US4468808A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Feature extraction system for digitized character information
Assignee: National Institute of Advanced Industrial Science and Technology
•
Filing: 1981-01-19
•
Granted: 1984-08-28
•
Expires: 2001-01-19
Abstract
Abstract Disclosed is an extraction method for optical character recognition systems in which the necessary character features for recognition process are extracted from a character information detected and digitized by an optical scanner. In accordance with this method, the local features of every adjacent two rows or columns of a digitized character pattern on a two-dimensional plane are extracted row by row or column by column and are successively integrated so as to extract in a global manner the respective feature types such as concavity and convexity, loop and connectivity of the character pattern. At the same time, the global feature regions or segments are separated and each of the separated feature segments is coded without any loss of the information. The feature quantity of each segment is obtained on the basis of the coded representation. Any local concavity or convexity noises are eliminated simultaneously with the extraction of the global features.
Inventors
- 👤 Shunji Mori
- 👤 Hiromitsu Yamada
- 👤 Kunihiko Takeuchi
- 👤 Michihisa Do
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-10-25 20:28
CPC Classifications
G; G06; G06V; G06V30/00; G06V30/10; G06V30/18; G06V30/182
Claims Analysis
Total Claims: 6
Claim 1
Independent Claim
228 words
A method of extracting, for the purpose of recognition processing, character features from a character pattern detected by an optical scanner, digitized into character digital bits and background digital bits and stored in a two-dimensional memory, said method comprising the steps of: (a) scanning said character pattern on the two-dimensional plane of said memory sequentially row by row to detect unit row pattern blocks each including a single row bit array; (b) separating from each of said unit row pattern blocks predetermined parameters including a run of a series of character digital bits and respective run length data and coordinates data for each run of each said block; (c) for each of said scanning steps, comparing the respective detected unit row pattern block with the unit row pattern block detected by the just preceding scanning such that in accordance with variations in the detected parameters in the compared unit pattern blocks, types of local features are detected in the form of a concavity opened upward, downward, left and right, and extracted and listed in correspondence with the compared unit pattern blocks; and (d) successively integrating said local features extracted by each said scanning step, coding the results obtained in the course of the step of successively integrating and the predetermined parameters including run length data and then listing feature quantities and coordinates of respective segments as global features.
Claim 2
Independent Claim
228 words
A method of extracting, for the purpose of recognition processing, character features from a character pattern detected by an optical scanner, digitized into character digital bits and background digital bits and stored in a two-dimensional memory, said method comprising the steps of: (a) scanning said character pattern on the two-dimensional plane of said memory sequentially row by row to detect unit row pattern blocks each including a single row bit array; (b) separating from each of said unit row pattern blocks predetermined parameters including a run of a series of character digital bits and respective run length data and coordinates data for each run of each said block; (c) for each of said scanning steps, comparing the respective detected unit row pattern block with the unit row pattern block detected by the just preceding scanning such that in accordance with variations in the detected parameters in the compared unit pattern blocks, types of local features are detected in the form of a concavity opened upward, downward, left and right, and extracted and listed in correspondence with the compared unit pattern blocks; and (d) successively integrating said local features extracted by each said scanning step, coding the results obtained in the course of the step of successively integrating and the predetermined parameters including run length data and then listing feature quantities and coordinates of respective segments as global features.
Claim 3
Dependent Claim
90 words
A method according to claim 1 further comprising a local feature list correcting step, wherein of the local feature lists produced by said step (c), the local feature lists by said just preceding scanning and the local feature lists by said respective scanning are used to perform a noise processing on the unit pattern block produced by said just preceding scanning so as to neglect concavity features of 1-bit depth in a direction opposite to the direction of the scanning and thereby to effect correction of the local feature lists.
Claim 4
Independent Claim
228 words
A method of extracting, for the purpose of recognition processing, character features from a character pattern detected by an optical scanner, digitized into character digital bits and background digital bits and stored in a two-dimensional memory, said method comprising the steps of: (a) scanning said character pattern on the two-dimensional plane of said memory sequentially column by column to detect unit column pattern blocks each including a single column bit array; (b) separating from each of said unit column pattern blocks predetermined parameters including a run of a series of character digital bits and respective run length data and coordinates data for each run of each said block; (c) for each of said scanning steps, comparing the respective detected unit column pattern block with the unit column pattern block detected by the just preceding scanning such that in accordance with variations in the detected parameters in the compared unit pattern blocks, types of local features are detected in the form of a concavity opened upward, downward, left and right, and extracted and listed in correspondence with the compared unit pattern blocks; and (d) successively integrating said local features extracted by each said scanning step, coding the results obtained in the course of the step of successively integrating and the predetermined parameters including run length data and then listing feature quantities and coordinates of respective segments as global features.
Claim 5
Independent Claim
228 words
A method of extracting, for the purpose of recognition processing, character features from a character pattern detected by an optical scanner, digitized into character digital bits and background digital bits and stored in a two-dimensional memory, said method comprising the steps of: (a) scanning said character pattern on the two-dimensional plane of said memory sequentially column by column to detect unit column pattern blocks each including a single column bit array; (b) separating from each of said unit column pattern blocks predetermined parameters including a run of a series of character digital bits and respective run length data and coordinates data for each run of each said block; (c) for each of said scanning steps, comparing the respective detected unit column pattern block with the unit column pattern block detected by the just preceding scanning such that in accordance with variations in the detected parameters in the compared unit pattern blocks, types of local features are detected in the form of a concavity opened upward, downward, left and right, and extracted and listed in correspondence with the compared unit pattern blocks; and (d) successively integrating said local features extracted by each said scanning step, coding the results obtained in the course of the step of successively integrating and the predetermined parameters including run length data and then listing feature quantities and coordinates of respective segments as global features.
Claim 6
Dependent Claim
90 words
A method according to claim 3 further comprising a local feature list correcting step, wherein of the local feature lists produced by said step (c), the local feature lists by said just preceding scanning and the local feature lists by said respective scanning are used to perform a noise processing on the unit pattern block produced by said just preceding scanning so as to neglect concavity features of 1-bit depth in a direction opposite to the direction of the scanning and thereby to effect correction of the local feature lists.
References Cited (Backward Citations)
US3178688A
Cited
US3541511A
Cited
US3713098A
Cited
US4061998A
Cited
US4048615A
Cited
US4155072A
Cited
US4193056A
Cited
US4189711A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 1987-08-28 | 1988-02-27 | 1988-08-28 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 1991-08-28 | 1992-02-27 | 1992-08-28 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 1995-08-28 | 1996-02-27 | 1996-08-28 | $7700.00 | Window Closed / Lapsed |