Patent US5050229A successfully refreshed from Google Patents.
US5050229A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Method and apparatus for thinning alphanumeric characters for optical character recognition
Assignee: Eastman Kodak Co
•
Filing: 1990-06-05
•
Granted: 1991-09-17
•
Expires: 2010-06-05
Abstract
Abstract The width of each stroke in a character image is reduced one pixel at a time over a series of repetitive cycles until the number of cycles exceeds a predetermined number proportional to the thinnest character stroke observed in the character image. The process is then continued until the character has been skeletonized, but with the added constraint that the existing end point pixels of character strokes are retained.
Inventors
- 👤 Lori L. Barski
- 👤 Hin-Leong Tan
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-06 15:23
CPC Classifications
G; G06; G06T; G06T5/00; G06T5/20; G06T5/30; G06V; G06V30/00; G06V30/10; G06V30/16; G06V30/168
Claims Analysis
Total Claims: 22
Claim 1
Independent Claim
87 words
In an optical character recognition system, a character thinning pre-processor comprising: means for determining a certain character stroke width from a character image; thinning means for removing successive layers of boundary pixels including boundary pixels which are also end point pixels from said character image; and control means for preventing said means for removing from removing boundary pixels which are also end point pixels of said character image whenever the number of layers removed is at least equal to a predetermined fraction of said certain stroke width.
Claim 2
Dependent Claim
87 words
In an optical character recognition system, a character thinning pre-processor comprising: means for determining a certain character stroke width from a character image; thinning means for removing successive layers of boundary pixels including boundary pixels which are also end point pixels from said character image; and control means for preventing said means for removing from removing boundary pixels which are also end point pixels of said character image whenever the number of layers removed is at least equal to a predetermined fraction of said certain stroke width.
Claim 3
Dependent Claim
59 words
The pre-processor of claim 1 wherein said thinning means comprises means for removing boundary pixels along boundaries in said character image facing a pair of orthogonal directions as one of said layers and means for removing boundary pixels along boundaries in said character image facing a pair of opposing orthogonal directions as a next one of said successive layers.
Claim 4
Dependent Claim
49 words
The pre-processor of claim 2 wherein said control means comprises: means for counting the number of layers removed by said thinning means; and means for sensing whether said number is less than the number of pixels in said image spanned by said pre-determined fraction of said certain stroke width.
Claim 5
Dependent Claim
43 words
The pre-processor of claim 1 wherein said image is bi-tonal and wherein the character in said image is represented by "ON" pixels in a field of "OFF" pixels, and wherein said thinning means comprise means for converting said boundary pixels to "OFF" pixels.
Claim 6
Dependent Claim
58 words
The pre-processor of claim 1 wherein said means for determining said certain stroke width comprise: means for locating all intersections of a set of orthogonal intercept lines with the boundaries of said character image, and means for determining the distances between pairs of said intersections along each of said intercept lines spanning individual character strokes of said image.
Claim 7
Dependent Claim
24 words
The pre-processor of claim 5 wherein said means for determining said certain stroke width comprise means for determining the shortest one of said distances.
Claim 8
Dependent Claim
22 words
The pre-processor of claim 6 wherein said predetermined fraction of said certain stroke width equals half said shortest one of said distances.
Claim 9
Dependent Claim
40 words
The pre-processor of claim 1 wherein said thinning means comprises: means responsive whenever immediate neighboring pixels of one boundary pixel form more than one object upon removal of said one boundary pixel for preventing removal of said one boundary pixel.
Claim 10
Dependent Claim
39 words
The pre-processor of claim 9 wherein said control means comprise means for detecting whether said one boundary pixel comprises an end pixel, said means for detecting comprising: means for determining whether only one of said neighboring pixels is "ON".
Claim 11
Dependent Claim
97 words
In an optical character recognition system, a character thinning pre-processor, comprising: thinning means for iteratively reducing the width of each character stroke in a character image by one pixel by the removal of boundary pixels including boundary pixels which are also endpoint pixels of character strokes in said image; control means responsive whenever said width has been reduced by a predetermined number of pixels for preventing said thinning means from removing boundary pixels which are end point pixels of character strokes in said image; and means for determining said predetermined number of pixels from said character image.
Claim 12
Dependent Claim
97 words
In an optical character recognition system, a character thinning pre-processor, comprising: thinning means for iteratively reducing the width of each character stroke in a character image by one pixel by the removal of boundary pixels including boundary pixels which are also endpoint pixels of character strokes in said image; control means responsive whenever said width has been reduced by a predetermined number of pixels for preventing said thinning means from removing boundary pixels which are end point pixels of character strokes in said image; and means for determining said predetermined number of pixels from said character image.
Claim 13
Dependent Claim
24 words
The pre-processor of claim 10 wherein said means for determining comprises means for determining a certain stroke width from character strokes in said image.
Claim 14
Dependent Claim
38 words
The pre-processor of claim 11 wherein said certain character stroke width is the smallest character stroke width in said image and wherein said predetermined number is a fraction of the number of pixels spanned by said smallest width.
Claim 15
Dependent Claim
89 words
In an optical character recognition system, a character thinning pre-processing method, comprising: determining a certain character stroke width from character strokes in a character image; removing successive layers of boundary pixels from said character image including boundary pixels which are also end point pixels of said character image until the number of layers removed equals or exceeds a predetermined fraction of said certain stroke width; and thereafter preventing removal of such end point pixels from said character image while removing successive layers of boundary pixels from said character image.
Claim 16
Dependent Claim
89 words
In an optical character recognition system, a character thinning pre-processing method, comprising: determining a certain character stroke width from character strokes in a character image; removing successive layers of boundary pixels from said character image including boundary pixels which are also end point pixels of said character image until the number of layers removed equals or exceeds a predetermined fraction of said certain stroke width; and thereafter preventing removal of such end point pixels from said character image while removing successive layers of boundary pixels from said character image.
Claim 17
Dependent Claim
52 words
The method of claim 13 wherein said removing step comprises removing boundary pixels along boundaries facing a pair of orthogonal directions as one of said layers and then removing boundary pixels along edges of said character image facing a pair of opposing orthogonal directions as a next one of said successive layers.
Claim 18
Dependent Claim
41 words
The method of claim 14 wherein said preventing step comprises: counting the number of layers removed; and sensing whether said number is less than the number of pixels in said image spanned by said predetermined fraction of said certain stroke width.
Claim 19
Dependent Claim
41 words
The method of claim 13 wherein said image is bi-tonal and wherein the character in said image is represented by "ON" pixels in a field of "OFF" pixels, and wherein said removal step comprises converting said boundary pixels to "OFF" pixels.
Claim 20
Dependent Claim
53 words
The method of claim 13 wherein said step of determining said certain stroke width comprises: locating all intersections of a set of orthogonal intercept lines with character stroke boundaries in said character image; and determining the distances between adjacent pairs of said intersections spanning individual character strokes along each of said intercept lines.
Claim 21
Dependent Claim
23 words
The method of claim 17 wherein said step of determining said typical stroke width further comprises determining the shortest one of said distances.
Claim 22
Dependent Claim
36 words
The method of claim 13 wherein said step of removal comprises: preventing removal of a boundary pixel whenever immediate neighboring pixels of said boundary pixel form more than one object upon removal of said boundary pixel.
References Cited (Backward Citations)
US3196398A
Cited
US3668637A
Cited
US3735349A
Cited
US3975709A
Cited
US3940737A
Cited
US3846754A
Cited
US4034344A
Cited
US4115760A
Cited
US4162482A
Cited
JPS55131876A
Cited
US4520505A
Cited
JPS62160595A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 1994-09-16 | 1995-03-18 | 1995-09-17 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 1998-09-16 | 1999-03-18 | 1999-09-17 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2002-09-16 | 2003-03-18 | 2003-09-17 | $7700.00 | Window Closed / Lapsed |