Intelligent font rendering co-processor

• US 5,301,267 A
• Filed: 09/27/1991
• Issued: 04/05/1994
• Est. Priority Date: 09/27/1991
• Status: Expired due to Term

• Current Assignee: Adobe Systems Incorporated (Adobe Inc.)
• Original Assignee: Adobe Systems Incorporated (Adobe Inc.)
• Inventors: Collins, Harry J., Hassett, Christopher R., Nogrady, John W.
• Primary Examiner(s): Harkcom, Gary V.
• Assistant Examiner(s): FEILD, JOSEPH H
• Application Number:
  US07/767,259
• Time in Patent Office: 921 Days
• Field of Search: 395/150, 395/151, 395/142, 395/143, 340/730, 340/731, 340/735, 340/750, 340/751, 340/799, 345/144
• US Class Current: 345/469
• CPC Class Codes: G06T 11/203 Drawing of straight lines or curves

Abstract

The present invention provides an apparatus and method for converting font outlines to rasterized bit maps. The method accesses stored outline data representing the object in a first coordinate space and transforms the outline data to corresponding data representing the object in a second coordinate space, maintaining regional relationship information in both coordinate spaces, through a non-linear transformation expressed as a plurality of linear transformation matrices, to generate a bit map suitable for displaying the object.

The present invention includes an apparatus to analyze Bezier curves and subdivide them as necessary until each portion is sufficiently flat to be approximated as a straight line, and then to calculate where line segments cross pixel midlines in order to fill the outline and generate the bit map.

From another perspective, the method takes an outline of an object in a first coordinate space, scales the outline to a second coordinate space, identifies the coordinates of one or more select points in the second coordinate space and compares those coordinates with desired coordinates in the second coordinate space, calculates the difference in device space for the desired versus the actual coordinate in the second coordinate space, derives a plurality of piecewise linear transformation matrices to approximate a non-linear transformation, applies an appropriate linear transformation matrix to map essentially any point on the outline in the first coordinate space to corresponding coordinates in the second coordinate space, and fills and stores the outline of the object in a form suitable for display on a raster device.

Claims

1. A method for displaying rasterized objects comprising:

accessing outline data from a computer memory or storage medium, said outline data representing said object in a first coordinate space,

deriving from said outline data representing said object in a first coordinate space the corresponding data representing said object in a second coordinate space, said outline data in said first coordinate space possessing regional relationship information for each of a plurality of regions, including the steps of;

(a) transforming said regional relationship information into said second coordinate space using a linear transformation;

(b) deriving from the transformed regional relationship information for each of the plurality of regions a non-linear transformation matrix expressed as a plurality of piecewise linear transformation matrices, one for each of the regions;

(c) applying said non-linear transformation matrix to said outline data representing said object in said first coordinate space to derive a second representation of said object in said second coordinate space;

(d) converting said second representation of said object to raster data describing said object in a form to be displayed on a display device; and

(e) displaying said second representation of said object on a raster device.

2. A method for rasterizing an object defined by outline data representing said object in a first coordinate space containing regional relationship information, comprising:

accessing said outline data from a computer memory or storage medium;

scaling said regional relationship information in said first coordinate space to derive scaled regional relationship information in a second coordinate space;

on a region-by-region basis, identifying a plurality of points, each defined by a pair of coordinates of said scaled regional relationship information within a region in said second coordinate space, wherein one of the two coordinates of such pair of coordinates does not coincide with a predetermined coordinate in said second coordinate space;

for each such point having a non-coinciding coordinate, measuring in the units of said second coordinate space the distance in said second coordinate space between said one coordinate of such point and said predetermined coordinate;

deriving from a plurality of distances measured for a plurality of said points, a non-linear transformation matrix approximated by a plurality of piecewise linear transformation matrices;

applying said non-linear transformation matrix to said outline data of said object in said first coordinate space to convert said outline data to said second coordinate space;

rasterizing the converted outline data to drive a form of said object to be displayed on a raster device; and

displaying the rasterized form of said object on a raster device.

3. The method of claim 2 wherein said outline data defines stems and counters in a character and wherein at least two points are identified, a first point lying on one of said stems and a second point lying on one of said counters, further comprising:

applying said regional relationship information in said second coordinate space to adjust the coordinates said first point on one of said stems in said first coordinate space before identifying said first point in said second coordinate space, and

applying said regional relationship information in said second coordinate space to adjust the coordinates of said second point on one of said counters in said first coordinate space before identifying said second point in said second coordinate space.

4. The method of claim 2 wherein said outline data contains information defining counters, further comprising

after accessing said outline data containing information defining counters and prior to applying said non-linear transformation matrix to said outline data,

grouping said counters with any adjacent counters of approximately similar height;

rounding the average height of each group of counters to an integer value;

summing the total height of all counters as rounded,

if the total height exceeds or does not fill the available space, adjusting the height of one or more of said groups of counters by changing the height of a group of counters and repeating this step with one or more additional groups of counters until the total height of the groups of counters fills the available space.

5. The method of claim 2 wherein said outline data contains information defining counters, further comprising

after accessing said outline data containing information defining counters and prior to applying said non-linear transformation matrix to said outline data,

grouping said counters with any adjacent counters of approximately similar width;

rounding the average width of each group of counters to an integer value;

summing the total width of all counters as rounded,

if the total width exceeds or does not fill the available space, adjusting the width of one or more of said groups of counters

by changing the width of a group of counters and repeating this step with one or more additional groups of counters until the total width of the groups of counters fills the available space.

6. The method of claim 2 wherein said outline data contains information defining counters, further comprising

after accessing said outline data containing information defining counters and prior to applying said non-linear transformation matrix to said outline data,

grouping said counters with any adjacent counters of approximately similar height;

rounding the average height of each group of counters to an integer value;

summing the total height of all counters as rounded and if the total height exceeds or does not fill the available space, adjusting the height of one or more of said groups of counters by dividing a group of counters into smaller groups and rounding the average size of each divided group and repeating this step until the total height of the averaged groups fills the available space.

7. The method of claim 2 wherein said outline data contains information defining counters, further comprising

after accessing said outline data containing information defining counters and prior to applying said non-linear transformation matrix to said outline data,

grouping said counters with any adjacent counters of approximately similar width;

rounding the average width of each group of counters to an integer value;

summing the total width of all counters as rounded and if the total width exceeds or does not fill the available space, adjusting the width of one or more of said groups of counters by dividing a group of counters into smaller groups and rounding the average size of each divided group and repeating this step until the total width of the averaged groups fills the available space.

8. Apparatus for displaying rasterized objects comprising:

means for accessing outline data from a computer memory or storage medium, said outline data representing said object in a first coordinate space,

means for deriving from said outline data representing said object in a first coordinate space the corresponding data representing said object in a second coordinate space, said outline data in said first coordinate space including regional relationship data to be maintained in said second coordinate space, including;

(a) means for transforming said regional relationship information into said second coordinate space using a linear transformation;

(b) means for deriving a non-linear transformation matrix expressed as a plurality of linear transformation matrices using the transformed regional relationship information;

(c) means for applying said non-linear transformation matrix to said outline data representing said object in said first coordinate space to derive a second representation of said object in said second coordinate space;

(d) means for converting said second representation of said object to raster data describing said object for display; and

(e) means for displaying said second representation of said object on a raster device.

9. Apparatus for rasterizing an object defined by outline data representing said object in a first coordinate space containing regional relationship information, comprising:

means for accessing said outline data from a computer memory or storage medium;

means for scaling said regional relationship information in said first coordinate space to derive scaled regional relationship information in a second coordinate space;

means for identifying a plurality of points, each defined by a pair of coordinates of said scaled regional relationship information in the second coordinate space, wherein one of the two coordinates of said pair of coordinates does not coincide with a predetermined coordinate in said second coordinate space;

for each such point, means for measuring in the units of said second coordinate space the distance in said second coordinate space between said one coordinate of such point and said predetermined coordinate;

means for deriving from a plurality of such measured distances a non-linear transformation matrix approximated by a plurality of piecewise linear transformation matrices;

means for applying said non-linear transformation matrix to said outline data of said object in said first coordinate space to convert said outline data to said second coordinate space;

means for rasterizing the converted outline data to derive a form of said object for display on a raster device; and

means for displaying said rasterized form of said object on a raster device.