Technical Notes, Zetafax

HOWTO: Improve the image quality of faxes sent using the Zetafax Automation Toolkit

ZTN2104

ID: ZTN2104

This Zetafax technical note applies to:

  • Zetafax version 17.0 and earlier

Summary

When using Zetafax with an application using the Zetafax Automation Toolkit for submitting faxes, the image quality of the sent fax is poor with the text and images including uneven edges and “speckling”.

This document describes how to adjust the algorithm which Zetafax uses for image conversion to improve the quality for certain document types.

Further information

Fax images are monochrome (each pixel is either black or white), have a fixed resolution of 204x196 dpi (for fine mode fax) or 204x98 dpi (for standard mode fax), and have a fixed width of 1728 pixels.  Zetafax stores fax images using an industry standard format called “TIFF-FAX”.  When sending an image file by fax, the highest image quality will result if the submitted image is in the correct format already.  If the image is submitted as a TIFF-FAX format file, Zetafax does not have to perform any further image conversion and the image quality is maintained.

If you submit an image file for sending which does not match the fax specification, then Zetafax has to convert it into the correct format before sending the fax.  This may require changing the resolution by combining pixels, or conversion from colour to monochrome.  Often the conversion process will result in a degradation of image quality.  For example, black text in a coloured or grayscale image often uses “anti-aliasing” at the edges of the letters, using varying shades of grey to make the edges appear smooth.  However converting this into pure monochrome can often result in a ragged edge, as the conversion algorithm attempts to represent the grey with alternating black and white dots.

Zetafax uses two different mechanisms for converting graphics files into the correct format for sending: the DOCTIFF document rendering engine, and a graphics library.  DOCTIFF supports a wide range of graphics and document file formats, but is only enabled for graphics file conversion on systems with the Document Rendering add-on installed.

The DOCTIFF application supports a number of different algorithms to perform this conversion process.  The default setting uses the “Bayer” algorithm which gives the best results for a range of file formats and image types.  However in systems which use a single image type, such as systems where faxes are being submitted automatically from another application using the Zetafax Automation Toolkit, then it may be possible to improve the image quality of sent faxes by changing to a different algorithm.

For documents which mainly comprise text, the “Diffusion” algorithm will often give a better result than the default setting.  This document gives instructions for changing the system to use this algorithm for all documents.

Implementation steps

To change the algorithm used for image conversion, first check that the Document Rendering add-on is installed on the system.  Run the Zetafax Configuration program, then display the licence details.  Check that the licence description includes the word “Render” or “Rendering” indicating that the add-on is installed – otherwise you will need to obtain and install this add-on before continuing with this change.

Next, using a text edition such as Notepad, open the system configuration file SETUP.INI (in folder zfax\SYSTEM\Z-DB), then scroll down the file and locate the [DOCTIFF] section.

Add the following line to the [DOCTIFF] section (or change the existing line if present):

IGFunc: DIFFUSION

The settings are re-read for every file conversion, so there is no need to restart the Zetafax Server after making the change.

Next submit a new fax from the application, and compare the image quality with a previous fax.  If the quality is improved then keep the setting.  Otherwise repeat the steps above substituting “HALFTONE” for “DIFFUSION”.  Keep the setting which gives the best result.

Last updated: 14th April 2016 (GW) 

Keywords: API