Method and apparatus for sorting and comparing linear configurations

• US 6,332,040 B1
• Filed: 11/04/1997
• Issued: 12/18/2001
• Est. Priority Date: 11/04/1997
• Status: Expired due to Fees

• Current Assignee: Jones J. Howard
• Original Assignee: Jones J. Howard
• Inventors: Jones, J. Howard
• Primary Examiner(s): Au, Amelia
• Assistant Examiner(s): Dastouri, Mehrdad
• Application Number:
  US08/964,298
• Time in Patent Office: 1,505 Days
• Field of Search: 382/119, 382/123, 382/185, 382/186, 382/187, 382/189, 382/193, 382/197, 382/203, 382/305
• US Class Current: 382/197
• CPC Class Codes: G06F 16/56 having vectorial format
• CPC Class Codes: G06V 30/1423 the instrument generating sequences of position coordinates corresponding to handwriting preprocessing or recognising digital ink G06V30/32

Abstract

A computing device to classify, manipulate, store, search, and selectively retrieve items from image databases, using a method & apparatus including a computer processor to analyze the configured disposition of a multiplicity of linear motion paths that constitute a glyph delineated within a defined planar rendering area and to process, manipulate, and compare data representing said configured disposition of said linear motion paths, comprising (a) a template analytic lattice constituted of a predetermined multiplicity of cell partitions, the totality of the analytic lattice being coextensive with the defined rendering area of the glyph and in the plane of motion of the linear motion paths of the glyph; (b) means of interpreting each linear motion path as a sequence of displacement vectors, determining which displacement vectors intersect specific boundaries of specific cells of the analytic lattice, and registering each cell boundary intersection by both recording the intersecting vector inclination gradient and incrementing an associated vector count for each cell bounded by the intersected boundary; (c) means of combining all of the vector inclination gradients registered for each cell, thereby producing an associated numeric clutch value as a cumulative total of the registered vector inclination gradients, and (d) means of deriving a numeric inclination index value for each cell from the associated vector count and the associated clutch value, the kind of inclination index being selected from a group defined herein.

Claims

1. In connection with a computer based system for classifying, manipulating, storing, searching, and selectively retrieving items from at least one image database which includes multiple glyphs, which computer based system includes

at least one computer processing unit,

programmable working memory associated with said at least one computer Processing unit,

a data storage system, containing the at least one image database, associated with said at least one computer processing unit,

persistent memory, and

input/output (“

I/O”

) means associated with at

least said at least one computer processing unit, a method of classifying, manipulating, storing, searching, and selectively retrieving items from said at least one image database, comprising the following steps;

a) setting up in working memory a template analytic lattice constituted of a predetermined multiplicity of cell partitions, the totality of the analytic lattice being coextensive with the defined rendering area of each of the glyphs and in the plane of motion of the linear motion paths of each of the glyphs;

b) interpreting in the working memory of the computer each linear motion path of each glyph as a sequence of displacement vectors, determining which displacement vectors intersect specific boundaries of specific cells of the analytic lattice, and registering in memory each cell boundary intersection by both recording the intersecting vector inclination gradient and incrementing an associated vector count for each cell bounded by the intersected boundary;

c) combining all of the vector inclination gradients registered for each cell in working memory, producing an associated numeric clutch value as a cumulative total of the registered vector inclination gradients;

d) deriving in working memory a numeric inclination index value for each cell from the associated vector count and the associated clutch value, the kind of the inclination index being selected from the group consisting of

(1) a numeric sum inclination index calculated as the specific rotational increment of a resultant vector inclination derived by adding the inclinations of all the vector displacements that intersect the cell boundaries;

(2) a numeric meld inclination index calculated as the specific rotational increment of an average vector inclination derived by dividing the cell clutch value by the associated vector count; and

(3) a numeric delta inclination index calculated as the specific rotational increment of differential vector inclination derived by subtracting the meld inclination index value from the sum inclination index value;

whereby there is no need to process representational content elements and incidental local mark features of the images in said at least one database, and whereby said at least one image database may be searched by filtering and eliminating non-viable search candidates prior to processor intensive content feature analysis through comparison of textual token keys.

2. The method of claim 1, wherein there is further included the step of:

using said at least one computer processing unit and working memory to derive for each said glyph for at least one of said kinds of inclination index-an associated textual token key comprising;

a multiplicity of text slot values corresponding to said multiplicity of said cells, including

means of representing each said cell inclination index value in each said slot as a text substring comprising a predetermined number of characters that expresses said inclination index in terms of a base number system when it has a value and denotes the absence of said inclination index value by some predetermined combination of characters; and

means of concatenating said slots in an order corresponding to a predetermined ordering of said analytic lattice cell partitions;

whereby results of a multiplicity of instances of said glyph analysis may be compared using conventional alphanumeric text string operations to derive and express useful match data regarding similarities of configured linear motion path dispositions among various subsets of said multiplicity of glyphs.

3. The method of claim 2, wherein, for a multiplicity of said glyphs contained in a collection of glyphs, each member of which is associated with at least one member of said at least one image database, and for a given kind of said textual token key, there is included the further steps of:

using said at least one computer processing unit and working memory to organize and search said at least one image database, including the sub-steps of

referencing, accessing, and retrieving said glyph by said associated textual token key;

referencing, accessing, and retrieving said associated member of said at least one image database by accessing said textual token key of said associated glyph, further including

A. compiling said textual token keys with said glyph referencing means into an alphanumerically sorted list to facilitate inserting and removing said textual token keys as glyphs are added to and removed from said collection of glyphs and further to enable rapid re-sorting, comparing, manipulating, and matching of said textual token keys as means to identifying and selecting subsets of similarly configured glyphs within said collection of glyphs; and

B. comparing corresponding slots of said textual token keys within a predetermined tolerance gamut to find and enumerate matching sets of said corresponding slots for the purpose of determining whether said enumeration of said matched slots satisfies a predetermined threshold stating the number of matching slots required;

whereby items held in said at least one image database may be screened for search and retrieval by comparing and manipulating said corresponding slots of associated glyph textual token keys.

4. The method of claim 3, wherein there is included the steps of using said at least one computer processing unit and working memory, combined with an image display and a graphic input enabling a user to visually select among groups of displayed items, to custom configure elements of a control interface, further comprising the sub-steps of:

variably setting said tolerance gamut, and said match threshold;

analyzing said configured disposition of linear motion paths of said glyph drawn with said input means to thereby derive said associated textual token key to serve as a reference search key;

comparing within said predetermined tolerance gamut said input reference search key with each of said multiplicity of pre-existing said textual token keys that each reference at least one said associated member of said at least one image database;

selecting on the basis of said comparison with input reference search key, ranking in match order, and displaying candidate items from said multiplicity of glyphs along with said associated members of said at least one image database;

organizing said ranked candidate items as an array of control elements in which a specific control function is associated with each said displayed item; and

activating said control function by said user executing some predetermined motion-with said graphic input;

whereby control elements of a user interface may be conveniently selected, deployed, and activated by said user approximating glyph linear configurations with drawn gestures that may be subsequently interpreted to a desired degree of accuracy.

5. The method of claim 4, wherein there is further included the step of:

using for said graphic input a stylus and touch sensitive tablet.

6. The method of claim 2, wherein there is included the steps of using said at least one computer processing unit and working memory, combined with a graphic input and other computer related elements to enable a user to guide multimedia operations and other computer system control events, further comprising the sub-steps of:

variably designating specific combinations of said textual token key slot positions;

associating specific computer system control events with specific values of said designated textual token key slot positions so that said specific values may cause said computer system to prepare and enable said associated specific computer system control events;

interpreting said glyph motion paths drawn by said user on said graphic input to produce said textual token keys that may be decoded into specific values that enable associated specific computer system control events; and

setting an actuator to control the pace at which said enabled associated specific computer system control events are automatically actuated at intervals signaled by said actuator;

whereby said user may variably select combinations of multimedia delivery vehicles, control their pace of delivery, and generally interact with said computer system control events by drawing various linear motion path configurations on said graphic input.

7. The method of claim 6, wherein there is further included the step of:

using for said graphic input a stylus and touch sensitive tablet.

8. A computer based system for classifying, manipulating, storing, searching, and selectively retrieving items from at least one image database which includes multiple glyphs, which computer based system includes

at least one computer processing unit,

programmable working memory associated with said at least one computer processing unit,

a data storage system, containing the image data base(s), associated with said at least one computer processing unit,

persistent memory, and

input/output (“

I/O”

) means associated with at least said at least one computer processing unit,

said persistent memory being programmed to provide in said computer system a method of classifying, manipulating, storing, searching, and selectively retrieving items from said at least one image database, comprising the following steps;

a) setting up in working memory a template analytic lattice constituted of a predetermined multiplicity of cell partitions, the totality of the analytic lattice being coextensive with the defined rendering area of each of the glyphs and in the plane of motion of the linear motion paths of each of the glyphs;

b) interpreting in the working memory of the computer each linear motion path of each glyph as a sequence of displacement vectors, determining which displacement vectors intersect specific boundaries of specific cells of the analytic lattice, and registering in memory each cell boundary intersection by both recording the intersecting vector inclination gradient and incrementing an associated vector count for each cell bounded by the intersected boundary;

c) combining all of the vector inclination gradients registered for each the cell in working memory, producing an associated numeric clutch value as a cumulative total of the registered vector inclination gradients;

d) deriving in working memory a numeric inclination index value for each cell from the associated vector count and the associated clutch value, the kind of the inclination index being selected from the group consisting of

(1) a numeric sum inclination index calculated as the specific rotational increment of a resultant vector inclination derived by adding the inclinations of all the vector displacements that intersect the cell boundaries;

(2) a numeric meld inclination index calculated as the specific rotational increment of an average vector inclination derived by dividing the cell clutch value by the associated vector count; and

(3) a numeric delta inclination index calculated as the specific rotational increment of a differential vector inclination derived by subtracting the meld inclination index value from the sum inclination index value;

whereby there is no need to process representational content elements and incidental local mark features of the images in said at least one database, and whereby said at least one image database may be searched by filtering and eliminating non-viable search candidates prior to processor intensive content feature analysis through comparison of textual token keys.

9. The computer system of claim 8, wherein said persistent memory is programmed to further include the step of:

using said at least one computer processing unit and working memory to derive for each said glyph for at least one of said kinds of inclination index an associated textual token key comprising;

a multiplicity of text slot values corresponding to said multiplicity of said cells, including

means of representing each said cell inclination index value in each said slot as a text substring comprising a predetermined number of characters that expresses said inclination index in terms of a base number system when it has a value and denotes the absence of said inclination index value by some predetermined combination of characters; and

means of concatenating said slots in an order corresponding to a predetermined ordering of said analytic lattice cell partitions;

whereby results of a multiplicity of instances of said glyph analysis may be compared using conventional alphanumeric text string operations to derive and express useful match data regarding similarities of configured linear motion path dispositions among various subsets of said multiplicity of glyphs.

10. The computer system of claim 9, wherein said persistent memory is programmed, for a multiplicity of said glyphs contained in a collection of glyphs, each member of which is associated with at least one member of said at least one image database, and for a given kind of said textual token key, to further include the steps of:

using said at least one computer processing unit and working memory to organize and search said at least one image database, to provide the further sub-steps of

referencing, accessing, and retrieving said glyph by said associated textual token key;

referencing, accessing, and retrieving said associated member of said at least one image database by accessing said textual token key of said associated glyph, further including

C. compiling said textual token keys with said glyph referencing means into an alphanumerically sorted list to facilitate inserting and removing said textual token keys as glyphs are added to and removed from said collection of glyphs and further to enable rapid re-sorting, comparing, manipulating, and matching of said textual token keys as means of identifying and selecting subsets of similarly configured glyphs within said collection of glyphs; and

D. comparing corresponding slots of said textual token keys within a predetermined tolerance gamut to find and enumerate matching sets of said corresponding slots for the purpose of determining whether said enumeration of said matched slots satisfies a predetermined threshold stating the number of matching slots required;

whereby items held in said at least one image database may be screened for search and retrieval by comparing and manipulating said corresponding slots of associated glyph textual token keys.

11. The computer system of claim 10, wherein said persistent memory is programmed to further include the steps of using said at least one computer processing unit and working memory, combined with an image display and a graphic input enabling a user to visually select among groups of displayed items, to custom configure elements of a control interface, to further provide the sub-steps of:

variably setting said tolerance gamut and said match threshold;

analyzing said configured disposition of linear motion paths of said glyph drawn with said input means to thereby derive said associated textual token key to serve as a reference search key;

comparing within said predetermined tolerance gamut said input reference search key with each of said multiplicity of pre-existing said textual token keys that each reference at least one said associated member of said at least one image database;

selecting on the basis of said comparison with input reference search key, ranking in match order, and displaying candidate items from said multiplicity of glyphs along with said associated members of said at least one image database;

organizing said ranked candidate items as an array of control elements in which a specific control function is associated with each said displayed item; and

activating said control function by said user executing some predetermined motion with said graphic input;

whereby control elements of a user interface may be conveniently selected, deployed, and activated by said user approximating glyph linear configurations with drawn gestures that may be subsequently interpreted to a desired degree of accuracy.

12. The computer system of claim 11, wherein said persistent memory is programmed to further include the step of:

using for said graphic input a stylus and touch sensitive tablet.

13. The computer system of claim 9, wherein said persistent memory is programmed to further include the steps of using said at least one computer processing unit and working memory, combined with a graphic input and other computer related elements to enable a user to guide multimedia operations and other computer system control events, to further provide the sub-steps of:

variably designating specific combinations of said textual token key slot positions;

associating specific computer system control events with specific values of said designated textual token key slot positions so that said specific values may cause said computer system to prepare and enable said associated specific computer system control events;

interpreting said glyph motion paths drawn by said user on said graphic input to produce said textual token keys that may be decoded into specific values that enable associated specific computer system control events; and

setting an actuator to control the pace at which said enabled associated specific computer system control events are automatically actuated at intervals signaled by said actuator;

whereby said user may variably select combinations of multimedia delivery vehicles, control their pace of delivery, and generally interact with said computer system control events by drawing various linear motion path configurations on said graphic input.

14. The computer system of claim 13, wherein said persistent memory is programmed to further included the step of:

using for said graphic input a stylus and touch sensitive tablet.