EP0079251A2
Expired / Lapsed
EP
⏱️ 0.0 years remaining
Geometric character recognition using skeleton and stroke width representation
Assignee: Unisys Corp
•
Filing: 1982-07-30
•
Granted: —
•
Expires: 2002-07-30
Abstract
Abstract character recognition system is shown which makes use of skeleton and stroke width data in digital form. Data representing certain features such as curves, lines and loops of the skeleton are extracted by a feature extractor to provide character codes representing the digits "0" through "9". The character codes are compared then with reference standards in a look-up table which provides a classification of the character codes according to whether there is (1) a direct match with standard character codes in the look-up table, (2) a "confused class" indicating that the particular character code represents more than one character or class of data in the table or (3) a "noise code" indicating the particular character code cannot be identified from data in the table. In case of (1) above the identification is transmitted for immediate use. In cases (2) and (3) further analysis of characteristics and modifications of characteristics of the lines, curves and loops as well as analysis of stroke width are employed to enable identification of additional character codes.
Inventors
- 👤 Pepe Siy
- 👤 James Ching-Ting Young
Bibliographic Details
Application Number
—
Jurisdiction
EP
Data Source
google_patents
Last Synchronized
2026-11-02 20:02
CPC Classifications
G; G06; G06V; G06V10/00; G06V10/20; G06V10/34; G06V10/30; G06V10/40; G06V10/70; G06V10/74; G06V10/75
Claims Analysis
Total Claims: 19
Claim 1
Independent Claim
72 words
A character recognition system for processing data in the form of signals representing geometric features of characters, including skeleton and stroke width characteristics, comprising: means for extracting a character code in the form of data representing selected features of a character from said signals; and means for comparing said character code with data representing character codes of known characters and, when positive identification is made, providing output signals designating the identified character.
Claim 2
Dependent Claim
72 words
A character recognition system for processing data in the form of signals representing geometric features of characters, including skeleton and stroke width characteristics, comprising: means for extracting a character code in the form of data representing selected features of a character from said signals; and means for comparing said character code with data representing character codes of known characters and, when positive identification is made, providing output signals designating the identified character.
Claim 3
Dependent Claim
77 words
The invention as claimed in Claim 1, in which: said means for comparing said data, when positive identification is not made, including means for transmitting said character code for further processing; processing means coupled responsive to receipt of said character code to make an analysis thereof and to produce an altered character code incorporating information generated from said analysis, and means for routing said altered character code to said means for comparing in search of positive identification.
Claim 4
Dependent Claim
76 words
The invention as claimed in Claim 1, in which: said means for comparing said data, when positive identification is not made, including means for transmitting said character code for further processing; processing means coupled responsive to said character code including a confused code processor for applying fine feature tests to distinguish between different characters having identical character codes to establish positive identification of a character; said confused code processor providing an output designating said identified character.
Claim 5
Dependent Claim
59 words
The invention as claimed in Claims 2 or 3, in which said processing means comprises: a noise code processor for modifying a character code which is not directly identifiable as belonging to any character class in the look-up table; said noise code processor providing a modified character code to be routed to the look-up table for comparison and identification.
Claim 6
Dependent Claim
29 words
The invention as claimed in Claim 4, in which the noise code processor comprises: means to remove selected elements of said character code and generate said modified character code.
Claim 7
Dependent Claim
53 words
The invention as claimed in Claim 5, in which: the noise code processor examines each node of a character code to which exactly two branches are attached; said processor concatenating the two branches, reducing them to a single branch, and routing the resulting modified character to the look-up table for comparison and identification.
Claim 8
Dependent Claim
32 words
The invention as claimed in Claim 6, in which: the noise code processor terminates the search in the event that two concatenations are attempted without securing a match in the look-up table.
Claim 9
Dependent Claim
35 words
The invention as claimed in Claim 4 in which the noise code processor includes means for deleting portions of branches without destroying connectivity, said branches including tip branches, big loop branches and big bubble branches.
Claim 10
Dependent Claim
83 words
A character recognition system for use with a document processing machine in which data is received in the form of signals representing geometric features of characters, comprising: means for extracting data representing distinct features of a character from said signals; processor means for comparing said data with data representing features of known characters; said processor means providing output signals designating a particular character when one is identified; and said processor means routing said data to additional processors when no particular character is recognized.
Claim 11
Dependent Claim
83 words
A character recognition system for use with a document processing machine in which data is received in the form of signals representing geometric features of characters, comprising: means for extracting data representing distinct features of a character from said signals; processor means for comparing said data with data representing features of known characters; said processor means providing output signals designating a particular character when one is identified; and said processor means routing said data to additional processors when no particular character is recognized.
Claim 12
Dependent Claim
46 words
The invention claimed in Claim 9, in which: the data represents the shape of the skeleton including node points interconnected by a plurality of branch points; and the means for extracting data includes means for distinguishing between data representing a line, a curve and a loop.
Claim 13
Dependent Claim
57 words
The invention as claimed in Claim 9 or 10, in which: the additional processors respond to receipt of said data to make an analysis thereof and produce altered data incorporating information generated from said analysis, and means for routing said altered data to said processor means for comparing said data with data representing features of known characters.
Claim 14
Dependent Claim
56 words
The invention as claimed in Claim 9 or 10, in which: the additional processors coupled responsive to said altered data include a confused code processor for applying fine feature tests to distinguish between different characters having identical character codes to establish positive identification of a character; said confused code processor providing an output designating identified characters.
Claim 15
Dependent Claim
57 words
The invention as claimed in Claims 9 or 10, in which said processing means comprises: a noise code processor for modifying a character code which is not directly identifiable as belonging to any character class in the processor; said noise code processor providing a modified character code to be routed to the processor for comparison and identification.
Claim 16
Dependent Claim
29 words
The invention as claimed in Claim 13, in which the noise code processor comprises: means to remove selected elements of said character code and generate said modified character code.
Claim 17
Dependent Claim
53 words
The invention as claimed in Claim 13, in which: the noise code processor examines each node of a character code to which exactly two branches are attached; said processor concatenating the two branches, reducing them to a single branch, and routing the resulting modified character to the look-up table for comparison and identification.
Claim 18
Dependent Claim
32 words
The invention as claimed in Claim 15, in which: the noise code processor terminates the search in the event that two concatenations are attempted without securing a match in the look-up table.
Claim 19
Dependent Claim
39 words
The invention as claimed in Claim 13 in which the noise code includes means for modifying a character code by deleting portions of branches without destroying connectivity, said branches including tip branches, big loop branches and big bubble branches.
References Cited (Backward Citations)
No citation records indexed.
USPTO Annuity & Maintenance Windows
Maintenance schedule is calculated for granted US utility patents.