Convolution filtering for decoding self-clocking glyph shape codes

• US 5,128,525 A
• Filed: 07/31/1990
• Issued: 07/07/1992
• Est. Priority Date: 07/31/1990
• Status: Expired due to Fees

• Current Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Original Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Inventors: Stearns, Richard G., Hecht, David L., Bloomberg, Dan S.
• Primary Examiner(s): PITTS, HAROLD I
• Application Number:
  US07/560,654
• Time in Patent Office: 707 Days
• Field of Search: 235/454, 235/456, 235/494, 382/9, 382/30, 382/40, 382/54, 364/413.21
• US Class Current: 235/454
• CPC Class Codes: G06K 19/06037 multi-dimensional coding
• CPC Class Codes: G06K 7/143 Glyph-codes
• CPC Class Codes: G06V 30/224 of printed characters having additional code marks or containing code marks

Abstract

Weighted and unweighted convolution filtering processes are provided for decoding bitmap image space representations of self-clocking glyph shape codes and for tracking the number and locations of the ambiquities or "errors" that are encountered during the decoding. This error detection may be linked to or compared against the error statistics from an alternative decoding process, such as the binary image processing techniques that are described herein to increase the reliability of the decoding that is obtained.

Claims

1. A process for decoding a bitmap image space representation of a self-clocking glyph shape code composed of glyphs having shapes that encode digital values, such that every distinct digital value that is encoded by said code is represented by the shape of a respective glyph;

said glyphs being selected from a set of n permissible glyph shapes, with each of said glyph shapes being preassigned to the encoding of a predetermined digital value;

said glyphs being spatially distributed in said bitmap image space, at an unknown skew angle, in substantial accordance with a spatial formatting rule;

said process comprising the steps of

locating, in coordinates that define said bitmap image space, at least three non-colinear points of known spatial relationship to said bitmap image space representation of said code;

calculating said skew angle from the coordinates of said non-colinear points;

applying said spatial formatting rule, at said skew angle, to said bitmap image space representation of said code for locating said glyphs in said bitmap image space;

convolving said bit map image space representation of said glyphs with at least n convolution filters, each of said convolution filters being relatively strongly matched to a respective one of said permissible glyph shapes and relatively poorly matched to all of the other of said permissibly glyph shapes, thereby providing at least n convolved images of each of said glyphs;

evaluating said convolved images glyph-by-glyph for determining a convolution value for each glyph as convolved with each filter;

comparing the convolution values for each of said glyphs for classifying said glyphs by their shapes; and

assigning decoded data values to said shape classified glyphs in accordance with the data values preassigned to said glyph shapes.

2. The decoding process of claim 1 wherein said convolution filters are weighted to emphasize distinctive features of the shapes to which they are strongly matched and to de-emphasize features of shapes to which they are poorly matched.

3. The decoding process of any of claims 1 and 2 wherein the convolution values that are compared are local maximum values.

4. The decoding process of any of claims 1 and 2 wherein the convolution values that are compared are sums of convolution values from a predetermined area surrounding local maximum values.