Patent US4453408A successfully refreshed from Google Patents.
Patent US5809167A successfully refreshed from Google Patents.
US5809167A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Page segmentation and character recognition system
Assignee: Canon Inc
•
Filing: 1997-01-08
•
Granted: 1998-09-15
•
Expires: 2017-01-08
Abstract
Abstract A personal imaging computer system, which is connectable to and operable with a computerized local or wide area network, identifies characters in a document on which the characters are formed. The system scans the document to obtain a gray-scale image of the document, generates a binary image from the gray-scale image by comparing the gray-scale image with the threshold, segments the binary image to locate individual characters within the binary image and to determine the shape of the individual characters, extracts gray-scale image information from the gray-scale image for each such individual character based on the location and shape of the character in the binary image, recognition-processes the extracted gray scale image information to determine the identity of the character, and stores the identity of the character.
Inventors
- 👤 Hussein Al-Hussein
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-09 10:52
CPC Classifications
G; G06; G06V; G06V30/00; G06V30/40; G06V30/41; G06V30/414; G06V30/10; G06V30/14; G06V30/146; G06V30/1475; G06V30/1478; G06V30/148; G06V30/155; G06V30/24; G06V30/248
Claims Analysis
Total Claims: 192
Claim 1
Independent Claim
117 words
A character recognition apparatus for identifying characters in a gray-scale image which includes gray-scale character images of the characters, comprising: thresholding means for thresholding the gray-scale image to obtain a binary image which includes binary character images that correspond to the gray-scale character images; segmenting means for segmenting the binary image to locate individual binary character images; extracting means for extracting the gray-scale character images from the gray-scale image using corresponding ones of the individual binary character images segmented by said segmenting means as templates; and recognition-processing means for recognition-processing the extracted gray-scale character images, which are extracted from the gray-scale image using the binary character image templates, so as to determine the identity of the characters.
Claim 2
Dependent Claim
117 words
A character recognition apparatus for identifying characters in a gray-scale image which includes gray-scale character images of the characters, comprising: thresholding means for thresholding the gray-scale image to obtain a binary image which includes binary character images that correspond to the gray-scale character images; segmenting means for segmenting the binary image to locate individual binary character images; extracting means for extracting the gray-scale character images from the gray-scale image using corresponding ones of the individual binary character images segmented by said segmenting means as templates; and recognition-processing means for recognition-processing the extracted gray-scale character images, which are extracted from the gray-scale image using the binary character image templates, so as to determine the identity of the characters.
Claim 3
Dependent Claim
38 words
An apparatus according to claim 1, further comprising obtaining means for obtaining a histogram of density distributions in the gray-scale image, and for deriving the threshold for said thresholding means based on adjacency of bins in the histogram.
Claim 4
Dependent Claim
29 words
An apparatus according to claim 1, further comprising removing means for removing underlines from the binary image, wherein the gray-scale character images are extracted from underlined gray-scale character image.
Claim 5
Dependent Claim
18 words
An apparatus according to claim 1, wherein said segmenting means performs connected component analysis of the binary image.
Claim 6
Dependent Claim
23 words
An apparatus according to claim 4, further comprising reconstructing means for reconstructing lines of characters so as to reconnect improperly separated connected components.
Claim 7
Dependent Claim
29 words
An apparatus according to claim 4, further comprising separating means for separating text-type connected components from non-text-type connected components, wherein said recognition processing means recognition-processes only text-type connected components.
Claim 8
Dependent Claim
32 words
An apparatus according to claim 1, further comprising reconstructing means for reconstructing reading order of characters on a document page and storing the identities of characters based on the page-reconstructed reading order.
Claim 9
Dependent Claim
15 words
An apparatus according to claim 1, further comprising inputting means for inputting the gray-scale image.
Claim 10
Dependent Claim
26 words
An apparatus according to claim 8, wherein said inputting means step comprises scanning means for scanning in a document page on which the characters are formed.
Claim 11
Dependent Claim
23 words
An apparatus according to claim 8, wherein said inputting means comprises retrieving means for retrieving a gray-scale document image from a computer file.
Claim 12
Dependent Claim
44 words
An apparatus according to claim 10, wherein said inputting means comprises reading means for reading a computer file across a computerized local area network, and further comprising storing means for storing the identities of characters in a computer file on the local area network.
Claim 13
Dependent Claim
22 words
A character recognition apparatus according to claim 1, wherein the binary character image templates are enlarged for use by said extracting means.
Claim 14
Dependent Claim
20 words
A character recognition apparatus according to claim 12, wherein the binary character image templates are enlarged by approximately two pixels.
Claim 15
Dependent Claim
43 words
A character recognition apparatus according to claim 1, wherein the gray-scale character images are extracted by extracting individual gray-scale character images from the gray-scale image using corresponding individual binary character images as templates, with one gray-scale character image extracted for each individual character.
Claim 16
Dependent Claim
158 words
A character recognition apparatus for identifying characters in a gray-scale image which includes gray-scale character images of the characters, comprising: a memory for storing the gray-scale image and for storing instruction sequences by which the gray-scale image is processed; and a processor for executing the stored instruction sequences; wherein the stored instruction sequences include process steps to cause said processor to (a) obtain a binary image which includes binary character images that correspond to the gray-scale character images by comparing the gray-scale image with a threshold, (b) segment the binary image to locate individual binary character images, (c) extract the gray-scale character images from the gray-scale image using corresponding ones of the individual binary character images segmented by the process steps to segment the binary image as templates, and (d) recognition-process the extracted gray-scale character images, which are extracted from the gray-scale image using the binary character image templates, so as to determine the identity of the characters.
Claim 17
Dependent Claim
158 words
A character recognition apparatus for identifying characters in a gray-scale image which includes gray-scale character images of the characters, comprising: a memory for storing the gray-scale image and for storing instruction sequences by which the gray-scale image is processed; and a processor for executing the stored instruction sequences; wherein the stored instruction sequences include process steps to cause said processor to (a) obtain a binary image which includes binary character images that correspond to the gray-scale character images by comparing the gray-scale image with a threshold, (b) segment the binary image to locate individual binary character images, (c) extract the gray-scale character images from the gray-scale image using corresponding ones of the individual binary character images segmented by the process steps to segment the binary image as templates, and (d) recognition-process the extracted gray-scale character images, which are extracted from the gray-scale image using the binary character image templates, so as to determine the identity of the characters.
Claim 18
Dependent Claim
43 words
An apparatus according to claim 15, wherein the stored instruction sequences further include process steps for obtaining a histogram of density distributions in the gray-scale image, and for deriving the threshold for said thresholding step based on adjacency of bins in the histogram.
Claim 19
Dependent Claim
34 words
An apparatus according to claim 15, wherein the stored instruction sequences further include process steps for removing underlines from the binary image, wherein the gray-scale character images are extracted from underlined gray-scale character image.
Claim 20
Dependent Claim
23 words
An apparatus according to claim 15, wherein the stored instruction sequences further include process steps for connected component analysis of the binary image.
Claim 21
Dependent Claim
28 words
An apparatus according to claim 18, wherein the stored instruction sequences further include process steps for reconstructing lines of characters so as to reconnect improperly separated connected components.
Claim 22
Dependent Claim
34 words
An apparatus according to claim 18, wherein the stored instruction sequences further include process steps for separating text-type connected components from non-text-type connected components, wherein recognition processing is performed only on text-type connected components.
Claim 23
Dependent Claim
37 words
An apparatus according to claim 15, wherein the stored instruction sequences further include process steps for reconstructing reading order of characters on a document page and storing the identities of characters based on the page-reconstructed reading order.
Claim 24
Dependent Claim
20 words
An apparatus according to claim 15, wherein the stored instruction sequences further include process steps for inputting the gray-scale image.
Claim 25
Dependent Claim
23 words
An apparatus according to claim 22, wherein the gray-scale image is input by scanning a document page on which the characters are formed.
Claim 26
Dependent Claim
22 words
An apparatus according to claim 22, wherein the gray-scale image is input by retrieving a gray-scale document image from a computer file.
Claim 27
Dependent Claim
47 words
An apparatus according to claim 24, wherein the gray-scale image is input by reading a computer file across a computerized local area network, and wherein the stored instruction sequences include a step to store the identities of characters in a computer file on the local area network.
Claim 28
Dependent Claim
28 words
A character recognition apparatus according to claim 15, wherein the binary character image templates are enlarged for use in extracting the gray-scale character images from the gray-scale image.
Claim 29
Dependent Claim
20 words
A character recognition apparatus according to claim 26, wherein the binary character image templates are enlarged by approximately two pixels.
Claim 30
Dependent Claim
43 words
A character recognition apparatus according to claim 15, wherein the gray-scale character images are extracted by extracting individual gray-scale character images from the gray-scale image using corresponding individual binary character images as template, with one gray-scale character image extracted for each individual character.
Claim 31
Dependent Claim
114 words
A character recognition method for identifying characters in a gray-scale image which includes gray-scale character images of the characters, said method comprising the steps of: thresholding the gray-scale image to obtain a binary image which includes binary character images that correspond to the gray-scale character images; segmenting the binary image to locate individual binary character images; extracting the gray-scale character images from the gray-scale image using corresponding ones of the individual binary character images segmented by the step of segmenting the binary image as templates; and recognition-processing the extracted gray-scale character images, which are extracted from the gray-scale image using the binary character image templates, so as to determine the identity of the characters.
Claim 32
Dependent Claim
114 words
A character recognition method for identifying characters in a gray-scale image which includes gray-scale character images of the characters, said method comprising the steps of: thresholding the gray-scale image to obtain a binary image which includes binary character images that correspond to the gray-scale character images; segmenting the binary image to locate individual binary character images; extracting the gray-scale character images from the gray-scale image using corresponding ones of the individual binary character images segmented by the step of segmenting the binary image as templates; and recognition-processing the extracted gray-scale character images, which are extracted from the gray-scale image using the binary character image templates, so as to determine the identity of the characters.
Claim 33
Dependent Claim
37 words
A method according to claim 29, further comprising the steps of obtaining a histogram of density distributions in the gray-scale image, and deriving the threshold for said thresholding step based on adjacency of bins in the histogram.
Claim 34
Dependent Claim
29 words
A method according to claim 29, further comprising the step of removing underlines from the binary image, wherein the gray-scale character images are extracted from underlined gray-scale character image.
Claim 35
Dependent Claim
21 words
A method according to claim 29, wherein said segmenting step includes the step of connected component analysis of the binary image.
Claim 36
Dependent Claim
23 words
A method according to claim 32, further comprising the step of reconstructing lines of characters so as to reconnect improperly separated connected components.
Claim 37
Dependent Claim
31 words
A method according to claim 32, further comprising the step of separating text-type connected components from non-text-type connected components, wherein said recognition processing step is performed only on text-type connected components.
Claim 38
Dependent Claim
32 words
A method according to claim 29, further comprising the step of reconstructing reading order of characters on a document page and storing the identities of characters based on the page-reconstructed reading order.
Claim 39
Dependent Claim
15 words
A method according to claim 29, further comprising the step of inputting the gray-scale image.
Claim 40
Dependent Claim
25 words
A method according to claim 36, wherein said inputting step comprises the step of scanning in a document page on which the characters are formed.
Claim 41
Dependent Claim
23 words
A method according to claim 36, wherein said inputting step comprises the step of retrieving a gray-scale document image from a computer file.
Claim 42
Dependent Claim
44 words
A method according to claim 38, wherein said inputting step comprises the step of reading a computer file across a computerized local area network, and further comprising the step of storing the identities of characters in a computer file on the local area network.
Claim 43
Dependent Claim
28 words
A character recognition method according to claim 29, wherein the binary character image templates are enlarged for use in extracting the gray-scale character images from the gray-scale image.
Claim 44
Dependent Claim
20 words
A character recognition method according to claim 40, wherein the binary character image templates are enlarged by approximately two pixels.
Claim 45
Dependent Claim
43 words
A character recognition method according to claim 29, wherein the gray-scale character images are extracted by extracting individual gray-scale character images from the gray-scale image using corresponding individual binary character images as templates, with one gray-scale character image extracted for each individual character.
Claim 46
Dependent Claim
128 words
Computer-executable process steps stored on a computer-readable medium, which identify characters in a gray-scale image which includes gray-scale character images of the characters, the process steps including: a thresholding step which thresholds the gray-scale image to obtain a binary image which includes binary character images that correspond to the gray-scale character images; a segmenting step which segments the binary image to locate individual binary character images; an extracting step which extracts the gray-scale character images from the gray-scale image using corresponding ones of the individual binary character images segmented by said segmenting step as templates; and a recognition-processing step which recognition-processes the extracted gray-scale character images, which are extracted from the gray-scale image using the binary character image templates, so as to determine the identity of the characters.
Claim 47
Dependent Claim
128 words
Computer-executable process steps stored on a computer-readable medium, which identify characters in a gray-scale image which includes gray-scale character images of the characters, the process steps including: a thresholding step which thresholds the gray-scale image to obtain a binary image which includes binary character images that correspond to the gray-scale character images; a segmenting step which segments the binary image to locate individual binary character images; an extracting step which extracts the gray-scale character images from the gray-scale image using corresponding ones of the individual binary character images segmented by said segmenting step as templates; and a recognition-processing step which recognition-processes the extracted gray-scale character images, which are extracted from the gray-scale image using the binary character image templates, so as to determine the identity of the characters.
Claim 48
Dependent Claim
40 words
Computer-executable process steps according to claim 43, further comprising an obtaining step which obtains a histogram of density distributions in the gray-scale image, and which derives the threshold for said thresholding step based on adjacency of bins in the histogram.
Claim 49
Dependent Claim
30 words
Computer-executable process steps according to claim 43, further a removing step which removes underlines from the binary image, wherein the gray-scale character images are extracted from underlined gray-scale character image.
Claim 50
Dependent Claim
22 words
Computer-executable process steps according to claim 43, wherein said segmenting step includes the step of connected component analysis of the binary image.
Claim 51
Dependent Claim
25 words
Computer-executable process steps according to claim 46, further comprising a reconstructing step which reconstructs lines of characters so as to reconnect improperly separated connected components.
Claim 52
Dependent Claim
33 words
Computer-executable process steps according to claim 46, further comprising a separating step which separates text-type connected components from non-text-type connected components, wherein said recognition processing step is performed only on text-type connected components.
Claim 53
Dependent Claim
38 words
Computer-executable process steps according to claim 43, further comprising a reconstructing step which reconstructs reading order of characters on a document page and a storing step which stores the identities of characters based on the page-reconstructed reading order.
Claim 54
Dependent Claim
17 words
Computer-executable process steps according to claim 43, further comprising a inputting step which inputs the gray-scale image.
Claim 55
Dependent Claim
27 words
Computer-executable process steps according to claim 50, wherein said inputting step comprises a step scanning step which in a document page on which the characters are formed.
Claim 56
Dependent Claim
25 words
Computer-executable process steps according to claim 50, wherein said inputting step comprises a retrieving step which retrieves a gray-scale document image from a computer file.
Claim 57
Dependent Claim
47 words
Computer-executable process steps according to claim 52, wherein said inputting step comprises a reading step which reads a computer file across a computerized local area network, and further comprising a storing step which stores the identities of characters in a computer file on the local area network.
Claim 58
Dependent Claim
27 words
Computer-executable process steps according to claim 43, wherein the binary character image templates are enlarged for use in extracting the gray-scale character images from the gray-scale image.
Claim 59
Dependent Claim
19 words
Computer-executable process steps according to claim 54, wherein the binary character image templates are enlarged by approximately two pixels.
Claim 60
Dependent Claim
42 words
Computer-executable process steps according to claim 43, wherein the gray-scale character images are extracted by extracting individual gray-scale character images from the gray-scale image using corresponding individual binary character images as templates, with one gray-scale character image extracted for each individual character.
Claim 61
Dependent Claim
159 words
A character recognition apparatus for identifying characters in a document on which the characters are formed, said apparatus comprising: scanning means for scanning the document to obtain a gray-scale image of the document; generating means for generating a binary image from the gray-scale image by comparing the gray-scale image with a threshold; segmenting means for segmenting the binary image to locate individual characters within the binary image and to determine the shape of the individual characters; extracting means for extracting gray-scale image information from the gray-scale image for each such individual character based on the location and shape of the character segmented by said segmenting means; processing means for processing the extracted gray-scale image information, which is extracted from the gray-scale image based on the location and shape of each individual character in the binary image, so as to determine the identity of the character; and storing means for storing the identity of the character in a text file.
Claim 62
Dependent Claim
159 words
A character recognition apparatus for identifying characters in a document on which the characters are formed, said apparatus comprising: scanning means for scanning the document to obtain a gray-scale image of the document; generating means for generating a binary image from the gray-scale image by comparing the gray-scale image with a threshold; segmenting means for segmenting the binary image to locate individual characters within the binary image and to determine the shape of the individual characters; extracting means for extracting gray-scale image information from the gray-scale image for each such individual character based on the location and shape of the character segmented by said segmenting means; processing means for processing the extracted gray-scale image information, which is extracted from the gray-scale image based on the location and shape of each individual character in the binary image, so as to determine the identity of the character; and storing means for storing the identity of the character in a text file.
Claim 63
Dependent Claim
38 words
An apparatus according to claim 57, further comprising obtaining means for obtaining a histogram of density distributions in the gray-scale image, and for deriving the threshold for said thresholding step based on adjacency of bins in the histogram.
Claim 64
Dependent Claim
29 words
An apparatus according to claim 57, further comprising removing means for removing underlines from the binary image, wherein the gray-scale character images are extracted from underlined gray-scale character image.
Claim 65
Dependent Claim
18 words
An apparatus according to claim 57, wherein said segmenting means performs connected component analysis of the binary image.
Claim 66
Dependent Claim
23 words
An apparatus according to claim 60, further comprising reconstructing means for reconstructing lines of characters so as to reconnect improperly separated connected components.
Claim 67
Dependent Claim
30 words
An apparatus according to claim 60, further comprising separating means for separating text-type connected components from non-text-type connected components, wherein said recognition processing means recognition-processes only on text-type connected components.
Claim 68
Dependent Claim
32 words
An apparatus according to claim 57, further comprising reconstructing means for reconstructing reading order of characters on a document page and storing the identities of characters based on the page-reconstructed reading order.
Claim 69
Dependent Claim
32 words
A character recognition apparatus according to claim 57, wherein said extraction means extracts the gray-scale image information from the gray-scale image based on an enlargement of the character in the binary image.
Claim 70
Dependent Claim
30 words
A character recognition apparatus according to claim 64, wherein the enlargement of the character in the binary image is approximately two pixels larger than the character in the binary image.
Claim 71
Dependent Claim
189 words
A character recognition apparatus for identifying characters in a document on which the characters are formed, said apparatus comprising: a scanner for scanning the document to obtain a gray-scale image of the document; a processor; and a memory for storing (a) the gray-scale image obtained by the scanner, (b) a text file containing the identity of characters in the gray-scale image, and (c) process steps adapted to cause said processor to (i) generate a binary image from the gray-scale image by comparing the gray-scale image with a threshold; (ii) segment the binary image to locate individual characters within the binary image and to determine the shape of the individual characters; (iii) extract gray-scale image information from the gray-scale image for each such individual character based on the location and shape of the character segmented by the process steps to segment the binary image; (iv) process the extracted gray-scale image information, which is extracted from the gray-scale image based on the location and shape of each individual character in the binary image, so as to determine the identity of the character; and (v) store the identity in the text file.
Claim 72
Dependent Claim
189 words
A character recognition apparatus for identifying characters in a document on which the characters are formed, said apparatus comprising: a scanner for scanning the document to obtain a gray-scale image of the document; a processor; and a memory for storing (a) the gray-scale image obtained by the scanner, (b) a text file containing the identity of characters in the gray-scale image, and (c) process steps adapted to cause said processor to (i) generate a binary image from the gray-scale image by comparing the gray-scale image with a threshold; (ii) segment the binary image to locate individual characters within the binary image and to determine the shape of the individual characters; (iii) extract gray-scale image information from the gray-scale image for each such individual character based on the location and shape of the character segmented by the process steps to segment the binary image; (iv) process the extracted gray-scale image information, which is extracted from the gray-scale image based on the location and shape of each individual character in the binary image, so as to determine the identity of the character; and (v) store the identity in the text file.
Claim 73
Dependent Claim
46 words
An apparatus according to claim 66, wherein the process steps stored in said memory further include steps adapted to cause said processor to obtain a histogram of density distributions in the gray-scale image, and to derive the threshold based on adjacency of bins in the histogram.
Claim 74
Dependent Claim
41 words
An apparatus according to claim 66, wherein the process steps stored in said memory further include steps adapted to cause said processor to remove underlines from the binary image, wherein the gray-scale character images are extracted from underlined gray-scale character image.
Claim 75
Dependent Claim
31 words
An apparatus according to claim 66, wherein the process steps stored in said memory further include steps adapted to cause said processor to perform connected component analysis of the binary image.
Claim 76
Dependent Claim
35 words
An apparatus according to claim 69, wherein the process steps stored in said memory further include steps adapted to cause said processor to reconstruct lines of characters so as to reconnect improperly separated connected components.
Claim 77
Dependent Claim
40 words
An apparatus according to claim 69, wherein the process steps stored in said memory further include steps adapted to cause said processor to separate text-type connected components from non-text-type connected components, wherein recognition-processing is performed only on text-type connected components.
Claim 78
Dependent Claim
44 words
An apparatus according to claim 66, wherein the process steps stored in said memory further include steps adapted to cause said processor to reconstruct reading order of characters on a document page and storing the identities of characters based on the page-reconstructed reading order.
Claim 79
Dependent Claim
30 words
A character recognition apparatus according to claim 66, wherein the gray-scale image information is extracted from the gray-scale image based on an enlargement of the character in the binary image.
Claim 80
Dependent Claim
30 words
A character recognition apparatus according to claim 73, wherein the enlargement of the character in the binary image is approximately two pixels larger than the character in the binary image.
Claim 81
Dependent Claim
144 words
A character recognition method for identifying characters in a document on which the characters are formed, said method comprising the steps of: scanning the document to obtain a gray-scale image of the document; generating a binary image from the gray-scale image by comparing the gray-scale image with a threshold; segmenting the binary image to locate individual characters within the binary image and to determine the shape of the individual characters; extracting gray-scale image information from the gray-scale image for each such individual character based on the location and shape of the character segmented by the step of segmenting the binary image; processing the extracted gray-scale image information, which is extracted from the gray-scale image based on the location and shape of each individual character in the binary image, so as to determine the identity of the character; and storing the identity of the character.
Claim 82
Dependent Claim
144 words
A character recognition method for identifying characters in a document on which the characters are formed, said method comprising the steps of: scanning the document to obtain a gray-scale image of the document; generating a binary image from the gray-scale image by comparing the gray-scale image with a threshold; segmenting the binary image to locate individual characters within the binary image and to determine the shape of the individual characters; extracting gray-scale image information from the gray-scale image for each such individual character based on the location and shape of the character segmented by the step of segmenting the binary image; processing the extracted gray-scale image information, which is extracted from the gray-scale image based on the location and shape of each individual character in the binary image, so as to determine the identity of the character; and storing the identity of the character.
Claim 83
Dependent Claim
37 words
A method according to claim 75, further comprising the steps of obtaining a histogram of density distributions in the gray-scale image, and deriving the threshold for said thresholding step based on adjacency of bins in the histogram.
Claim 84
Dependent Claim
29 words
A method according to claim 75, further comprising the step of removing underlines from the binary image, wherein the gray-scale character images are extracted from underlined gray-scale character image.
Claim 85
Dependent Claim
21 words
A method according to claim 75, wherein said segmenting step includes the step of connected component analysis of the binary image.
Claim 86
Dependent Claim
23 words
A method according to claim 78, further comprising the step of reconstructing lines of characters so as to reconnect improperly separated connected components.
Claim 87
Dependent Claim
31 words
A method according to claim 78, further comprising the step of separating text-type connected components from non-text-type connected components, wherein said recognition processing step is performed only on text-type connected components.
Claim 88
Dependent Claim
32 words
A method according to claim 75, further comprising the step of reconstructing reading order of characters on a document page and storing the identities of characters based on the page-reconstructed reading order.
Claim 89
Dependent Claim
30 words
A character recognition method according to claim 75, wherein the gray-scale image information is extracted from the gray-scale image based on an enlargement of the character in the binary image.
Claim 90
Dependent Claim
30 words
A character recognition method according to claim 82, wherein the enlargement of the character in the binary image is approximately two pixels larger than the character in the binary image.
Claim 91
Dependent Claim
166 words
Computer-executable process steps stored on a computer-readable medium, which identify characters in a document on which the characters are formed, the process steps including: a scanning step which scans the document to obtain a gray-scale image of the document; a generating step which generates a binary image from the gray-scale image by comparing the gray-scale image with a threshold; a segmenting step which segments the binary image to locate individual characters within the binary image and to determine the shape of the individual characters; an extracting step which extracts gray-scale image information from the gray-scale image for each such individual character based on the location and shape of the character segmented by said segmenting step; a processing step which processes the extracted gray-scale image information, which is extracted from the gray-scale image based on the location and shape of each individual character in the binary image, so as to determine the identity of the character; and a storing step which stores the identity of the character.
Claim 92
Dependent Claim
166 words
Computer-executable process steps stored on a computer-readable medium, which identify characters in a document on which the characters are formed, the process steps including: a scanning step which scans the document to obtain a gray-scale image of the document; a generating step which generates a binary image from the gray-scale image by comparing the gray-scale image with a threshold; a segmenting step which segments the binary image to locate individual characters within the binary image and to determine the shape of the individual characters; an extracting step which extracts gray-scale image information from the gray-scale image for each such individual character based on the location and shape of the character segmented by said segmenting step; a processing step which processes the extracted gray-scale image information, which is extracted from the gray-scale image based on the location and shape of each individual character in the binary image, so as to determine the identity of the character; and a storing step which stores the identity of the character.
Claim 93
Dependent Claim
40 words
Computer-executable process steps according to claim 84, further comprising an obtaining step which obtains a histogram of density distributions in the gray-scale image, and which derives the threshold for said thresholding step based on adjacency of bins in the histogram.
Claim 94
Dependent Claim
31 words
Computer-executable process steps according to claim 84, further comprising a removing step which removes underlines from the binary image, wherein the gray-scale character images are extracted from underlined gray-scale character image.
Claim 95
Dependent Claim
22 words
Computer-executable process steps according to claim 84, wherein said segmenting step includes the step of connected component analysis of the binary image.
Claim 96
Dependent Claim
25 words
Computer-executable process steps according to claim 87, further comprising a reconstructing step which reconstructs lines of characters so as to reconnect improperly separated connected components.
Claim 97
Dependent Claim
33 words
Computer-executable process steps according to claim 87, further comprising a separating step which separates text-type connected components from non-text-type connected components, wherein said recognition processing step is performed only on text-type connected components.
Claim 98
Dependent Claim
38 words
Computer-executable process steps according to claim 84, further comprising a reconstructing step which reconstructs reading order of characters on a document page and a storing step which stores the identities of characters based on the page-reconstructed reading order.
Claim 99
Dependent Claim
30 words
Computer executable process steps according to claim 84, wherein the gray-scale image information is extracted from the gray-scale image based on an enlargement of the character in the binary image.
Claim 100
Dependent Claim
30 words
Computer executable process steps according to claim 91, wherein the enlargement of the character in the binary image is approximately two pixels larger than the character in the binary image.
Claim 101
Dependent Claim
160 words
A character recognition apparatus for identifying characters in a document on which the characters are formed, comprising: scanning means for scanning the document to obtain a gray-scale image of the document; thresholding means for thresholding the gray-scale image to obtain a binary image; segmenting means for segmenting the binary image to locate individual binary character images and to determine attributes of the binary character images; extracting means for extracting a gray-scale character image from the gray-scale image using one of the binary character images as a template; first recognition-processing means for recognition-processing the extracted gray-scale character image, which is extracted from the gray-scale image using the binary character image template, so as to obtain a first identity for the character; second recognition processing means for recognition-processing said one binary character image to obtain a second identity for the character; and means for resolving ambiguities between the first and second identity based on the character attributes determined by said segmenting means.
Claim 102
Dependent Claim
160 words
A character recognition apparatus for identifying characters in a document on which the characters are formed, comprising: scanning means for scanning the document to obtain a gray-scale image of the document; thresholding means for thresholding the gray-scale image to obtain a binary image; segmenting means for segmenting the binary image to locate individual binary character images and to determine attributes of the binary character images; extracting means for extracting a gray-scale character image from the gray-scale image using one of the binary character images as a template; first recognition-processing means for recognition-processing the extracted gray-scale character image, which is extracted from the gray-scale image using the binary character image template, so as to obtain a first identity for the character; second recognition processing means for recognition-processing said one binary character image to obtain a second identity for the character; and means for resolving ambiguities between the first and second identity based on the character attributes determined by said segmenting means.
Claim 103
Dependent Claim
38 words
An apparatus according to claim 93, further comprising obtaining means for obtaining a histogram of density distributions in the gray-scale image, and for deriving the threshold for said thresholding step based on adjacency of bins in the histogram.
Claim 104
Dependent Claim
29 words
An apparatus according to claim 93, further comprising removing means for removing underlines from the binary image, wherein the gray-scale character images are extracted from underlined gray-scale character image.
Claim 105
Dependent Claim
18 words
An apparatus according to claim 93, wherein said segmenting means performs connected component analysis of the binary image.
Claim 106
Dependent Claim
23 words
An apparatus according to claim 96, further comprising reconstructing means for reconstructing lines of characters so as to reconnect improperly separated connected components.
Claim 107
Dependent Claim
28 words
An apparatus according to claim 96, further comprising separating means for separating text-type connected components from non-text-type connected components, wherein recognition-processing is performed only on text-type connected components.
Claim 108
Dependent Claim
32 words
An apparatus according to claim 93, further comprising reconstructing means for reconstructing reading order of characters on a document page and storing the identities of characters based on the page-reconstructed reading order.
Claim 109
Dependent Claim
35 words
An apparatus according to claim 93, further comprising means for performing connected component analysis of the binary image, and wherein the character attributes relate to physical image attributes of connected components in the binary image.
Claim 110
Dependent Claim
22 words
A character recognition apparatus according to claim 93, wherein the binary character image template is enlarged for use by said extracting means.
Claim 111
Dependent Claim
20 words
A character recognition apparatus according to claim 101, wherein the binary character image template is enlarged by approximately two pixels.
Claim 112
Dependent Claim
185 words
A character recognition apparatus for identifying characters in a document on which the characters are formed, comprising: a scanner for scanning the document to obtain a gray-scale image of the document; a memory for storing the gray-scale image and for storing instruction sequences by which the gray-scale image is processed; and a processor for executing the stored instruction sequences; wherein the stored instruction sequences include process steps to cause said processor to (a) generate a binary image by comparing the gray-scale image with a threshold, (b) segment the binary image to locate individual binary character images and to determine attributes of the binary character images, (c) extract a gray-scale character image from the gray-scale image using one of the binary character images as a template, (d) recognition-process the extracted gray-scale character image, which is extracted from the gray-scale image using the binary character image template, so as to obtain a first identity for the character, (e) recognition-process said one binary character image to obtain a second identity for the character, and (f) resolve ambiguities between the first and second identity based on the character attributes.
Claim 113
Dependent Claim
185 words
A character recognition apparatus for identifying characters in a document on which the characters are formed, comprising: a scanner for scanning the document to obtain a gray-scale image of the document; a memory for storing the gray-scale image and for storing instruction sequences by which the gray-scale image is processed; and a processor for executing the stored instruction sequences; wherein the stored instruction sequences include process steps to cause said processor to (a) generate a binary image by comparing the gray-scale image with a threshold, (b) segment the binary image to locate individual binary character images and to determine attributes of the binary character images, (c) extract a gray-scale character image from the gray-scale image using one of the binary character images as a template, (d) recognition-process the extracted gray-scale character image, which is extracted from the gray-scale image using the binary character image template, so as to obtain a first identity for the character, (e) recognition-process said one binary character image to obtain a second identity for the character, and (f) resolve ambiguities between the first and second identity based on the character attributes.
Claim 114
Dependent Claim
47 words
An apparatus according to claim 103, wherein the stored instruction sequences further include process steps to cause said processor to obtain a histogram of density distributions in the gray-scale image, and to derive the threshold for said thresholding step based on adjacency of bins in the histogram.
Claim 115
Dependent Claim
38 words
An apparatus according to claim 103, wherein the stored instruction sequences further include process steps to cause said processor to remove underlines from the binary image, wherein the gray-scale character images are extracted from underlined gray-scale character image.
Claim 116
Dependent Claim
28 words
An apparatus according to claim 103, wherein the stored instruction sequences further include process steps to cause said processor to perform connected component analysis of the binary image.
Claim 117
Dependent Claim
32 words
An apparatus according to claim 106, wherein the stored instruction sequences further include process steps to cause said processor to reconstruct lines of characters so as to reconnect improperly separated connected components.
Claim 118
Dependent Claim
37 words
An apparatus according to claim 106, wherein the stored instruction sequences further include process steps to cause said processor to separate text-type connected components from non-text-type connected components, wherein recognition-processing is performed only on text-type connected components.
Claim 119
Dependent Claim
41 words
An apparatus according to claim 103, wherein the stored instruction sequences further include process steps to cause said processor to reconstruct reading order of characters on a document page and storing the identities of characters based on the page-reconstructed reading order.
Claim 120
Dependent Claim
45 words
An apparatus according to claim 103, wherein the stored instruction sequences further include process steps to cause said processor to perform connected component analysis of the binary image, and wherein the character attributes relate to physical image attributes of connected components in the binary image.
Claim 121
Dependent Claim
28 words
A character recognition apparatus according to claim 103, wherein the binary character image template is enlarged for use in extracting the gray-scale character image from the gray-scale image.
Claim 122
Dependent Claim
20 words
A character recognition apparatus according to claim 111, wherein the binary character image template is enlarged by approximately two pixels.
Claim 123
Dependent Claim
142 words
A character recognition method for identifying characters in a document on which the characters are formed, said method comprising the steps of: scanning the document to obtain a gray-scale image of the document; thresholding the gray-scale image to obtain a binary image; segmenting the binary image to locate individual binary character images and to determine attributes of the binary character images; extracting a gray-scale character image from the gray-scale image using one of the binary character images as a template; recognition-processing the extracted gray-scale character image, which is extracted from the gray-scale image using the binary character image template, so as to obtain a first identity for the character; recognition-processing said one binary character image to obtain a second identity for the character; and resolving ambiguities between the first and second identity based on the character attributes determined in said segmenting step.
Claim 124
Dependent Claim
142 words
A character recognition method for identifying characters in a document on which the characters are formed, said method comprising the steps of: scanning the document to obtain a gray-scale image of the document; thresholding the gray-scale image to obtain a binary image; segmenting the binary image to locate individual binary character images and to determine attributes of the binary character images; extracting a gray-scale character image from the gray-scale image using one of the binary character images as a template; recognition-processing the extracted gray-scale character image, which is extracted from the gray-scale image using the binary character image template, so as to obtain a first identity for the character; recognition-processing said one binary character image to obtain a second identity for the character; and resolving ambiguities between the first and second identity based on the character attributes determined in said segmenting step.
Claim 125
Dependent Claim
37 words
A method according to claim 113, further comprising the step of obtaining a histogram of density distributions in the gray-scale image, and deriving the threshold for said thresholding step based on adjacency of bins in the histogram.
Claim 126
Dependent Claim
29 words
A method according to claim 113, further comprising the step of removing underlines from the binary image, wherein the gray-scale character images are extracted from underlined gray-scale character image.
Claim 127
Dependent Claim
21 words
A method according to claim 113, wherein said segmenting step includes the step of connected component analysis of the binary image.
Claim 128
Dependent Claim
23 words
A method according to claim 116, further comprising the step of reconstructing lines of characters so as to reconnect improperly separated connected components.
Claim 129
Dependent Claim
31 words
A method according to claim 116, further comprising the step of separating text-type connected components from non-text-type connected components, wherein said recognition processing step is performed only on text-type connected components.
Claim 130
Dependent Claim
32 words
A method according to claim 113, further comprising the step of reconstructing reading order of characters on a document page and storing the identities of characters based on the page-reconstructed reading order.
Claim 131
Dependent Claim
35 words
A method according to claim 113, further comprising the step of connected component analysis of the binary image, and wherein the character attributes relate to physical image attributes of connected components in the binary image.
Claim 132
Dependent Claim
28 words
A character recognition method according to claim 113, wherein the binary character image template is enlarged for use in extracting the gray-scale character image from the gray-scale image.
Claim 133
Dependent Claim
20 words
A character recognition method according to claim 121, wherein the binary character image template is enlarged by approximately two pixels.
Claim 134
Dependent Claim
174 words
Computer-executable process steps stored on a computer-readable medium, which identify characters in a document on which the characters are formed, the process steps including: a scanning step which scans the document to obtain a gray-scale image of the document; a thresholding step which thresholds the gray-scale image to obtain a binary image; a segmenting step which segments the binary image to locate individual binary character images and which determines attributes of the binary character images; an extracting step which extracts a gray-scale character image from the gray-scale image using one of the binary character images as a template; a first recognition-processing step which recognition-processes the extracted gray-scale character image, which is extracted from the gray-scale image using the binary character image template, so as to obtain a first identity for the character; a second recognition-processing step which recognition-processes said one binary character image to obtain a second identity for the character; and a resolving step which resolves ambiguities between the first and second identity based on the character attributes determined in said segmenting step.
Claim 135
Dependent Claim
174 words
Computer-executable process steps stored on a computer-readable medium, which identify characters in a document on which the characters are formed, the process steps including: a scanning step which scans the document to obtain a gray-scale image of the document; a thresholding step which thresholds the gray-scale image to obtain a binary image; a segmenting step which segments the binary image to locate individual binary character images and which determines attributes of the binary character images; an extracting step which extracts a gray-scale character image from the gray-scale image using one of the binary character images as a template; a first recognition-processing step which recognition-processes the extracted gray-scale character image, which is extracted from the gray-scale image using the binary character image template, so as to obtain a first identity for the character; a second recognition-processing step which recognition-processes said one binary character image to obtain a second identity for the character; and a resolving step which resolves ambiguities between the first and second identity based on the character attributes determined in said segmenting step.
Claim 136
Dependent Claim
39 words
Computer-executable process steps according to claim 123, further comprising an obtaining step which obtains a histogram of density distributions in the gray-scale image, and derives the threshold for said thresholding step based on adjacency of bins in the histogram.
Claim 137
Dependent Claim
31 words
Computer-executable process steps according to claim 123, further comprising a removing step which removes underlines from the binary image, wherein the gray-scale character images are extracted from underlined gray-scale character image.
Claim 138
Dependent Claim
22 words
Computer-executable process steps according to claim 123, wherein said segmenting step includes the step of connected component analysis of the binary image.
Claim 139
Dependent Claim
25 words
Computer-executable process steps according to claim 126, further comprising a reconstructing step which reconstructs lines of characters so as to reconnect improperly separated connected components.
Claim 140
Dependent Claim
33 words
Computer-executable process steps according to claim 126, further comprising a separating step which separates text-type connected components from non-text-type connected components, wherein said recognition processing step is performed only on text-type connected components.
Claim 141
Dependent Claim
38 words
Computer-executable process steps according to claim 123, further comprising a reconstructing step which reconstructs reading order of characters on a document page and a storing step which stores the identities of characters based on the page-reconstructed reading order.
Claim 142
Dependent Claim
37 words
Computer-executable process steps according to claim 123, further comprising a step which performs connected component analysis of the binary image, and wherein the character attributes relate to physical image attributes of connected components in the binary image.
Claim 143
Dependent Claim
27 words
Computer-executable process steps according to claim 123, wherein the binary character image template is enlarged for use in extracting the gray-scale character image from the gray-scale image.
Claim 144
Dependent Claim
19 words
Computer-executable process steps according to claim 131, wherein the binary character image template is enlarged by approximately two pixels.
Claim 145
Dependent Claim
118 words
A character recognition apparatus for identifying characters in an image of the characters, comprising: locating means for locating lines of characters in the image; determining means for determining for each line font characteristics of the characters in the line, wherein the font characteristics include character appearance and character spacing of characters in the line; selecting means for selecting for each line one of plural different character recognition-processing techniques based on the font characteristics determined for the line by said determining means; extracting means for extracting individual character images from each line; and recognition-processing means for recognition-processing each extracted character image in accordance with the character recognition-processing technique selected for the line from which the character image is extracted.
Claim 146
Dependent Claim
118 words
A character recognition apparatus for identifying characters in an image of the characters, comprising: locating means for locating lines of characters in the image; determining means for determining for each line font characteristics of the characters in the line, wherein the font characteristics include character appearance and character spacing of characters in the line; selecting means for selecting for each line one of plural different character recognition-processing techniques based on the font characteristics determined for the line by said determining means; extracting means for extracting individual character images from each line; and recognition-processing means for recognition-processing each extracted character image in accordance with the character recognition-processing technique selected for the line from which the character image is extracted.
Claim 147
Dependent Claim
38 words
An apparatus according to claim 133, further comprising obtaining means for obtaining a histogram of density distributions in the gray-scale image, and for deriving the threshold for said thresholding step based on adjacency of bins in the histogram.
Claim 148
Dependent Claim
29 words
An apparatus according to claim 133, further comprising removing means for removing underlines from the binary image, wherein the gray-scale character images are extracted from underlined gray-scale character image.
Claim 149
Dependent Claim
18 words
An apparatus according to claim 133, wherein said segmenting means performs connected component analysis of the binary image.
Claim 150
Dependent Claim
23 words
An apparatus according to claim 136, further comprising reconstructing means for reconstructing lines of characters so as to reconnect improperly separated connected components.
Claim 151
Dependent Claim
30 words
An apparatus according to claim 136, further comprising separating means for separating text-type connected components from non-text-type connected components, wherein said recognition processing is performed only on text-type connected components.
Claim 152
Dependent Claim
32 words
An apparatus according to claim 133, further comprising reconstructing means for reconstructing reading order of characters on a document page and storing the identities of characters based on the page-reconstructed reading order.
Claim 153
Dependent Claim
15 words
An apparatus according to claim 133, further comprising inputting means for inputting the gray-scale image.
Claim 154
Dependent Claim
25 words
An apparatus according to claim 140, wherein said inputting means comprises scanning means for scanning in a document page on which the characters are formed.
Claim 155
Dependent Claim
23 words
An apparatus according to claim 140, wherein said inputting means comprises retrieving means for retrieving a gray-scale document image from a computer file.
Claim 156
Dependent Claim
44 words
An apparatus according to claim 142, wherein said inputting means comprises reading means for reading a computer file across a computerized local area network, and further comprising storing means for storing the identities of characters in a computer file on the local area network.
Claim 157
Dependent Claim
146 words
A character recognition apparatus for identifying characters in an image of the characters, comprising: a memory for storing the gray-scale image and for storing instruction sequences by which the gray-scale image is processed; and a processor for executing the stored instruction sequences; wherein the stored instruction sequences include process steps adapted to cause said processor to (a) locate lines of characters in the image, (b) determine for each line font characteristics of the characters in the line, wherein the font characteristics include character appearance and character spacing of characters in the line (c) select for each line one of plural different character recognition-processing techniques based on the font characteristics determined for the line, (d) extract individual character images from each line, and (e) recognition-process each extracted character image in accordance with the character recognition-processing technique selected for the line from which the character image is extracted.
Claim 158
Dependent Claim
146 words
A character recognition apparatus for identifying characters in an image of the characters, comprising: a memory for storing the gray-scale image and for storing instruction sequences by which the gray-scale image is processed; and a processor for executing the stored instruction sequences; wherein the stored instruction sequences include process steps adapted to cause said processor to (a) locate lines of characters in the image, (b) determine for each line font characteristics of the characters in the line, wherein the font characteristics include character appearance and character spacing of characters in the line (c) select for each line one of plural different character recognition-processing techniques based on the font characteristics determined for the line, (d) extract individual character images from each line, and (e) recognition-process each extracted character image in accordance with the character recognition-processing technique selected for the line from which the character image is extracted.
Claim 159
Dependent Claim
47 words
An apparatus according to claim 144, wherein the stored instruction sequences further include process steps to cause said processor to obtain a histogram of density distributions in the gray-scale image, and to derive the threshold for said thresholding step based on adjacency of bins in the histogram.
Claim 160
Dependent Claim
38 words
An apparatus according to claim 144, wherein the stored instruction sequences further include process steps to cause said processor to remove underlines from the binary image, wherein the gray-scale character images are extracted from underlined gray-scale character image.
Claim 161
Dependent Claim
21 words
An apparatus according to claim 144, wherein said segmenting step includes the step of connected component analysis of the binary image.
Claim 162
Dependent Claim
32 words
An apparatus according to claim 147, wherein the stored instruction sequences further include process steps to cause said processor to reconstruct lines of characters so as to reconnect improperly separated connected components.
Claim 163
Dependent Claim
38 words
An apparatus according to claim 147, wherein the stored instruction sequences further include process steps to cause said processor to separate text-type connected components from non-text-type connected components, wherein recognition processing is performed only on text-type connected components.
Claim 164
Dependent Claim
41 words
An apparatus according to claim 147, wherein the stored instruction sequences further include process steps to cause said processor to reconstruct reading order of characters on a document page and storing the identities of characters based on the page-reconstructed reading order.
Claim 165
Dependent Claim
24 words
An apparatus according to claim 144, wherein the stored instruction sequences further include process steps to cause said processor to input the gray-scale image.
Claim 166
Dependent Claim
23 words
An apparatus according to claim 151, wherein the gray-scale image is input by scanning a document page on which the characters are formed.
Claim 167
Dependent Claim
22 words
An apparatus according to claim 151, wherein the gray-scale image is input by retrieving a gray-scale document image from a computer file.
Claim 168
Dependent Claim
47 words
An apparatus according to claim 153, wherein the gray-scale image is input by reading a computer file across a computerized local area network, and wherein the stored instruction sequences include a step to store the identities of characters in a computer file on the local area network.
Claim 169
Dependent Claim
109 words
A character recognition method for identifying characters in an image of the characters, said method comprising the steps of: processing the image to locate lines of characters; determining for each line font characteristics of the characters in the line, wherein the font characteristics include character appearance and character spacing of characters in the line; selecting for each line one of plural different character recognition-processing techniques based on the font characteristics determined for the line in said determining step; extracting individual character images from each line; and recognition-processing each extracted character image in accordance with the character recognition-processing technique selected for the line from which the character image is extracted.
Claim 170
Dependent Claim
109 words
A character recognition method for identifying characters in an image of the characters, said method comprising the steps of: processing the image to locate lines of characters; determining for each line font characteristics of the characters in the line, wherein the font characteristics include character appearance and character spacing of characters in the line; selecting for each line one of plural different character recognition-processing techniques based on the font characteristics determined for the line in said determining step; extracting individual character images from each line; and recognition-processing each extracted character image in accordance with the character recognition-processing technique selected for the line from which the character image is extracted.
Claim 171
Dependent Claim
37 words
A method according to claim 155, further comprising the steps of obtaining a histogram of density distributions in the gray-scale image, and deriving the threshold for said thresholding step based on adjacency of bins in the histogram.
Claim 172
Dependent Claim
29 words
A method according to claim 155, further comprising the step of removing underlines from the binary image, wherein the gray-scale character images are extracted from underlined gray-scale character image.
Claim 173
Dependent Claim
21 words
A method according to claim 155, wherein said segmenting step includes the step of connected component analysis of the binary image.
Claim 174
Dependent Claim
23 words
A method according to claim 158, further comprising the step of reconstructing lines of characters so as to reconnect improperly separated connected components.
Claim 175
Dependent Claim
29 words
A method according to claim 158, further comprising the step of separating text-type connected components from non-text-type connected components, wherein recognition processing is performed only on text-type connected components.
Claim 176
Dependent Claim
32 words
A method according to claim 155, further comprising the step of reconstructing reading order of characters on a document page and storing the identities of characters based on the page-reconstructed reading order.
Claim 177
Dependent Claim
15 words
A method according to claim 155, further comprising the step of inputting the gray-scale image.
Claim 178
Dependent Claim
25 words
A method according to claim 162, wherein said inputting step comprises the step of scanning in a document page on which the characters are formed.
Claim 179
Dependent Claim
23 words
A method according to claim 162, wherein said inputting step comprises the step of retrieving a gray-scale document image from a computer file.
Claim 180
Dependent Claim
44 words
A method according to claim 164, wherein said inputting step comprises the step of reading a computer file across a computerized local area network, and further comprising the step of storing the identities of characters in a computer file on the local area network.
Claim 181
Dependent Claim
130 words
Computer-executable process steps stored on a computer-readable medium, which identify characters in an image of the characters, the process steps including: a processing step which processes the image to locate lines of characters; a determining step which determines for each line font characteristics of the characters in line, wherein the font characteristics include character appearance and character spacing of characters in the line; a selecting step which selects for each line one of plural different character recognition-processing techniques based on the font characteristics determined for the line in said determining step; an extracting step which extracts individual character images from each line; and a recognition-processing step which recognition-processes each extracted character image in accordance with the character recognition-processing technique selected for the line from which the character image is extracted.
Claim 182
Dependent Claim
130 words
Computer-executable process steps stored on a computer-readable medium, which identify characters in an image of the characters, the process steps including: a processing step which processes the image to locate lines of characters; a determining step which determines for each line font characteristics of the characters in line, wherein the font characteristics include character appearance and character spacing of characters in the line; a selecting step which selects for each line one of plural different character recognition-processing techniques based on the font characteristics determined for the line in said determining step; an extracting step which extracts individual character images from each line; and a recognition-processing step which recognition-processes each extracted character image in accordance with the character recognition-processing technique selected for the line from which the character image is extracted.
Claim 183
Dependent Claim
40 words
Computer-executable process steps according to claim 166, further comprising an obtaining step which obtains a histogram of density distributions in the gray-scale image, and which derives the threshold for said thresholding step based on adjacency of bins in the histogram.
Claim 184
Dependent Claim
31 words
Computer-executable process steps according to claim 166, further comprising a removing step which removes underlines from the binary image, wherein the gray-scale character images are extracted from underlined gray-scale character image.
Claim 185
Dependent Claim
22 words
Computer-executable process steps according to claim 166, wherein said segmenting step includes the step of connected component analysis of the binary image.
Claim 186
Dependent Claim
25 words
Computer-executable process steps according to claim 169, further comprising a reconstructing step which reconstructs lines of characters so as to reconnect improperly separated connected components.
Claim 187
Dependent Claim
33 words
Computer-executable process steps according to claim 169, further comprising a separating step which separates text-type connected components from non-text-type connected components, wherein said recognition processing step is performed only on text-type connected components.
Claim 188
Dependent Claim
38 words
Computer-executable process steps according to claim 166, further comprising a reconstructing step which reconstructs reading order of characters on a document page and a storing step which stores the identities of characters based on the page-reconstructed reading order.
Claim 189
Dependent Claim
17 words
Computer-executable process steps according to claim 166, further comprising an inputting step which inputs the gray-scale image.
Claim 190
Dependent Claim
27 words
Computer-executable process steps according to claim 173, wherein said inputting step comprises a scanning step which scans in a document page on which the characters are formed.
Claim 191
Dependent Claim
25 words
Computer-executable process steps according to claim 173, wherein said inputting step comprises a retrieving step which retrieves a gray-scale document image from a computer file.
Claim 192
Dependent Claim
47 words
Computer-executable process steps according to claim 175, wherein said inputting step comprises a reading step which reads a computer file across a computerized local area network, and further comprising a storing step which stores the identities of characters in a computer file on the local area network.
References Cited (Backward Citations)
US4461027A
Cited
US4741045A
Cited
US4633502A
Cited
US4742556A
Cited
US4823194A
Cited
JPS63155390A
Cited
US4944022A
Cited
US5278918A
Cited
US5073953A
Cited
US4912569A
Cited
US5048096A
Cited
US5048109A
Cited
US5081690A
Cited
WO1992006449A1
Cited
US5181255A
Cited
US5317652A
Cited
US5325447A
Cited
US5350303A
Cited
US5321770A
Cited
US5384863A
Cited
US5491760A
Cited
US5303313A
Cited
EP0555023A2
Cited
US5337370A
Cited
US5438630A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 2001-09-14 | 2002-03-16 | 2002-09-15 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2005-09-14 | 2006-03-16 | 2006-09-15 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2009-09-14 | 2010-03-16 | 2010-09-15 | $7700.00 | Window Closed / Lapsed |