Patent US4379283A successfully refreshed from Google Patents.
← Back to Patent Explorer
US4379283A Expired / Lapsed US ⏱️ 0.0 years remaining

Type font optical character recognition system

Assignee: Toyo Keiki Co Ltd
•
Filing: 1981-01-30
•
Granted: 1983-04-05
•
Expires: 2001-01-30
📄 Original PDF 🌐 Google Patents 🔄 Sync Now

Abstract

Abstract Stored in a two-dimensional memory is a read-in character information of one character which was scanned by an optical scanner and digitized into character digital bits and background digital bits. The character information on the storage plane of the memory is divided into a plurality of segments using 4×4 bit meshes as a unit and the number of the character digital bit meshes included in each of the segments is counted. A collection of segment information obtained from these count values is used as an input information and is compared by pattern matching with each of a collection of standard character information preliminarily prepared for a set of characters in the same information form as the input information, thereby effecting the desired type font character recognition.

Inventors

  • 👤 Koji Ito
  • 👤 Akira Kondo

Bibliographic Details

Application Number —
Jurisdiction US
Data Source google_patents
Last Synchronized 2026-10-27 10:32
CPC Classifications
G; G06; G06V; G06V10/00; G06V10/70; G06V10/74; G06V10/75; G06V10/751

Claims Analysis

Total Claims: 11
Claim 1
Independent Claim 274 words
An optical character recognition system comprising: first memory means for temporarily storing an input character information read by optical scanning of a printed character on a paper in the form of a digitized data including background digital bits indicative of a paper white background area and character digital bits indicative of a character black data area; location normalizing means for normalizing a location occupied by said input character information on a storage plane of said first memory means; counting means for dividing said normalized input character information on said storage plane into a plurality of segments each having an area of 4×4 meshes, counting the number of character digital bit indicative meshes in each of said segments and generating the resulting count values as a set of segment information; subtracting means for subtracting 1 from each of said count values of said segment information set when the same is greater than 1; second memory means for temporarily storing as an input information a collection of said segment information for one character applied via said subtracting means; third memory means preliminarily storing permanently a set of standard character information representing a plurality of different characters in the same segment information form as the information form of said input information; matching means for performing the process of pattern matching between said input information and each of said set of standard character information; and decision means responsive to the result of said pattern matching process to determine whether said printed character is to be recognized as one corresponding to the most matched one of said set of standard character information or the same is to be rejected.
Claim 2
Dependent Claim 274 words
An optical character recognition system comprising: first memory means for temporarily storing an input character information read by optical scanning of a printed character on a paper in the form of a digitized data including background digital bits indicative of a paper white background area and character digital bits indicative of a character black data area; location normalizing means for normalizing a location occupied by said input character information on a storage plane of said first memory means; counting means for dividing said normalized input character information on said storage plane into a plurality of segments each having an area of 4×4 meshes, counting the number of character digital bit indicative meshes in each of said segments and generating the resulting count values as a set of segment information; subtracting means for subtracting 1 from each of said count values of said segment information set when the same is greater than 1; second memory means for temporarily storing as an input information a collection of said segment information for one character applied via said subtracting means; third memory means preliminarily storing permanently a set of standard character information representing a plurality of different characters in the same segment information form as the information form of said input information; matching means for performing the process of pattern matching between said input information and each of said set of standard character information; and decision means responsive to the result of said pattern matching process to determine whether said printed character is to be recognized as one corresponding to the most matched one of said set of standard character information or the same is to be rejected.
Claim 3
Dependent Claim 76 words
A system according to claim 1, wherein each of said set of standard character information for one character comprises a collection of standard segment information each thereof including an upper limit value and a lower limit value, and wherein said matching means calculates a correlation distance between said input information and each of said set of standard character information as a total sum of distances calculated by comparison between corresponding ones of said segment information thereof.
Claim 4
Dependent Claim 41 words
A system according to claim 2, wherein the upper limit of said distances calculated by comparison between corresponding ones of said segment information of said input information and each said standard character information is limited to less than a predetermined value.
Claim 5
Dependent Claim 146 words
A system according to claim 1, wherein each of said set of standard character information for one character comprises a collection of standard segment information each thereof including an upper limit value and a lower limit value, and wherein said matching means comprises an arithmetic unit for comparing said input information and each said standard character information such that the difference between said upper limit value or lower limit value of each of said standard segment information and each of said segment information is obtained and maximum one of said difference values and zero is generated as a parameter, a function generator responsive to said parameter to generate a predetermined characteristic function, and an adder for producing a sum of said characteristic function values of said segment information for one character to generate a correlation distance between said input information and each said standard character information.
Claim 6
Dependent Claim 22 words
A system according to claim 4, wherein said function generator generates f(X)=X 2 as said characteristic function corresponding to a parameter X.
Claim 7
Dependent Claim 61 words
A system according to claim 4, wherein said function generator comprises a memory device for receiving a parameter X as an address information, said memory device storing as said characteristic function f(X)=X 2 for each of said parameters X of values less than a predetermined value and a fixed value for each of said parameters X greater than said predetermined value.
Claim 8
Dependent Claim 69 words
A system according to claim 4, wherein each of said plurality of segment information of said input information is represented by a 4-bit binary number, and each of said plurality of standard segment information of each said standard character information is represented by an 8-bit binary number of which four bits are assigned to said upper limit value and another four bits are assigned to said lower limit value.
Claim 9
Dependent Claim 46 words
A system according to claim 1, wherein said location normalizing means detects a left end side and an upper end side of said character digital bit area of said input character information on said storage plane and alignes each of said sides with a reference position.
Claim 10
Dependent Claim 39 words
A system according to claim 8, wherein said detection is accomplished by using a window of 2×2 mesh area such that said detection is accomplished when all the meshes of said window are occupied by said character digital bits.
Claim 11
Dependent Claim 35 words
A system according to claim 1, wherein the storage plane of said first memory means comprises an area of 32×20 bits per character, and wherein the number of said segment information per character is 40.

References Cited (Backward Citations)

US3522586A Cited
US3634823A Cited
US3634822A Cited
US3805239A Cited
US3831146A Cited
US4288779A Cited
US4274079A Cited

USPTO Annuity & Maintenance Windows

Annuity Stage Window Opens Payment Deadline Surcharge Period Until Estimated Fee Status
3.5 Years (Year 4) 1986-04-04 1986-10-04 1987-04-05 $2000.00 Window Closed / Lapsed
7.5 Years (Year 8) 1990-04-04 1990-10-04 1991-04-05 $3760.00 Window Closed / Lapsed
11.5 Years (Year 12) 1994-04-04 1994-10-04 1995-04-05 $7700.00 Window Closed / Lapsed

Assigned Categories & Tags

No categories assigned to this patent yet.

Assign Category Tag