Browse Prior Art Database

Digital Font Scaling Method

IP.com Disclosure Number: IPCOM000041681D
Original Publication Date: 1984-Feb-01
Included in the Prior Art Database: 2005-Feb-02
Document File: 1 page(s) / 12K

Publishing Venue

IBM

Related People

Wade, BW: AUTHOR

Abstract

This invention relates to improvement in processing speed and quality of dot matrix fonts from their stored to their printed forms using gray-scale rather than black/white techniques. The method includes multiple passes in which the first pass involves forming a gray-scale matrix and the remaining passes serve to clean up the target image. The clean-up passes comprise resolution of pels of ambiguous value at the matrix edges and the identification of saddle points for elimination. This method transforms a binary bit map (BBM) into a pseudo gray-scale image (PGSI). The following is a method sequence of pattern transformations in order to scale a discrete character array originated as a BBM: (1) Input = BBM(1) T PGSI(1), using an area count algorithm. (2) PGSI(1) T PGSI(2), using a "fuzzy edge" algorithm.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 93% of the total text.

Page 1 of 1

Digital Font Scaling Method

This invention relates to improvement in processing speed and quality of dot matrix fonts from their stored to their printed forms using gray-scale rather than black/white techniques. The method includes multiple passes in which the first pass involves forming a gray-scale matrix and the remaining passes serve to clean up the target image. The clean-up passes comprise resolution of pels of ambiguous value at the matrix edges and the identification of saddle points for elimination. This method transforms a binary bit map (BBM) into a pseudo gray- scale image (PGSI). The following is a method sequence of pattern transformations in order to scale a discrete character array originated as a BBM:
(1) Input = BBM(1) T PGSI(1), using an area count algorithm. (2) PGSI(1) T PGSI(2), using a "fuzzy edge" algorithm. This algorithm finds any sequence of gray values in the form 0,m,100,100,...,100,n,0 , where m and n are gray values less than 100. This step then replaces the larger of m and n by the sum min (m+n,100), while replacing the smaller by m+n-min (m+n,100). (3) PGSI(2) T PGSI(3), using a "saddle point" transformation. This step checks for gray pels in the configuration. A B X C D and sets X to zero if X is less than both B and C and greater than both A and D, or if X is less than both A and D and greater than both B and C. (4) PGSI(3) T BBM(2), using thresholding. Step 1 recites the conversion of input to gray-scale representation. Step 2...