Browse Prior Art Database

Font Compression in High Resolution Printers

IP.com Disclosure Number: IPCOM000045139D
Original Publication Date: 1983-Feb-01
Included in the Prior Art Database: 2005-Feb-06
Document File: 3 page(s) / 83K

Publishing Venue

IBM

Related People

Bresenham, JE: AUTHOR [+3]

Abstract

This article concerns a font compression technique for microprocessor based high resolution printers of the kind which print "swathe-wise", i.e., where a linear array of dot-making devices pass across a page along a line of characters perpendicular to the direction of the array. Such printers include resistive ribbon, molecular matrix, ink jet and electro-erosion printers.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 57% of the total text.

Page 1 of 3

Font Compression in High Resolution Printers

This article concerns a font compression technique for microprocessor based high resolution printers of the kind which print "swathe-wise", i.e., where a linear array of dot-making devices pass across a page along a line of characters perpendicular to the direction of the array. Such printers include resistive ribbon, molecular matrix, ink jet and electro-erosion printers.

In the present technique, each character in the full font set is initially quantized into 'on' and 'off' pels (picture elements) in a fixed size rectangular dot matrix. Thus each character is represented by a matrix C of dimensions M rows by N columns in which matrix entries C(ij) are constrained to values of 0 or 1 and M and N are the product of integers such that M=am and N=bn: C(11) .......... C(1N)

C equals

C(M1) .......... C(MN).

Each such character is then decomposed into fractional vertical slices within its original dot matrix, that is, into aN submatrices P each of dimensions m by 1: P(11) .......... P(1N)

C equals

P(a1) .......... P(aN)
where

P(ij) equals C(1+(i-1)m,j)

C(im,j).

The above decomposition of the original matrix C assumes multi-swathe characters, each character being a times as high as a swathe which contains m dots. If a=1, however, the slices will be full rather than fractional (M=m), and each character may be fully reconstructed in a single swathe. Furthermore, interchanging M and N would provide a basis for comparable horizontal compression.

Next, the number of unique slices is determined, and a table is constructed containing one copy of every unique slice in the font. The slices are ordered to bring into sequence in the table those slices which commonl...