US5052044A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Correlated masking process for deskewing, filtering and recognition of vertically segmented characters
Assignee: Eastman Kodak Co
•
Filing: 1990-02-02
•
Granted: 1991-09-24
•
Expires: 2010-02-02
Abstract
Abstract Vertically segmented characters are pre-processed for optical character recognition so as to remove skew and noise by convolution with a mask consisting of parallel vertical lines with a one-pixel spatial frequency. The horizontal alignment of the mask with respect to each segmented character is adjusted until a maximum cross-correlation is found. Then, the product of the mask and the character image is formed. A histogram of the number of pixels in the product image along each vertical line of the mask is compared with a library of such histograms of known symbols and the best match identifies the symbol represented by the character image. Alternatively, the histogram is converted to a binary codeword which is compared with a library of such codewords corresponding to a set of known symbols.
Inventors
- 👤 Roger S. Gaborski
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-10-24 11:58
CPC Classifications
G; G06; G06V; G06V30/00; G06V30/10; G06V30/22; G06V30/224; G06V30/2247
Claims Analysis
Total Claims: 22
Claim 1
Independent Claim
140 words
An optical character recognition system for reading vertically segmented characters in a document image of parallel horizontal rows and vertical columns of pixels, said system comprising: means for defining a mask image comprising a plurality of parallel vertical lines; incremental mask stepping means for positioning said mask image at one of a plurality of incremental steps with respect to said document image; means for generating a product image by multiplying together corresponding pixels pairs of said document image an said mask image; means for determining which one of said steps corresponds to a maximum correlation between said document and mask images; means for generating a histogram of said product image corresponding to an incremental step having maximum correlation between said document and mask images; and means for determining from said histogram the identity of individual characters in said document image.
Claim 2
Dependent Claim
140 words
An optical character recognition system for reading vertically segmented characters in a document image of parallel horizontal rows and vertical columns of pixels, said system comprising: means for defining a mask image comprising a plurality of parallel vertical lines; incremental mask stepping means for positioning said mask image at one of a plurality of incremental steps with respect to said document image; means for generating a product image by multiplying together corresponding pixels pairs of said document image an said mask image; means for determining which one of said steps corresponds to a maximum correlation between said document and mask images; means for generating a histogram of said product image corresponding to an incremental step having maximum correlation between said document and mask images; and means for determining from said histogram the identity of individual characters in said document image.
Claim 3
Dependent Claim
28 words
The system of claim 1 wherein said means for generating a histogram comprises means for computing sums of "on" pixels in plural vertical columns in said product image.
Claim 4
Dependent Claim
80 words
The system of claim 2 wherein said means for generating a product image generates a product image for each one of said incremental steps and said means for generating a histogram generates a histogram from the product image of each one of said incremental steps, and wherein said means for determining a maximum correlation comprises: means for determining which one of said incremental steps corresponds to the greatest sums computed by said means for computing the sum of "on" pixels.
Claim 5
Dependent Claim
51 words
The system of claim 1 wherein said incremental mask stepping means moves said mask image across said document image in a horizontal direction with respect to said parallel vertical mask lines, wherein said vertical mask lines and said vertical character segments are at least nearly parallel to within a skew angle.
Claim 6
Dependent Claim
43 words
The system of claim 4 wherein different pairs of adjacent ones of said character segments are separated by different separation lengths and wherein said vertical mask lines are located at spatial intervals d which is a common denominator of said different separation lengths.
Claim 7
Dependent Claim
22 words
The system of claim 5 wherein said vertical mask lines have a thickness less than the thickness of said vertical character segments.
Claim 8
Dependent Claim
30 words
The system of claim 6 wherein said vertical mask lines have a thickness equal to one pixel and said spatial interval d of said vertical mask lines is one pixel.
Claim 9
Dependent Claim
51 words
The system of claim 1 wherein said means for determining said identity of said character comprises: means for storing a set of reference histograms; and means for comparing each histogram corresponding to said maximum correlation with said set of reference histograms and identifying the reference histogram having the greatest correlation therewith.
Claim 10
Dependent Claim
77 words
The system of claim 1 wherein said means for determining said identity of said character comprises: means for converting each sum in said histogram corresponding to said maximum correlation to a binary bit and forming an image codeword of the binary bits corresponding to one histogram; means for storing a set of reference codewords; and means for comparing the image codeword with said set of reference codewords and identifying the reference codeword having the greatest correlation therewith.
Claim 11
Dependent Claim
32 words
The system of claim 9 wherein said means for converting to binary bits converts each sum to a one or a zero depending upon whether the sum is non-zero or zero, respectively.
Claim 12
Dependent Claim
24 words
The system of claim 2 wherein said plural vertical columns along which said sums of said histograms are computed comprise said vertical mask lines.
Claim 13
Dependent Claim
128 words
In an optical character recognition system, a method for reading vertically segmented characters in a document image of parallel horizontal rows and vertical columns of pixels, said method comprising: defining a mask image comprising a plurality of parallel vertical lines; positioning said mask image at one of a plurality of incremental steps with respect to said document image; generating a product image by multiplying together corresponding pixels pairs of said document image and said mask image; determining which one of said steps corresponds to a maximum correlation between said document and mask images; generating a histogram of said product image corresponding to an incremental step having maximum correlation between said document and mask images; and determining from said histogram the identity of individual characters in said document image.
Claim 14
Dependent Claim
128 words
In an optical character recognition system, a method for reading vertically segmented characters in a document image of parallel horizontal rows and vertical columns of pixels, said method comprising: defining a mask image comprising a plurality of parallel vertical lines; positioning said mask image at one of a plurality of incremental steps with respect to said document image; generating a product image by multiplying together corresponding pixels pairs of said document image and said mask image; determining which one of said steps corresponds to a maximum correlation between said document and mask images; generating a histogram of said product image corresponding to an incremental step having maximum correlation between said document and mask images; and determining from said histogram the identity of individual characters in said document image.
Claim 15
Dependent Claim
26 words
The method of claim 12 wherein said step of generating a histogram comprises computing sums of "on" pixels in plural vertical columns in said product image.
Claim 16
Dependent Claim
81 words
The method of claim 13 wherein said step of generating a product image comprises generating a product image for each one of said incremental steps and said step of generating a histogram comprises generating a histogram from the product image of each one of said incremental steps, and wherein said step of determining a maximum correlation comprises: determining which one of said incremental steps corresponds to the greatest sums computed by said step of computing the sum of "on " pixels.
Claim 17
Dependent Claim
49 words
The method of claim 12 wherein said incremental mask positioning step comprises moving said mask image across said document image in a horizontal direction with respect to said parallel vertical mask lines, wherein said vertical mask lines and said vertical character segments are parallel to within a skew angle.
Claim 18
Dependent Claim
43 words
The method of claim 15 wherein different pairs of adjacent ones of said character segments are separated by different separation lengths and wherein said vertical mask lines are located at spatial intervals d which is a common denominator of said different separation lengths.
Claim 19
Dependent Claim
22 words
The method of claim 16 wherein said vertical mask lines have a thickness less than the thickness of said vertical character segments.
Claim 20
Dependent Claim
30 words
The method of claim 17 wherein said vertical mask lines have a thickness equal to one pixel and said spatial interval d of said vertical mask lines is one pixel.
Claim 21
Dependent Claim
47 words
The method of claim 12 wherein said step of determining said identity of said character comprises: storing a set of reference histograms; and comparing each histogram corresponding to said maximum correlation with said set of reference histograms and identifying the reference histogram having the greatest correlation therewith.
Claim 22
Dependent Claim
71 words
The method of claim 12 wherein said step of determining said identity of said character comprises: converting each sum in said histogram corresponding to said maximum correlation to a binary bit and forming an image codeword of the binary bits corresponding to one histogram; storing a set of reference codewords; and comparing the image codeword with said set of reference codewords and identifying the reference codeword having the greatest correlation therewith.
References Cited (Backward Citations)
US3044696A
Cited
US3283303A
Cited
US3309667A
Cited
US3423730A
Cited
US3539989A
Cited
US3644889A
Cited
US3688955A
Cited
US3717848A
Cited
US4053737A
Cited
US4185271A
Cited
US4239151A
Cited
US4282426A
Cited
US4513440A
Cited
US4567610A
Cited
US4736437A
Cited
US4806741A
Cited
US4809344A
Cited
US4949392A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 1994-09-23 | 1995-03-25 | 1995-09-24 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 1998-09-23 | 1999-03-25 | 1999-09-24 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2002-09-23 | 2003-03-25 | 2003-09-24 | $7700.00 | Window Closed / Lapsed |