US5077809A — Optical character recognition

Claim 1

Independent Claim 205 words

A method for recognition of a designated character by an optical scanner comprising the steps of: providing an array R of pixels of minimal size, the array containing the designated character therein; providing three pixel slices across the array at three predetermined positions, one of which is oriented at right angles to the other two slices, to produce first, second and third slice sequences of alternating bands of dark and light pixels for the designated character; providing four pixel masks, to determine first, second, third and fourth mask sequences of alternating bands of dark and light pixels at peripheral regions of the pixel array for the designated character; providing a candidate set of characters, each candidate character being characterized by first, second and third slice light and dark sequences and first, second, third and fourth mask light and dark sequences that correspond, respectively, to the first, second and third slice sequences and the first, second, third and fourth mask sequences for the designated character; forming an identification set of all characters in the candidate set whose slice sequences and mask sequences agree with the slice sequences and mask sequences for the designated character; and identifying a character in the identification set as the designated character.

Claim 2

Independent Claim 205 words

A method for recognition of a designated character by an optical scanner comprising the steps of: providing an array R of pixels of minimal size, the array containing the designated character therein; providing three pixel slices across the array at three predetermined positions, one of which is oriented at right angles to the other two slices, to produce first, second and third slice sequences of alternating bands of dark and light pixels for the designated character; providing four pixel masks, to determine first, second, third and fourth mask sequences of alternating bands of dark and light pixels at peripheral regions of the pixel array for the designated character; providing a candidate set of characters, each candidate character being characterized by first, second and third slice light and dark sequences and first, second, third and fourth mask light and dark sequences that correspond, respectively, to the first, second and third slice sequences and the first, second, third and fourth mask sequences for the designated character; forming an identification set of all characters in the candidate set whose slice sequences and mask sequences agree with the slice sequences and mask sequences for the designated character; and identifying a character in the identification set as the designated character.

Claim 3

Dependent Claim 24 words

The method of claim 1, wherein said candidate set comprises all upper case letters, all lower case letters and all numerals in a font.

Claim 4

Dependent Claim 16 words

The method of claim 1, wherein said candidate set further comprises at least one punctuation mark.

Claim 5

Dependent Claim 42 words

The method of claim 1, wherein said four peripheral regions of said pixel array are adjacent to the top boundary of said array, the left boundary of said array, the bottom boundary of said array and the right boundary of said array.

Claim 6

Dependent Claim 118 words

The method of claim 1, wherein said step of forming said identification set comprises the steps of: forming first, second and third sets of said candidate characters that comprise all candidate characters whose first, second and third slice sequences agree, respectively, with said corresponding first, second and third slice sequences of said designated character; forming fourth, fifth, sixth and seventh sets of said candidate characters that comprise all candidate characters whose first, second, third and fourth mask sequences agree, respectively, with said corresponding first, second, third and fourth mask sequences for said designated character; and forming the intersection of the first, second, third, fourth, fifth, sixth and seventh sets and identifying this intersection set with said identification set.

Claim 7

Dependent Claim 164 words

The method of claim 1, wherein said step of forming said identification set comprises the steps of: defining a set SO of all said candidate characters; for said first, second, third and fourth mask sequences numbered j=1,2,3 and 4, respectively, of said designated character and said candidate characters, defining a set Sj (j=1,2,3,4) of all said candidate characters c for which (i) c belongs to the set S(j-l) and (ii) said mask sequence No. j for the character c is identical to said mask sequence No. j for said designated character; for said first, second and third slice sequences numbered j=5,6 and 7, respectively, of said designated character and said candidate characters, defining a set Sj (j=5,6,7) of all said candidate characters c' for which (i) c' belongs to the set S(j-l) and (ii) said slice sequence No. j for the character c' is identical to said slice sequence No. j for said designated character; and identifying the set S7 with said identification set.

Claim 8

Dependent Claim 164 words

The method of claim 1, wherein said step of forming said identification set comprises the steps of: defining a set SO of all said candidate characters; for said first, second and third slice sequences numbered j=1,2 and 3, respectively, of said designated character and said candidate characters, defining a set Sj (j=1,2,3) of all said candidate characters c for which (i) c belongs to the set S(j-1) and (ii) said slice sequence No. j for the character c is identical to said slice sequence No. j for said designated character; for said first, second, third and fourth mask sequences numbered j=4,5,6 and 7, respectively, of said designated character and said candidate characters, defining a set Sj (j=4,5,6,7) of all said candidate characters c' for which (i) c' belongs to the set S(j-l) and (ii) said mask sequence No. j for the character c' is identical to said mask sequence No. j for said designated character; and identifying the set S7 with said identification set.

Claim 9

Dependent Claim 176 words

The method of claim 1, further comprising the steps of: representing each character in said identification set as an array of dark and light pixels; for each character in said identification set, identifying a common sub-region SR in the pixel array representation for which (i) said identification set contains at least two identification characters and (ii) the sub-region SR for one identification character contains a substantially different fraction of dark pixels than the fraction of dark pixels contained in the sub-region SR for a second identification character; determining the number of dark pixels in the sub-region SR of said array R of pixels that represents said designated character; comparing the number of dark pixels in the sub-region SR of said designated character, the first identification character and the second identification character; and identifying said designated character with the first identification character or with the second identification character, according as the number of dark pixels in the sub-region SR for said designated character is closer to the first identification character or to the second identification character, respectively.

Claim 10

Dependent Claim 181 words

The method of claim 1 , further comprising the steps of: representing each character in said identification set as an array of dark and light pixels; for each character in said identification set, identifying a first common sub-region SR1 and a second common sub-region SR2 in the pixel array representation for which (i) said identification set contains at least two identification characters and (ii) the sub-regions SR and SR2 for a first identification character contain approximately the same number of dark pixels, and the sub-regions SR1 and SR2 for a second identification character contain substantially different numbers of dark pixels; determining the number of dark pixels in the sub-regions SR1 and SR2 of said array R of pixels that represents said designated character; comparing the number of dark pixels in the sub-regions SR1 and SR2 of said designated character; and identifying said designated character with the first identification character or with the second identification character, according as the number of dark pixels in the sub-regions SR1 and SR2 for said designated character are approximately the same or are substantially different, respectively.

Claim 11

Dependent Claim 141 words

The method of claim 1, further comprising the steps of: representing each character in said identification set as an array of dark and light pixels; for each character in said identification set, identifying a common sub-region SR in the pixel array representation for which (i) said identification set contains at least two identification characters and (ii) the sub-region SR for a first identification character contains no dark pixels therein and contains a plurality of dark pixels therein for a second identification character; determining the number of dark pixels contained in the sub-region SR of said array of pixels R that represents said designated character; identifying said designated character with the first identification character or with the second identification character, according as the number of dark pixels in the sub-region SR of said designated character is zero or greater than zero, respectively.

Claim 12

Dependent Claim 173 words

The method of claim 1, further comprising the steps of: representing each character in said identification set as an array of dark and light pixels; for each character in said identification set, identifying a common sub-region SR in the pixel array representation for which (i) said identification set contains at least two identification characters and (ii) within the sub-region SR, a first identification character has a continuous band of dark pixels whose vertical height is substantially different from the vertical height of a continuous band of dark pixels for a second identification character; determining the vertical height of a continuous band of dark pixels contained in the sub-region SR of said array R of pixels that represents said designated character; and identifying said designated character with the first identification character or with the second identification character, according as the vertical height of the continuous dark pixel band of said designated character is closer to the vertical height of the continuous dark pixel band of the first identification character or the second identification character.

Claim 13

Dependent Claim 173 words

The method of claim 1, further comprising the steps of: representing each character in said identification set as an array of dark and light pixels; for each character in said identification set, identifying a common sub-region SR in the pixel array representation for which (i) said identification set contains at least two identification characters and (ii) within the sub-region SR, a first identification character has a continuous band of dark pixels whose horizontal width is substantially different from the horizontal width of a continuous band of dark pixels for a second identification character; determining the horizontal width of a continuous band of dark pixels contained in the sub-region SR of said array R of pixels that represents said designated character; and identifying said designated character with the first identification character or with the second identification character, according as the horizontal width of the continuous dark pixel band of said designated character is closer to the horizontal width of the continuous dark pixel band of the first identification character or the second identification character.

Claim 14

Dependent Claim 175 words

The method of claim 1, further comprising the steps of: representing each character in said identification set as an array of dark and light pixels; for each character in said identification set, identifying a common sub-region SR in the pixel array representation for which (i) said identification set contains at least two identification characters and (ii) within the interior of the sub-region SR, a first identification character is represented as a single, continuous band of dark pixels and a second identification character is represented as two or more continuous bands of dark pixels that are spaced apart from each other within the sub-region SR; determining whether said designated character within the interior of the sub-array SR is represented as a single, continuous band of dark pixels; and identifying said designated character with the first identification character if said designated character is represented within the interior of the sub-array SR as a single, continuous band of dark pixels, and identifying said designated character with the second identification character if said designated character is not so represented.

Claim 15

Dependent Claim 88 words

The method of claim 1, further comprising the steps of: representing each character in said identification set as an array of dark and light pixels; for each character in said identification set, determining the horizontal width in pixels of the character; determining the horizontal width in pixels of said designated character; and identifying a character in said identification set with said designated character if the horizontal width of the character in said identification set differs by at most one pixel from the horizontal width of said designated character.

Claim 16

Dependent Claim 131 words

The method of claim 1, wherein said identification set for said designated character contains an upper case character and a lower case character for the same letter, said designated character is part of a word comprising at least three letters, and the method further comprises the steps of: determining whether at least one other character in the word that includes said designated character is a lower case letter; identifying said designated character with a lower case letter in said identification set, if at least one other character in the word that contains said designated character is a lower case letter; and identifying said designated character with an upper case letter in said identification set, if no other character in the word that contains said designated character is a lower case letter.

Claim 17

Dependent Claim 119 words

The method of claim 1, wherein said identification set for said designated number contains a letter and a numeral, said designated character is part of a word comprising at least one character in addition to said designated character, and the method further comprises the steps of: determining whether at least one other character in the word that includes said designated character is a numeral; identifying said designated character with a numeral in said identification set, if at least one other character in the word that contains said designated character is a numeral; and identifying said designated character with a numeral in said identification set, if no other character in the word that contains said designated character is a numeral.

Claim 18

Independent Claim 810 words

A method for recognition of a designated character by an optical scanner comprising the steps of: providing an approximately rectangular array R of pixels of minimal size representing the designated character, the array being M pixels in height by N pixels in width; providing a bottom mask of pixels that is an approximately rectangular sub-array of the array R, approximately N pixels in width and approximately M/4 pixels in height, that extends upward from the bottom of the array R; forming a bottom mask sequence from the intersection of the bottom mask with the designated character in the array R; providing a left mask of pixels that is approximately a rectangular sub-array of the array R, approximately M pixels in height and approximately N/4 pixels in width, that extends rightward from the left side of the array R; forming a left mask sequence from the intersection of the left mask with the designated character in the array R; providing a right mask of pixels that is an approximately rectangular sub-array of the array R, approximately M pixels in height and approximately N/4 pixels in width, that extends leftward from the right side of the array R; forming a right mask sequence from the intersection of the right mask with the designated character in the array R; providing a top mask of pixels that is an approximately rectangular sub-array of the array R, approximately N pixels in width and approximately M/4 pixels in height, that extends downward from the top of the array R; forming a top mask sequence from the intersection of the top mask with the designated character in the array R; providing a vertical slice of pixels that is an approximately rectangular sub-array of the array R, at least one pixel in width, extending from the top of the array R to the bottom of the array R at a position approximately equidistant from the two sides of the array R; forming a vertical slice sequence from the intersection of the vertical slice with the designated character in the array R; providing a first horizontal slice of pixels that is an approximately rectangular sub-array of the array R, at least one pixel in height, extending from the left side of the array R to the right side of the array R at a position approximately M/3 pixels from the bottom of the array R; forming a first horizontal slice sequence from the intersection of the first horizontal slice with the designated character in the array R; providing a second horizontal slice of pixels that is an approximately horizontal sub-array of the array R, at least one pixel in height, extending from the left side of the array R to the right side of the array R at a position approximately M/3 pixels from the top of the array R; forming a second horizontal slice sequence from the intersection of the second horizontal slice with the designated character in the array R; providing a candidate set of characters, each represented by an array of dark and light pixels and, for each such candidate character, providing a bottom mask sequence, a left mask sequence, a right mask sequence, a top mask sequence, a vertical slice sequence, a first horizontal slice sequence and a second horizontal slice sequence as the corresponding sequence is formed for the designated character; determining a first subset of the candidate set of characters, each character of which has a bottom mask sequence that agrees with the bottom mask sequence of the designated character; determining a second subset of the candidate set of characters, each character of which has a left mask sequence that agrees with the left mask sequence of the designated character; determining a third subset of the candidate set of characters, each character of which has a right mask sequence that agrees with the right mask sequence of the designated character; determining a fourth subset of the candidate set of characters, each character of which has a top mask sequence that agrees with the top mask sequence of the designated character; determining a fifth subset of the candidate set of characters, each character of which has a vertical slice sequence that agrees with the vertical slice sequence of the designated character; determining a sixth subset of the candidate set of characters, each character of which has a first horizontal slice sequence that agrees with the first horizontal slice sequence of the designated character; determining a seventh subset of the candidate set of characters, each character of which has a second horizontal slice sequence that agrees with the second horizontal slice sequence of the designated character; determining an identification set of characters that belong to all of the first, second, third, fourth, fifth, sixth and seventh subsets of the candidate set of characters; and identifying a character in the identification set as the designated character.

Claim 19

Independent Claim 810 words

A method for recognition of a designated character by an optical scanner comprising the steps of: providing an approximately rectangular array R of pixels of minimal size representing the designated character, the array being M pixels in height by N pixels in width; providing a bottom mask of pixels that is an approximately rectangular sub-array of the array R, approximately N pixels in width and approximately M/4 pixels in height, that extends upward from the bottom of the array R; forming a bottom mask sequence from the intersection of the bottom mask with the designated character in the array R; providing a left mask of pixels that is approximately a rectangular sub-array of the array R, approximately M pixels in height and approximately N/4 pixels in width, that extends rightward from the left side of the array R; forming a left mask sequence from the intersection of the left mask with the designated character in the array R; providing a right mask of pixels that is an approximately rectangular sub-array of the array R, approximately M pixels in height and approximately N/4 pixels in width, that extends leftward from the right side of the array R; forming a right mask sequence from the intersection of the right mask with the designated character in the array R; providing a top mask of pixels that is an approximately rectangular sub-array of the array R, approximately N pixels in width and approximately M/4 pixels in height, that extends downward from the top of the array R; forming a top mask sequence from the intersection of the top mask with the designated character in the array R; providing a vertical slice of pixels that is an approximately rectangular sub-array of the array R, at least one pixel in width, extending from the top of the array R to the bottom of the array R at a position approximately equidistant from the two sides of the array R; forming a vertical slice sequence from the intersection of the vertical slice with the designated character in the array R; providing a first horizontal slice of pixels that is an approximately rectangular sub-array of the array R, at least one pixel in height, extending from the left side of the array R to the right side of the array R at a position approximately M/3 pixels from the bottom of the array R; forming a first horizontal slice sequence from the intersection of the first horizontal slice with the designated character in the array R; providing a second horizontal slice of pixels that is an approximately horizontal sub-array of the array R, at least one pixel in height, extending from the left side of the array R to the right side of the array R at a position approximately M/3 pixels from the top of the array R; forming a second horizontal slice sequence from the intersection of the second horizontal slice with the designated character in the array R; providing a candidate set of characters, each represented by an array of dark and light pixels and, for each such candidate character, providing a bottom mask sequence, a left mask sequence, a right mask sequence, a top mask sequence, a vertical slice sequence, a first horizontal slice sequence and a second horizontal slice sequence as the corresponding sequence is formed for the designated character; determining a first subset of the candidate set of characters, each character of which has a bottom mask sequence that agrees with the bottom mask sequence of the designated character; determining a second subset of the candidate set of characters, each character of which has a left mask sequence that agrees with the left mask sequence of the designated character; determining a third subset of the candidate set of characters, each character of which has a right mask sequence that agrees with the right mask sequence of the designated character; determining a fourth subset of the candidate set of characters, each character of which has a top mask sequence that agrees with the top mask sequence of the designated character; determining a fifth subset of the candidate set of characters, each character of which has a vertical slice sequence that agrees with the vertical slice sequence of the designated character; determining a sixth subset of the candidate set of characters, each character of which has a first horizontal slice sequence that agrees with the first horizontal slice sequence of the designated character; determining a seventh subset of the candidate set of characters, each character of which has a second horizontal slice sequence that agrees with the second horizontal slice sequence of the designated character; determining an identification set of characters that belong to all of the first, second, third, fourth, fifth, sixth and seventh subsets of the candidate set of characters; and identifying a character in the identification set as the designated character.

Claim 20

Dependent Claim 21 words

The method of claim 17, wherein said candidate set comprises all upper case letters, all lower case letters and all numerals.

Claim 21

Dependent Claim 16 words

The method of claim 18, wherein said candidate set further comprises at least one punctuation mark.

Claim 22

Dependent Claim 176 words

The method of claim 17, further comprising the steps of: representing each character in said identification set as an array of dark and light pixels; for each character in said identification set, identifying a common sub-region SR in the pixel array representation for which (i) said identification set contains at least two identification characters and (ii) the sub-region SR for one identification character contains a substantially different fraction of dark pixels than the fraction of dark pixels contained in the sub-region SR for a second identification character; determining the number of dark pixels in the sub-region SR of said array R of pixels that represents said designated character; comparing the number of dark pixels in the sub-region SR of said designated character, the first identification character and the second identification character; and identifying said designated character with the first identification character or with the second identification character, according as the number of dark pixels in the sub-region SR for said designated character is closer to the first identification character or to the second identification character, respectively.

Claim 23

Dependent Claim 180 words

The method of claim 17, further comprising the steps of: representing each character in said identification set as an array of dark and light pixels; for each character in said identification set, identifying a first common sub-region SR1 and a second common sub-region SR2 in the pixel array representation for which (i) said identification set contains at least two identification characters and (ii) the sub-regions SR and SR2 for a first identification character contain approximately the same number of dark pixels, and the sub-regions SR1 and SR2 for a second identification character contain substantially different numbers of dark pixels; determining the number of dark pixels in the sub-regions SR1 and SR2 of said array R of pixels that represents said designated character; comparing the number of dark pixels in the sub-regions SR1 and SR2 of said designated character; and identifying said designated character with the first identification character or with the second identification character, according as the number of dark pixels in the sub-regions SR1 and SR2 for said designated character are approximately the same or are substantially different, respectively.

Claim 24

Dependent Claim 141 words

The method of claim 17, further comprising the steps of: representing each character in said identification set as an array of dark and light pixels; for each character in said identification set, identifying a common sub-region SR in the pixel array representation for which (i) said identification set contains at least two identification characters and (ii) the sub-region SR for a first identification character contains no dark pixels therein and contains a plurality of dark pixels therein for a second identification character; determining the number of dark pixels contained in the sub-region SR of said array of pixels R that represents said designated character; identifying said designated character with the first identification character or with the second identification character, according as the number of dark pixels in the sub-region SR of said designated character is zero or greater than zero, respectively.

Claim 25

Dependent Claim 173 words

The method of claim 17, further comprising the steps of: representing each character in said identification set as an array of dark and light pixels; for each character in said identification set, identifying a common sub-region SR in the pixel array representation for which (i) said identification set contains at least two identification characters and (ii) within the sub-region SR, a first identification character has a continuous band of dark pixels whose vertical height is substantially different from the vertical height of a continuous band of dark pixels for a second identification character; determining the vertical height of a continuous band of dark pixels contained in the sub-region SR of said array R of pixels that represents said designated character; and identifying said designated character with the first identification character or with the second identification character, according as the vertical height of the continuous dark pixel band of said designated character is closer to the vertical height of the continuous dark pixel band of the first identification character or the second identification character.

Claim 26

Dependent Claim 173 words

The method of claim 17, further comprising the steps of: representing each character in said identification set as an array of dark and light pixels; for each character in said identification set, identifying a common sub-region SR in the pixel array representation for which (i) said identification set contains at least two identification characters and (ii) within the sub-region SR, a first identification character has a continuous band of dark pixels whose horizontal width is substantially different from the horizontal width of a continuous band of dark pixels for a second identification character; determining the horizontal width of a continuous band of dark pixels contained in the sub-region SR of said array R of pixels that represents said designated character; and identifying said designated character with the first identification character or with the second identification character, according as the horizontal width of the continuous dark pixel band of said designated character is closer to the horizontal width of the continuous dark pixel band of the first identification character or the second identification character.

Claim 27

Dependent Claim 175 words

The method of claim 17, further comprising the steps of: representing each character in said identification set as an array of dark and light pixels; for each character in said identification set, identifying a common sub-region SR in the pixel array representation for which (i) said identification set contains at least two identification characters and (ii) within the interior of the sub-region SR, a first identification character is represented as a single, continuous band of dark pixels and a second identification character is represented as two or more continuous bands of dark pixels that are spaced apart from each other within the sub-region SR; determining whether said designated character within the interior of the sub-array SR is represented as a single, continuous band of dark pixels; and identifying said designated character with the first identification character if said designated character is represented within the interior of the sub-array SR as a single, continuous band of dark pixels, and identifying said designated character with the second identification character if said designated character is not so represented.

Claim 28

Dependent Claim 88 words

The method of claim 17, further comprising the steps of: representing each character in said identification set as an array of dark and light pixels; for each character in said identification set, determining the horizontal width in pixels of the character; determining the horizontal width in pixels of said designated character; and identifying a character in said identification set with said designated character if the horizontal width of the character in said identification set differs by at most one pixel from the horizontal width of said designated character.

Claim 29

Dependent Claim 131 words

The method of claim 17, wherein said identification set for said designated character contains an upper case character and a lower case character for the same letter, said designated character is part of a word comprising at least three letters, and the method further comprises the steps of: determining whether at least one other character in the word that includes said designated character is a lower case letter; identifying said designated character with a lower case letter in said identification set, if at least one other character in the word that contains said designated character is a lower case letter; and identifying said designated character with an upper case letter in said identification set, if no other character in the word that contains said designated character is a lower case letter.

Claim 30

Dependent Claim 119 words

The method of claim 17, wherein said identification set for said designated number contains a letter and a numeral, said designated character is part of a word comprising at least one character in addition to said designated character, and the method further comprises the steps of: determining whether at least one other character in the word that includes said designated character is a numeral; identifying said designated character with a numeral in said identification set, if at least one other character in the word that contains said designated character is a numeral; and identifying said designated character with a letter in said identification set, if no other character in the word that contains said designated character is a numeral.

Claim 31

Independent Claim 201 words

A method for recognition of a designated character by an optical scanner comprising the steps of: providing an array R of pixels of minimal size, the array containing the designated character therein; providing two pixel slices across the array at two predetermined positions, one slice being oriented at right angles to the other slices, to produce first and second slice sequences of alternating bands of dark and light pixels for the designated character; providing four pixel masks, to determine first, second, third and fourth mask sequences of alternating bands of dark and light pixels at peripheral regions of the pixel array for the designated character; providing a candidate set of characters, each candidate character being characterized by first and second slice light and dark sequences and first, second, third and fourth mask light and dark sequences that correspond, respectively, to the first, second and third slice sequences and the first, second, third and fourth mask sequences for the designated character; forming an identification set of all characters in the candidate set whose slice sequences and mask sequences agree with the slice sequences and mask sequences for the designated character; and identifying a character in the identification set as the designated character.

Claim 32

Independent Claim 201 words

A method for recognition of a designated character by an optical scanner comprising the steps of: providing an array R of pixels of minimal size, the array containing the designated character therein; providing two pixel slices across the array at two predetermined positions, one slice being oriented at right angles to the other slices, to produce first and second slice sequences of alternating bands of dark and light pixels for the designated character; providing four pixel masks, to determine first, second, third and fourth mask sequences of alternating bands of dark and light pixels at peripheral regions of the pixel array for the designated character; providing a candidate set of characters, each candidate character being characterized by first and second slice light and dark sequences and first, second, third and fourth mask light and dark sequences that correspond, respectively, to the first, second and third slice sequences and the first, second, third and fourth mask sequences for the designated character; forming an identification set of all characters in the candidate set whose slice sequences and mask sequences agree with the slice sequences and mask sequences for the designated character; and identifying a character in the identification set as the designated character.

Claim 33

Independent Claim 695 words

A method for recognition of a designated character by an optical scanner comprising the steps of: providing an approximately rectangular array R of pixels of minimal size representing the designated character, the array being M pixels in height by N pixels in width; providing a bottom mask of pixels that is an approximately rectangular sub-array of the array R, approximately N pixels in width and approximately M/4 pixels in height, that extends upward from the bottom of the array R; forming a bottom mask sequence from the intersection of the bottom mask with the designated character in the array R; providing a left mask of pixels that is approximately a rectangular sub-array of the array R, approximately M pixels in height and approximately N/4 pixels in width, that extends rightward from the left side of the array R; forming a left mask sequence from the intersection of the left mask with the designated character in the array R; providing a right mask of pixels that is an approximately rectangular sub-array of the array R, approximately M pixels in height and approximately N/4 pixels in width, that extends leftward from the right side of the array R; forming a right mask sequence from the intersection of the right mask with the designated character in the array R; providing a top mask of pixels that is an approximately rectangular sub-array of the array R, approximately N pixels in width and approximately M/4 pixels in height, that extends downward from the top of the array R; forming a top mask sequence from the intersection of the top mask with the designated character in the array R; providing a vertical slice of pixels that is an approximately rectangular sub-array of the array R, at least one pixel in width, extending from the top of the array R to the bottom of the array R at a position approximately equidistant from the two sides of the array R; forming a vertical slice sequence from the intersection of the vertical slice with the designated character in the array R; providing a horizontal slice of pixels that is an approximately rectangular sub-array of the array R, at least one pixel in height, extending from the left side of the array R to the right side of the array R at a position that is a predetermined number of pixels from the bottom of the array R; forming a horizontal slice sequence from the intersection of the horizontal slice with the designated character in the array R; providing a candidate set of characters, each represented by an array of dark and light pixels and, for each such candidate character, providing a bottom mask sequence, a left mask sequence, a right mask sequence, a top mask sequence, a vertical slice sequence and a horizontal slice sequence as the corresponding sequence is formed for the designated character; determining a first subset of the candidate set of characters, each character of which has a bottom mask sequence that agrees with the bottom mask sequence of the designated character; determining a second subset of the candidate set of characters, each character of which has a left mask sequence that agrees with the left mask sequence of the designated character; determining a third subset of the candidate set of characters, each character of which has a right mask sequence that agrees with the right mask sequence of the designated character; determining a fourth subset of the candidate set of characters, each character of which has a top mask sequence that agrees with the top mask sequence of the designated character; determining a fifth subset of the candidate set of characters, each character of which has a vertical slice sequence that agrees with the vertical slice sequence of the designated character; determining a sixth subset of the candidate set of characters, each character of which has a horizontal slice sequence that agrees with the horizontal slice sequence of the designated character; determining an identification set of characters that belong to all of the first, second, third, fourth, fifth and sixth subsets of the candidate set of characters; and identifying a character in the identification set as the designated character.

Claim 34

Independent Claim 695 words

A method for recognition of a designated character by an optical scanner comprising the steps of: providing an approximately rectangular array R of pixels of minimal size representing the designated character, the array being M pixels in height by N pixels in width; providing a bottom mask of pixels that is an approximately rectangular sub-array of the array R, approximately N pixels in width and approximately M/4 pixels in height, that extends upward from the bottom of the array R; forming a bottom mask sequence from the intersection of the bottom mask with the designated character in the array R; providing a left mask of pixels that is approximately a rectangular sub-array of the array R, approximately M pixels in height and approximately N/4 pixels in width, that extends rightward from the left side of the array R; forming a left mask sequence from the intersection of the left mask with the designated character in the array R; providing a right mask of pixels that is an approximately rectangular sub-array of the array R, approximately M pixels in height and approximately N/4 pixels in width, that extends leftward from the right side of the array R; forming a right mask sequence from the intersection of the right mask with the designated character in the array R; providing a top mask of pixels that is an approximately rectangular sub-array of the array R, approximately N pixels in width and approximately M/4 pixels in height, that extends downward from the top of the array R; forming a top mask sequence from the intersection of the top mask with the designated character in the array R; providing a vertical slice of pixels that is an approximately rectangular sub-array of the array R, at least one pixel in width, extending from the top of the array R to the bottom of the array R at a position approximately equidistant from the two sides of the array R; forming a vertical slice sequence from the intersection of the vertical slice with the designated character in the array R; providing a horizontal slice of pixels that is an approximately rectangular sub-array of the array R, at least one pixel in height, extending from the left side of the array R to the right side of the array R at a position that is a predetermined number of pixels from the bottom of the array R; forming a horizontal slice sequence from the intersection of the horizontal slice with the designated character in the array R; providing a candidate set of characters, each represented by an array of dark and light pixels and, for each such candidate character, providing a bottom mask sequence, a left mask sequence, a right mask sequence, a top mask sequence, a vertical slice sequence and a horizontal slice sequence as the corresponding sequence is formed for the designated character; determining a first subset of the candidate set of characters, each character of which has a bottom mask sequence that agrees with the bottom mask sequence of the designated character; determining a second subset of the candidate set of characters, each character of which has a left mask sequence that agrees with the left mask sequence of the designated character; determining a third subset of the candidate set of characters, each character of which has a right mask sequence that agrees with the right mask sequence of the designated character; determining a fourth subset of the candidate set of characters, each character of which has a top mask sequence that agrees with the top mask sequence of the designated character; determining a fifth subset of the candidate set of characters, each character of which has a vertical slice sequence that agrees with the vertical slice sequence of the designated character; determining a sixth subset of the candidate set of characters, each character of which has a horizontal slice sequence that agrees with the horizontal slice sequence of the designated character; determining an identification set of characters that belong to all of the first, second, third, fourth, fifth and sixth subsets of the candidate set of characters; and identifying a character in the identification set as the designated character.

Claim 35

Dependent Claim 29 words

The method of claim 30, further comprising the step of choosing said position of said horizontal slice sequence as approximately M/3 pixels above said bottom of said array R.

Claim 36

Dependent Claim 29 words

The method of claim 30, further comprising the step of choosing said position of said horizontal slice sequence as approximately M/3 pixels below said top of said array R.

Annuity Stage	Window Opens	Payment Deadline	Surcharge Period Until	Estimated Fee	Status
3.5 Years (Year 4)	1994-12-30	1995-07-01	1995-12-31	$2000.00	Window Closed / Lapsed
7.5 Years (Year 8)	1998-12-30	1999-07-01	1999-12-31	$3760.00	Window Closed / Lapsed
11.5 Years (Year 12)	2002-12-30	2003-07-01	2003-12-31	$7700.00	Window Closed / Lapsed

Optical character recognition

Abstract

Inventors

Bibliographic Details

Claims Analysis

References Cited (Backward Citations)

USPTO Annuity & Maintenance Windows

Assigned Categories & Tags

Assign Category Tag