Patent US5113452A successfully refreshed from Google Patents.
US5113452A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Hand-written character recognition apparatus and method
Assignee: Sony Corp
•
Filing: 1989-08-18
•
Granted: 1992-05-12
•
Expires: 2009-08-18
Abstract
Abstract A hand-written character recognition apparatus and method in which line element data obtained from an output hand-written stroke is converted to a plurality of sub-strokes satisfying a tracing condition in accordance with a tracing command described in a character dictionary data base, substrokes satisfying the tracing condition are selected in accordance with a selection command described in the character dictionary data base, then on the basis of an assumption that the hand-written stroke is a set of sub-strokes, the hand-written character is recognized by evaluating the shape of the sub-stroke by employing a shape evaluation command and by evaluating a position evaluation command derived from the character dictionary data base, whereby a hand-written character, such as a distorted hand-written character and so on, can be recognized with ease.
Inventors
- 👤 Masayuki Chatani
- 👤 Kimiyoshi Yoshida
- 👤 Akio Sakano
- 👤 Hirofumi Tamori
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-03 20:41
CPC Classifications
G; G06; G06V; G06V30/00; G06V30/10; G06V30/14; G06V30/142; G06V30/1423; G06V30/32; G06V30/333; G06V30/347; G06F; G06F18/00; G06V30/36; G06V30/373
Claims Analysis
Total Claims: 10
Claim 1
Independent Claim
267 words
A hand-written character recognition apparatus comprising: means for receiving dot data obtained when writing a character and for generating corresponding line element data therefrom; character dictionary data base means which stores character data as command instruction sets each expressing the configuration of a different character, the character data including tracing commands for extracting sub-stroke data from the line element data, sub-stroke data selecting commands for selecting one or more sub-stroke data for later evaluation processing, and checking commands for partially evaluating the thus selected one or more sub-stroke data by making a comparison to a partial shape of a standard character, and for totally evaluating the hand-written character by comparing it to the standard character; sub-stroke data generating means connected to the character dictionary data base means for receiving the tracing commands and generating sub-stroke data by tracing the line element data based on the tracing commands inherent to each standard character derived from the character dictionary data base means; sub-stroke data selecting means connected to the character dictionary data base means for receiving the character selection commands and supplied with the thus generated sub-stroke data for selecting sub-stroke data inherent to each standard character based on the sub-stroke data selecting commands; and sub-stroke data evaluating means connected to the sub-stroke data selecting means and the character dictionary data means, so as to receive character checking commands, for evaluating partial similarity of the selected sub-stroke data compared to a partial sub-stroke data of the standard character, and for evaluating the total similarity of the hand-written character based on the results of the evaluation of the partial similarity.
Claim 2
Dependent Claim
267 words
A hand-written character recognition apparatus comprising: means for receiving dot data obtained when writing a character and for generating corresponding line element data therefrom; character dictionary data base means which stores character data as command instruction sets each expressing the configuration of a different character, the character data including tracing commands for extracting sub-stroke data from the line element data, sub-stroke data selecting commands for selecting one or more sub-stroke data for later evaluation processing, and checking commands for partially evaluating the thus selected one or more sub-stroke data by making a comparison to a partial shape of a standard character, and for totally evaluating the hand-written character by comparing it to the standard character; sub-stroke data generating means connected to the character dictionary data base means for receiving the tracing commands and generating sub-stroke data by tracing the line element data based on the tracing commands inherent to each standard character derived from the character dictionary data base means; sub-stroke data selecting means connected to the character dictionary data base means for receiving the character selection commands and supplied with the thus generated sub-stroke data for selecting sub-stroke data inherent to each standard character based on the sub-stroke data selecting commands; and sub-stroke data evaluating means connected to the sub-stroke data selecting means and the character dictionary data means, so as to receive character checking commands, for evaluating partial similarity of the selected sub-stroke data compared to a partial sub-stroke data of the standard character, and for evaluating the total similarity of the hand-written character based on the results of the evaluation of the partial similarity.
Claim 3
Dependent Claim
50 words
A hand-written character recognition apparatus as recited in claim 1, wherein the sub-stroke data selecting commands include a selecting condition command, storing commands, and output selecting commands, and further comprising a buffer memory connected to the sub-stroke data selecting means for storing selected sub-stroke data satisfying the selecting condition commands.
Claim 4
Dependent Claim
75 words
A hand-written character recognition apparatus as recited in claim 2, wherein the character checking commands include partial checking commands and total checking commands, and wherein the sub-stroke data evaluating means receives the partial checking commands and generates a partial evaluation and receives the total checking command and, using the results of the partial evaluation, generates a final evaluation of the written character relative to the character data stored in the character dictionary data base means.
Claim 5
Dependent Claim
89 words
A hand-written character recognition apparatus as recited in claim 3, wherein the character checking commands include shape evaluation commands and positional evaluation commands and wherein the sub-stroke data selecting means receives the shape evaluation commands and evaluates the shape of the selected sub-stroke data of the written character compared to the shape of the standard character and receives the positional evaluation commands and evaluates the positional relation of the selected sub-stroke data of the written character compared to the positional relations of the sub-stroke data of the standard character.
Claim 6
Dependent Claim
206 words
A hand-written character recognition method comprising the steps of: receiving dot data obtained when writing a character and generating corresponding line element data therefrom; storing character data in a character dictionary data base memory as command sets each expressing the configuration of a different standard character, the character data including tracing commands for extracting sub-stroke data from the line element data, sub-stroke data selecting commands for selecting one or more su-stroke data for later evaluation processing, and checking commands for partially evaluating the thus selected one or more sub-stroke data by making a comparison to a partial shape of a standard character, and for totally evaluating the hand-written character by comparing it to the standard character; generating sub-stroke data by tracing the line element data based on the tracing commands inherent to each standard character derived from the character dictionary data base memory; selecting sub-stroke data inherent to each standard character based on the sub-stroke data selecting commands; and evaluating partial similarity of the selected sub-stroke data compared to a partial sub-stroke data of the standard character, and for evaluating the total similarity of the hand-written character compared to the shape of the standard character based on the results of the evaluation of the partial similarity.
Claim 7
Dependent Claim
206 words
A hand-written character recognition method comprising the steps of: receiving dot data obtained when writing a character and generating corresponding line element data therefrom; storing character data in a character dictionary data base memory as command sets each expressing the configuration of a different standard character, the character data including tracing commands for extracting sub-stroke data from the line element data, sub-stroke data selecting commands for selecting one or more su-stroke data for later evaluation processing, and checking commands for partially evaluating the thus selected one or more sub-stroke data by making a comparison to a partial shape of a standard character, and for totally evaluating the hand-written character by comparing it to the standard character; generating sub-stroke data by tracing the line element data based on the tracing commands inherent to each standard character derived from the character dictionary data base memory; selecting sub-stroke data inherent to each standard character based on the sub-stroke data selecting commands; and evaluating partial similarity of the selected sub-stroke data compared to a partial sub-stroke data of the standard character, and for evaluating the total similarity of the hand-written character compared to the shape of the standard character based on the results of the evaluation of the partial similarity.
Claim 8
Dependent Claim
46 words
A hand-written character recognition method as recited in claim 5, wherein the sub-stroke data selecting commands include a selecting condition command, storing commands, and output selecting commands, and further comprising the step of storing in a buffer memory selected sub-stroke data satisfying the selecting condition commands.
Claim 9
Dependent Claim
76 words
A hand-written character recognition method as recited in claim 6, wherein the character checking commands include shape evaluation commands and positional evaluation commands, and further comprising the steps of evaluating the shape of the selected sub-stroke data of the written character compared to the shape of the standard character and evaluating the positional relation of the selected sub-stroke data of the written character compared to the positional relations of the sub-stroke data of the standard character.
Claim 10
Dependent Claim
84 words
A hand-written character recognition method as recited in claim 7, wherein the character checking commands derived from the character dictionary data base memory include a partial checking command and a total checking command, and further comprising the step of using the partial checking command to partially evaluate the written character relative to the stored character data and using the total checking command to generate a final evaluation of the written character relative to the stored characters based on results of the partial evaluation step.
References Cited (Backward Citations)
GB1144319A
Cited
US4045773A
Cited
US4193056A
Cited
US4162482A
Cited
US4284975A
Cited
US4317109A
Cited
WO1980002761A1
Cited
US4527283A
Cited
US4377803A
Cited
US4654873A
Cited
US4607385A
Cited
US4653107A
Cited
EP0166271A2
Cited
US4850025A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 1995-05-12 | 1995-11-11 | 1996-05-12 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 1999-05-12 | 1999-11-11 | 2000-05-12 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2003-05-12 | 2003-11-11 | 2004-05-12 | $7700.00 | Window Closed / Lapsed |