Radical-base classification of East Asian handwriting
AssigneeMicrosoft Corporation
InventorsFrank J. Eisenhart, James A. Pittman, Patrice Y. Simard
Life of this patent
Estimating the term…
Abstract
A system and method for labeling radicals in East Asian characters is described. The identity of the radical and the location of the radical in a character may be stored for future reference.
Claims16
A radical labeling system comprising:
- an input that receives an East Asian character;
- a processor that identifies a suspected radical in said East Asian character and searches a list of radicals for said suspected radical and determines whether the suspected radical is present in the list of radicals, wherein said suspected radical is a subset of strokes within said East Asian character;
- and
- a storage database that, upon the determination that the suspected radical is present in the list of radicals, stores, with said character, an association between an identity of said suspected radical and a position in said character of said suspected radical in relation to other radicals, wherein said position in said character of said suspected radical is based on a spatial layout of Unicode that uses a Unicode value to identify the position of the suspected radical in relation to other radicals within the character, and wherein said spatial layout includes left-and-right layout information, above-and-below layout information and bracketing information.
Computer-storage media having computer-useable instructions embodied thereon for performing a method of labeling radicals in East Asian characters, the method comprising the steps of:
- receiving a character;
- finding a plurality of strokes in said character that form a suspected radical based on a search within said character for at least closed shapes, singular lines, parallel lines, intersection points and spaces between lines and shapes;
- searching a list of radicals for said suspected radical;
- based on a spatial layout of Unicode, identifying a position of said suspected radical in relation to other radicals in said character, wherein said spatial layout includes left-and-right layout information, above-and-below layout information and bracketing information, and wherein the spatial layout of Unicode identifies whether the suspected radical is positioned within the character to the left, to the right, above, below, or is bracketed, in relation to other radicals within the character;
- providing an association between said position of said suspected radical in relation to other radicals in said character and an identity of said character;
- and
- storing, with said character, the association between said position of said suspected radical in said character and said identity of said character.
Computer-storage media having computer-useable instructions embodied thereon for performing a method of labeling radicals of East Asian characters, the method comprising:
- creating a class of radicals;
- partitioning said class of radicals into regions, wherein the regions are parts of a character layout in which the radicals appear;
- identifying radicals in said regions;
- minimizing a number of possible layouts of regions;
- and
- minimizing a number of radicals per region.
Dependencies are inferred from the claim text.