US5455871A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Detecting function words without converting a scanned document to character codes
Assignee: Xerox Corp
•
Filing: 1994-05-16
•
Granted: 1995-10-03
•
Expires: 2014-05-16
Abstract
Abstract A method and apparatus detects function words in a first image of a scanned document without first converting the image to character codes. Function words include determiners, prepositions, articles, and other words that play a largely grammatical role, as opposed to words such as nouns and verbs that convey topic information. Non-content based morphological characteristics of image units are predetermined as well as the presence or omission of character ascenders and descenders in image units. Predetermined characteristics of function word image units are compared with the image units of an image and when a match occurs, the image unit is identified as a function word. Conversely when no matching characteristics occur, the image unit is identified as a non-function word. Additionally, image units are classified and identified as containing only upper case characters, only lower case characters, only digits, and mixed character types.
Inventors
- 👤 Dan S. Bloomberg
- 👤 John W. Tukey
- 👤 M. Margaret Withgott
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-10-27 13:26
CPC Classifications
G; G06; G06V; G06V30/00; G06V30/10; G06V30/19; G06V30/191; G06V30/19173
Claims Analysis
Total Claims: 11
Claim 1
Independent Claim
182 words
A method of processing an image of a scanned document, said document image having been previously scanned and segmented into image units, said method comprising the steps of: (a) determining a length and height of each image unit; (b) defining short image units as image units having a length no greater than a predetermined length; (c) deleting image units from the image that are not said short image units, using the results of step (b); (d) blurring the image in a horizontal direction to form a blurred image; (e) eroding the blurred image using a set of ascender matching structuring elements and a set of descender matching structuring elements to form two eroded images; (f) taking a UNION of the two eroded images; (g) performing a seed filling operation using the image resulting from step (f) to provide 1) short image unit masks filled for short image units with at least one ascender, and 2) short image unit masks filled for short image units with at least one descender; and (h) identifying image units corresponding to said filled short image unit masks.
Claim 2
Independent Claim
182 words
A method of processing an image of a scanned document, said document image having been previously scanned and segmented into image units, said method comprising the steps of: (a) determining a length and height of each image unit; (b) defining short image units as image units having a length no greater than a predetermined length; (c) deleting image units from the image that are not said short image units, using the results of step (b); (d) blurring the image in a horizontal direction to form a blurred image; (e) eroding the blurred image using a set of ascender matching structuring elements and a set of descender matching structuring elements to form two eroded images; (f) taking a UNION of the two eroded images; (g) performing a seed filling operation using the image resulting from step (f) to provide 1) short image unit masks filled for short image units with at least one ascender, and 2) short image unit masks filled for short image units with at least one descender; and (h) identifying image units corresponding to said filled short image unit masks.
Claim 3
Dependent Claim
29 words
The method as recited in claim 1, wherein the step of blurring the image includes blurring the image by morphologically DILATING the image using at least one structuring element.
Claim 4
Dependent Claim
21 words
The method as recited in claim 1, wherein the step of blurring the image includes blurring the image by threshold reduction.
Claim 5
Dependent Claim
40 words
The method as recited in claim 1, further comprising the step of subtracting a result of the seed filling step from an input to the blurring step to identify image units which consist of all upper case letters and digits.
Claim 6
Dependent Claim
16 words
The method as recited in claim 1, wherein the seed filling step includes fillClipping the image.
Claim 7
Independent Claim
166 words
An apparatus comprising: an image scanning device for scanning a document image; processor means for segmenting the image into image units, determining a length and height of each image unit, defining short image units as image units having a length no greater than a predetermined length, deleting image units from the image that are not defined as said short image units, blurring the image in a horizontal direction to form a blurred image, eroding the blurred image using a set of ascender matching structuring elements and a set of descender matching structuring elements to form two eroded images, taking a UNION of the two eroded images, performing a seed filling operation to provide 1) short image unit masks filled for short image units with at least one ascender, and 2) short image unit masks filled for short image units with at least one descender, and identifying image units corresponding to said filled short image unit masks; and an output device for outputting said identified image units.
Claim 8
Independent Claim
166 words
An apparatus comprising: an image scanning device for scanning a document image; processor means for segmenting the image into image units, determining a length and height of each image unit, defining short image units as image units having a length no greater than a predetermined length, deleting image units from the image that are not defined as said short image units, blurring the image in a horizontal direction to form a blurred image, eroding the blurred image using a set of ascender matching structuring elements and a set of descender matching structuring elements to form two eroded images, taking a UNION of the two eroded images, performing a seed filling operation to provide 1) short image unit masks filled for short image units with at least one ascender, and 2) short image unit masks filled for short image units with at least one descender, and identifying image units corresponding to said filled short image unit masks; and an output device for outputting said identified image units.
Claim 9
Independent Claim
188 words
A method of processing an image of a scanned document without converting the scanned document to character codes, said document image having been previously scanned and segmented into image units, said method comprising the steps of: (a) determining a length and height of each image unit; (b) defining short image units as image units of no greater than a predetermined length; (c) deleting image units from the image that are not said short image units, using the results of step (b); (d) blurring the image in a horizontal direction to form a blurred image; (e) eroding the blurred image using a set of ascender matching structuring elements and a set of descender matching structuring elements to form two eroded images; (f) taking a UNION of the two eroded images; (g) performing a seed filling operation using the image resulting from step (f) to provide 1) short image unit masks filled for short image units with at least one ascender, and 2) short image unit masks filled for short image units with at least one descender; and (h) identifying image units corresponding to said filled short image unit masks.
Claim 10
Independent Claim
188 words
A method of processing an image of a scanned document without converting the scanned document to character codes, said document image having been previously scanned and segmented into image units, said method comprising the steps of: (a) determining a length and height of each image unit; (b) defining short image units as image units of no greater than a predetermined length; (c) deleting image units from the image that are not said short image units, using the results of step (b); (d) blurring the image in a horizontal direction to form a blurred image; (e) eroding the blurred image using a set of ascender matching structuring elements and a set of descender matching structuring elements to form two eroded images; (f) taking a UNION of the two eroded images; (g) performing a seed filling operation using the image resulting from step (f) to provide 1) short image unit masks filled for short image units with at least one ascender, and 2) short image unit masks filled for short image units with at least one descender; and (h) identifying image units corresponding to said filled short image unit masks.
Claim 11
Dependent Claim
21 words
The method as recited in claim 7, wherein the predetermined length is a length corresponding to a length of three characters.
References Cited (Backward Citations)
US2905927A
Cited
US4773098A
Cited
US4654873A
Cited
US4918740A
Cited
US4860376A
Cited
US5144682A
Cited
US5197107A
Cited
US5131049A
Cited
US5181255A
Cited
US5172422A
Cited
US5307422A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 1998-10-02 | 1999-04-03 | 1999-10-03 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2002-10-02 | 2003-04-03 | 2003-10-03 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2006-10-02 | 2007-04-03 | 2007-10-03 | $7700.00 | Window Closed / Lapsed |