Patent US5647022A successfully refreshed from Google Patents.
US5647022A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Method and apparatus for symbol recognition using multidimensional preprocessing and symbol sorting
Assignee: TELECOMMUNICATION LABORATORIES DGT MOTC REPUBLIC OF CHINA
•
Filing: 1992-10-13
•
Granted: 1997-07-08
•
Expires: 2012-10-13
Abstract
Abstract Data samples describing a plurality of micro-segments that compose a symbol to be recognized are received from a device such as an electronic pad. A preprocessor maps the micro-segments into cells of an array that has several feature dimensions. The preprocessor assigns values to the cells based on the length of a micro-segment associated with the cell, and how well the features of the associated micro-segment correspond to the feature label of the cell. The cell values are used as inputs to at least one of a plurality of neural networks, where each neural network is trained to identify symbols from a different group of symbols. A sorter examines the symbol to determine which neural network should be used to recognize the symbol. An output produced by the sorter controls a switching means that communicates the cell values to the proper neural network.
Inventors
- 👤 Yann Andre LeCun
- 👤 Quen-Zong Wu
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-08 08:56
CPC Classifications
G; G06; G06V; G06V30/00; G06V30/10; G06V30/32; G06V30/333; G06V30/347; G06V30/18; G06V30/1801; G06V30/18019; G06V30/18038; G06V30/18048; G06V30/18057
Claims Analysis
Total Claims: 24
Claim 1
Independent Claim
143 words
An apparatus for recognizing symbols, comprising: means for collecting data that describes a symbol comprising a plurality of micro-segments; means for sorting said symbol based on an evaluation of said data to produce a switch control output; means for associating a micro-segment with at least one cell in a plurality of cells that quantize micro-segment features, where each of said cells has an associated feature label; means for assigning a cell value to a cell, using a micro-segment value that is related to how well a feature of said micro-segment associated with said cell corresponds to said cell's feature label; a plurality of classifiers that receive said cell values assigned to said plurality of cells for recognizing said symbol; and switching means for communicating said cell values to at least one of said plurality of classifiers in response to said switch control output.
Claim 2
Independent Claim
143 words
An apparatus for recognizing symbols, comprising: means for collecting data that describes a symbol comprising a plurality of micro-segments; means for sorting said symbol based on an evaluation of said data to produce a switch control output; means for associating a micro-segment with at least one cell in a plurality of cells that quantize micro-segment features, where each of said cells has an associated feature label; means for assigning a cell value to a cell, using a micro-segment value that is related to how well a feature of said micro-segment associated with said cell corresponds to said cell's feature label; a plurality of classifiers that receive said cell values assigned to said plurality of cells for recognizing said symbol; and switching means for communicating said cell values to at least one of said plurality of classifiers in response to said switch control output.
Claim 3
Dependent Claim
14 words
The apparatus of claim 1, wherein said plurality of classifiers comprises a neural network.
Claim 4
Dependent Claim
15 words
The apparatus of claim 1, wherein said plurality of classifiers comprise a convolution neural network.
Claim 5
Dependent Claim
35 words
The apparatus of claim 1 wherein said means for assigning assigns said cell value to said cell using said micro-segment value that is also related to a length of said micro-segment associated with said cell.
Claim 6
Dependent Claim
29 words
The apparatus of claim 1 wherein said means for assigning assigns said cell value using a sum of micro-segment values from a plurality of micro-segments associated with said cell.
Claim 7
Dependent Claim
22 words
The apparatus of claim 1, wherein said means for assigning assigns said cell value based on a micro-segment associated with another cell.
Claim 8
Independent Claim
135 words
A method for recognizing symbols, comprising the steps of: collecting data that describes a symbol comprising a plurality of micro-segments; sorting said symbol based on an evaluation of said data to produce a switch control signal; associating a micro-segment with at least one cell in a plurality of cells that quantize micro-segment features, where each of said cells has an associated feature label; assigning a cell value to a cell, using a micro-segment value that is related to how well a feature of said micro-segment associated with said cell corresponds to said cell's feature label; using a plurality of classifiers that receive said cell values assigned to said plurality of cells for recognizing said symbol; and communicating said cell values to at least one of said plurality classifiers in response to said switch control signal.
Claim 9
Independent Claim
135 words
A method for recognizing symbols, comprising the steps of: collecting data that describes a symbol comprising a plurality of micro-segments; sorting said symbol based on an evaluation of said data to produce a switch control signal; associating a micro-segment with at least one cell in a plurality of cells that quantize micro-segment features, where each of said cells has an associated feature label; assigning a cell value to a cell, using a micro-segment value that is related to how well a feature of said micro-segment associated with said cell corresponds to said cell's feature label; using a plurality of classifiers that receive said cell values assigned to said plurality of cells for recognizing said symbol; and communicating said cell values to at least one of said plurality classifiers in response to said switch control signal.
Claim 10
Dependent Claim
19 words
The method of claim 7, wherein said step of using said plurality of classifiers comprises using a neural network.
Claim 11
Dependent Claim
20 words
The method of claim 7, wherein said step of using said plurality of classifiers comprises using a convolution neural network.
Claim 12
Dependent Claim
35 words
The method of claim 7 wherein said step of assigning assigns said cell value to said cell using said micro-segment value that is also related to a length of said micro-segment associated with said cell.
Claim 13
Dependent Claim
29 words
The method of claim 7 wherein said step of assigning assigns said cell value using a sum of micro-segment values from a plurality of micro-segments associated with said cell.
Claim 14
Dependent Claim
22 words
The method of claim 7, wherein said step of assigning assigns said cell value based on a micro-segment associated with another cell.
Claim 15
Independent Claim
156 words
A method for recognizing a group of symbols in a series of symbols, comprising the steps of: collecting data that describes said group of symbols, said group of symbols having symbols comprising a plurality of micro-segments; sorting said group of symbols based on an evaluation of said data to produce a switch control output; associating a micro-segment with at least one cell in a plurality of cells that quantize micro-segment features, where each of said cells has an associated feature label; assigning a cell value to a cell, using a micro-segment value that is related to how well a feature of said micro-segment associated with said cell corresponds to said cell's feature label; using a plurality of classifiers that receive said cell values assigned to said plurality of cells for recognizing said group of symbols; and communicating said cell values to at least one of said plurality of classifiers in response to said switch control output.
Claim 16
Independent Claim
156 words
A method for recognizing a group of symbols in a series of symbols, comprising the steps of: collecting data that describes said group of symbols, said group of symbols having symbols comprising a plurality of micro-segments; sorting said group of symbols based on an evaluation of said data to produce a switch control output; associating a micro-segment with at least one cell in a plurality of cells that quantize micro-segment features, where each of said cells has an associated feature label; assigning a cell value to a cell, using a micro-segment value that is related to how well a feature of said micro-segment associated with said cell corresponds to said cell's feature label; using a plurality of classifiers that receive said cell values assigned to said plurality of cells for recognizing said group of symbols; and communicating said cell values to at least one of said plurality of classifiers in response to said switch control output.
Claim 17
Dependent Claim
20 words
The method of claim 13, further comprising the step of separating said series of symbols into said group of symbols.
Claim 18
Dependent Claim
35 words
The method of claim 13 wherein said step of assigning assigns said cell value to said cell using said micro-segment value that is also related to a length of said micro-segment associated with said cell.
Claim 19
Dependent Claim
29 words
The method of claim 13 wherein said step of assigning assigns said cell value using a sum of micro-segment values from a plurality of micro-segments associated with said cell.
Claim 20
Dependent Claim
22 words
The method of claim 13, wherein said step of assigning assigns said cell value based on a micro-segment associated with another cell.
Claim 21
Dependent Claim
19 words
The method of claim 13, wherein said step of using said plurality of classifiers comprises using a neural network.
Claim 22
Dependent Claim
20 words
The method of claim 13, wherein said step of using said plurality of classifiers comprises using a convolution neural network.
Claim 23
Independent Claim
175 words
A method for recognizing a group of symbols in a series of symbols, comprising the steps of: collecting data that describes said group of symbols, said group of symbols having symbols comprising a plurality of micro-segments; separating said series of symbols into individual symbols; sorting said individual symbols based on an evaluation of said data to produce a switch control output; associating a micro-segment with at least one cell in a plurality of cells that quantize micro-segment features, where each of said cells has an associated feature label; assigning a cell value to a cell, using a micro-segment value that is related to how well a feature of said micro-segment associated with said cell corresponds to said cell's feature label; generating recognized symbols using a plurality of classifiers that receive said cell values assigned to said plurality of cells for recognizing said individual symbols; communicating said cell values to at least one of said plurality of classifiers in response to said switch control output; and postprocessing said recognized symbols to recognize said group of symbols.
Claim 24
Independent Claim
175 words
A method for recognizing a group of symbols in a series of symbols, comprising the steps of: collecting data that describes said group of symbols, said group of symbols having symbols comprising a plurality of micro-segments; separating said series of symbols into individual symbols; sorting said individual symbols based on an evaluation of said data to produce a switch control output; associating a micro-segment with at least one cell in a plurality of cells that quantize micro-segment features, where each of said cells has an associated feature label; assigning a cell value to a cell, using a micro-segment value that is related to how well a feature of said micro-segment associated with said cell corresponds to said cell's feature label; generating recognized symbols using a plurality of classifiers that receive said cell values assigned to said plurality of cells for recognizing said individual symbols; communicating said cell values to at least one of said plurality of classifiers in response to said switch control output; and postprocessing said recognized symbols to recognize said group of symbols.
References Cited (Backward Citations)
US3133266A
Cited
US4284975A
Cited
US4317109A
Cited
US4561105A
Cited
US4731857A
Cited
US4803736A
Cited
US4888814A
Cited
US5058182A
Cited
US5093803A
Cited
US5067164A
Cited
US5005205A
Cited
US5058179A
Cited
US5029223A
Cited
US5121441A
Cited
US5155801A
Cited
US5212739A
Cited
US5105468A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 2000-07-07 | 2001-01-06 | 2001-07-08 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2004-07-07 | 2005-01-06 | 2005-07-08 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2008-07-07 | 2009-01-06 | 2009-07-08 | $7700.00 | Window Closed / Lapsed |