Character region extracting method and apparatus capable of implementing the method

Claim 2

A character region extracting method for character recognition comprising the steps of; segmentating a first region (BL1) from a document image by extracting projection of the document image in a first direction parallel to a character string and detecting a range in which the projection is continuous in a second direction perpendicular to the first direction, the range defining the first region; segmentating a second region (sBL1) from the first region by extracting projection of the first region in the second direction and detecting a range in which the projection is continuous in the first direction, the range defining the second region; and segmentating a third region (BL2) from the second region by extracting projection of the second region in the first direction and detecting a range in which the projection is continuous in the second direction, the range defining the third region.

Claim 3

A character region extracting method as claimed in claim 1, further comprising the step of joining third regions in the first direction by comparing a relationship of a relative location between the third regions and a relationship of a relative size between the third regions, so that a fourth region (BL3) containing characters or a character string is extracted.

Claim 4

A character region extracting method as claimed in claim 2, further comprising the step of joining fourth regions in the second direction by comparing a relationship of a relative location between the fourth regions and a relationship of a relative size between the fourth regions, so that a fifth region (BL4) containing characters or a character string is extracted.

Claim 5

A region character extracting method as claimed in claim 3, further comprising the step of rearranging the fifth regions by comparing a relationship of a relative location between the fifth regions so that characters and character strings included in one character line are continuously arranged.

Claim 6

A character region extracting method as claimed in claim 4, further comprising the step of joining the rearranged fifth regions in the first direction by comparing a relationship of a relative location between the rearranged fifth regions and a relationship of a relative size between the rearranged fifth regions, so that a sixth region containing the characters or character strings in said one character line is extracted as one character region.

Claim 7

A character region extracting method as claimed in claim 5, wherein the fifth regions each related to the same category of information indicated by characters included in the fifth regions are joined together.

Claim 8

A character region extracting method as claimed in claim 1, wherein the projection extracted in each of the steps is represented by presence and non-presence of a black pixel in the document image in the first or second direction.

Claim 9

A character region extracting method as claimed in claim 1, wherein the projection extracted in each of the steps is represented by the number of black pixels in the first or second direction.

Claim 10

A character region extracting apparatus adapted to character recognition comprising; optical scanning means for extracting a document image in the form of an electric signal by optically scanning a document; an image memory for storing the document image extracted by the optical scanning means; projection extracting means for extracting projection of the document image in a first and second direction, the first direction corresponding to a scanning direction in the optical scanning means and the second direction being perpendicular to the first direction; a projection buffer for storing the projection in each of the first and second direction extracted by the projection extracting means; region segmentating means for segmentating a region by using the projection stored in the projection buffer, the region segmentating means further comprising a first means for segmentating a first region (BL1) from the document image by using the extracted projection of the document image in the first direction and detecting a range in which the projection is continuous in the second direction, said range defining the first region; a second means for segmentating a second region (sBL1) from the first region by using the extracted projection of the first region in the second direction and detecting a range in which the projection is continuous in the first direction, the range defining the second region; and a third means for segmentating a third region (BL2) from the second region by using the extracted projection of the second region in the first direction and detecting a range in which the projection is continuous in the second direction, the range defining the third region; and memory means for storing the extracted third region.

Claim 11

A character region extracting apparatus adapted to character recognition comprising; optical scanning means for extracting a document image in the form of an electric signal by optically scanning a document; an image memory for storing the document image extracted by the optical scanning means; projection extracting means for extracting projection of the document image in a first and second direction, the first direction corresponding to a scanning direction in the optical scanning means and the second direction being perpendicular to the first direction; a projection buffer for storing the projection in each of the first and second direction extracted by the projection extracting means; region segmentating means for segmentating a region by using the projection stored in the projection buffer, the region segmentating means further comprising a first means for segmentating a first region (BL1) from the document image by using the extracted projection of the document image in the first direction and detecting a range in which the projection is continuous in the second direction, said range defining the first region; a second means for segmentating a second region (sBL1) from the first region by using the extracted projection of the first region in the second direction and detecting a range in which the projection is continuous in the first direction, the range defining the second region; and a third means for segmentating a third region (BL2) from the second region by using the extracted projection of the second region in the first direction and detecting a range in which the projection is continuous in the second direction, the range defining the third region; and memory means for storing the extracted third region.

Claim 12

A character region extracting apparatus as claimed in claim 9, wherein the region segmentating means is implemented by a central processing unit.

Claim 13

A character region extracting apparatus as claimed in claim 10, wherein the region segmentating means further comprises a fifth means for joining fourth regions in the second direction by comparing a relationship of a relative location between the fourth regions and a relationship of a relative size between the fourth regions, so that a fifth region (BL4) containing characters or a character string is extracted.

Claim 14

A character region extracting apparatus as claimed in claim 11, wherein the region segmentating means further comprises means for rearranging the fifth regions by comparing a relationship of a relative location between the fifth regions so that characters and character strings included in one character line are continuously arranged.

Claim 15

A character region extracting apparatus as claimed in claim 12, wherein the region segmentating means further comprises for joining the rearranged fifth regions in the first direction by comparing a relationship of a relative location between the rearranged fifth regions and a relationship of a relative size between the rearranged fifth regions, so that a sixth region containing the characters or character strings in said one character line is extracted as one character region.

Claim 16

A character region extracting apparatus as claimed in claim 13, wherein the fifth regions each related to the same category of information indicated by characters included in the fifth regions are joined together.

Claim 17

A character region extracting apparatus as claimed in claim 9, wherein the region segmentating means further comprises a fourth means for joining third regions in the first direction by comparing a relationship of a relative location between the third regions and a relationship of a relative size between the third regions, so that a fourth region (BL3) containing characters or a character string is extracted.

Claim 18

A character region extracting apparatus as claimed in claim 9, wherein the projection extracted by the projection extracting means is represented by presence and non-presence of a black pixel in the document image in the first or second direction.

Claim 19

A character region extracting apparatus as claimed in claim 9, wherein the projection extracted by the projection extracting means is represented by the number of black pixels in the first or second direction.