Patent US4463435A successfully refreshed from Google Patents.
← Back to Patent Explorer
EP0334472A2 Expired / Lapsed EP ⏱️ 0.0 years remaining

Methods of detecting character strings

Assignee: International Business Machines Corp
•
Filing: 1989-02-06
•
Granted: —
•
Expires: 2009-02-06
📄 Original PDF 🌐 Google Patents 🔄 Sync Now

Abstract

Abstract A method for detecting potential character string regions for character strings extending in a predetermined direction in an image obtained by reading a document includes storing image data of the document in an image memory with in such a way that the predetermined direction in the image is aligned with a main scanning direction of the image data in the memory, accessing the image memory to scan the image whilst generating position data for the black runs in each scan line, comparing the position data relating to the black runs obtained for each pair of two adjoining scan lines, and, if a part of a black run which does not adjoin a black run in a preceding line exists in a succeeding line, deeming the part to be a boundary candidate of the first kind, generating position data thereof, and storing the data in a stored table, and if a part of a black run which does not adjoin a black run in a succeeding line exists in a preceding line, deeming the part to be a boundary candidate of the second kind, generating position data thereof, and storing the data in the stored table, looking up the table to determine a pair consisting of a boundary candidate of the first kind and of the second kind, which have a part whose position in the main scanning direction is common, deeming a rectangular region whose boundaries include at least a part of each boundary candidate of the determined pair to be a character string region, and generating position data thereof.

Inventors

  • 👤 Tomio Amano

Bibliographic Details

Application Number —
Jurisdiction EP
Data Source google_patents
Last Synchronized 2026-10-25 17:54
CPC Classifications
G; G06; G06V; G06V30/00; G06V30/10; G06V30/14; G06V30/148; G06V30/155

Claims Analysis

Total Claims: 9
Claim 1
Independent Claim 268 words
A method for detecting potential character string regions for character strings extending in a predetermined direction in an image obtained by reading a document comprising steps of: (a) storing image data of the document in an image memory with in such a way that the predetermined direction in the image is aligned with a main scanning direction of the image data in the memory, (b) accessing the image memory to scan the image whilst generating position data for the black runs in each scan line, (c) comparing the position data relating to the black runs obtained in step (b) for each pair of two adjoining scan lines, and, if a part of a black run which does not adjoin a black run in a preceding line exists in a succeeding line, deeming the part to be a boundary candidate of the first kind, generating position data thereof, and storing the data in a stored table, if a part of a black run which does not adjoin a black run in a succeeding line exists in a preceding line, deeming the part to be a boundary candidate of the second kind, generating position data thereof, and storing the data in the stored table, (d) looking up the table to determine a pair consisting of a boundary candidate of the first kind and of the second kind, which have a part whose position in the main scanning direction is common, deeming a rectangular region whose boundaries include at least a part of each boundary candidate of the determined pair to be a character string region, and generating position data thereof.
Claim 2
Independent Claim 268 words
A method for detecting potential character string regions for character strings extending in a predetermined direction in an image obtained by reading a document comprising steps of: (a) storing image data of the document in an image memory with in such a way that the predetermined direction in the image is aligned with a main scanning direction of the image data in the memory, (b) accessing the image memory to scan the image whilst generating position data for the black runs in each scan line, (c) comparing the position data relating to the black runs obtained in step (b) for each pair of two adjoining scan lines, and, if a part of a black run which does not adjoin a black run in a preceding line exists in a succeeding line, deeming the part to be a boundary candidate of the first kind, generating position data thereof, and storing the data in a stored table, if a part of a black run which does not adjoin a black run in a succeeding line exists in a preceding line, deeming the part to be a boundary candidate of the second kind, generating position data thereof, and storing the data in the stored table, (d) looking up the table to determine a pair consisting of a boundary candidate of the first kind and of the second kind, which have a part whose position in the main scanning direction is common, deeming a rectangular region whose boundaries include at least a part of each boundary candidate of the determined pair to be a character string region, and generating position data thereof.
Claim 3
Dependent Claim 31 words
2 A method as claimed in Claim 1 wherein the position data of each black run generated in step (b) includes the starting point position and length of the black run.
Claim 4
Independent Claim 67 words
A method as claimed in either preceding Claim wherein, in step (b), if the length of a white run sandwiched between black runs is not greater than a predetermined value, the white run is recognised to be a part of a grouped black run formed together with the two adjoining black runs, and position data of the grouped black run is generated and stored in the table.
Claim 5
Independent Claim 73 words
A method as claimed in any preceding Claim wherein, after step (c), if the space between the boundary candidates of the same kind detected on the same line has a length not greater than a predetermined value, the space is recognised to be a part of a grouped boundary candidate formed together with the two adjoining boundary candidates, the position data of the grouped boundary candidate replacing its component parts in the table.
Claim 6
Independent Claim 86 words
A method as claimed in any preceding Claim wherein, after step (c), if there is a space having a length not greater than a predetermined value with respect to a sub-scanning direction between two boundary candidates of the same kind which are contiguous with respect to the main scanning direction, one boundary candidate is recognised to be a part of a grouped boundary candidate formed together with the other boundary candidate, the position data of the grouped boundary candidate replacing its component parts in the table.
Claim 7
Independent Claim 89 words
A method as claimed in any preceding Claim wherein, in step (d), if the distance between the boundary candidates of the first kind and second kind of the determined pair is within a predetermined range and the gap between the two boundary candidates with respect to the main scanning direction is not greater than a predetermined value, a rectangular region whose boundaries include at least a part of each boundary candidate of the determined pair is judged to be a character string region, and position data thereof is generated.
Claim 8
Independent Claim 28 words
A method as claimed in any preceding Claim wherein, the position data of each boundary generated in step (c) includes the starting position and length of the boundary.
Claim 9
Independent Claim 39 words
A method as claimed in any preceding Claim wherein, the position data of a character string region generated in step (d) includes one apex and two orthogonal sides of the rectangular region deemed to be a character string region.

References Cited (Backward Citations)

No citation records indexed.

USPTO Annuity & Maintenance Windows

Maintenance schedule is calculated for granted US utility patents.

Assigned Categories & Tags

No categories assigned to this patent yet.

Assign Category Tag