from 300 dots per inch (d.p.i.) on a standard laser printer to 3300 d.p.i. or beyond on the highest quality imagesetters.

Part of the reason for the success of PostScript is that fact that it can be implemented on a wide range of print devices of varying resolutions. The printer may use a Raster Image Processor or RIP to convert the precise mathematical PostScript description into a raster image which fits the available resolution on the printer, or the image may be rasterised by software on the host computer.

Printing inevitably results in a loss of quality. Therefore, for the successful printing of bar codes, the resolution of the print device must be borne in mind. There is no hard and fast rule for the resolution that is required for bar codes as a number of factors must be considered:

1. The bar code symbology.

Each symbology has a different technical specification for the tolerance that is allowed for bar and space measurements.

2. The magnification (or size) of the bar code.

Generally speaking, the greater the magnification, the greater the tolerance that is allowed.

3. The end-use of the image.

If the image produced by the printer is the final image, such as with self-adhesive labels printed directly through a laser printer, a lower resolution will be sufficient than that which is necessary for an image which will be used for platemaking (and therefore suffer degradation in quality through platemaking and through the end print process).

4. The reliability of the print device.

In practice this is not often problematical. Some print devices have inconsistent paper transport mechanisms which will produce errors in the image in the direction of paper/film movement and so will only produce further error when the bar code is printed with the bars at 90 degrees to the feed direction.

With so many variable factors, it is impossible to offer clearly defined guidelines. However, here is one example:

A 100% magnification EAN or UPC bar code has a final printed tolerance of + or - 101 microns (1 micron is one-thousandth of a millimetre). For direct printing a resolution of 400 d.p.i. has been found to be sufficient for 100% EAN/UPC. For flexo platemaking of 100% EAN/UPC from a negative film, 1200 d.p.i has proved acceptable.

As all conditions vary, Computalabel International provide a free initial verification service to all registered users. Example output is scanned by a device which precisely measures all bars and spaces and the user is advised of the quality implications.

Printer Compensation and Bar Width Reduction

In order for raster based printers to produce solid areas of print from a series of circular dots, the dots must overlap. Therefore the dots must be larger than the distance between their centres. This affects the accuracy of the bars of a barcode.

However, our software offers a feature known as Printer Compensation, which allows the PostScript description of each bar to be thinned by a given amount to compensate for this effect. It must be noted that the amount of bar spread depends on the type of media that is used and the type of image that is set. For example, it has been noticed that bromide paper spreads more than film and that often positive images spread more than negatives.

A further spreading effect happens with traditional printing. If output is to be used for platemaking, the spread of the end print process must also be considered. This "Bar Width Reduction" factor (or BWR) depends on the type of print process (e.g. flexo, litho, etc..) the individual press and the conditions which apply (e.g. inks and type of plate material). Again, with so many variables, it is advisable to check the final printed sample - a free service offered by Computalabel International.

Whatever method is utilised for the production of bar code images, there can be no safer way to ensure the end readability of the code than to apply a quality check to all printed output. A number of quality checking devices are available, ranging from simple readers costing single hundreds of dollars, to complex laser scanners costing thousands.