Method for performing character recognition on a pixel matrix
First Claim
1. A method for recognizing ciphers, said method comprising the steps of:
- inputting a digital representation of a cipher, along with a bits-to-pixel ratio of said digital representation, the number of pixels per row of said digital representation, the number of rows in said digital representation, and an indicator which represents a block boundary condition of said digital representation;
transforming said digital representation so that each individual pixel of image information is represented by an addressable data block;
smoothing said transformation;
locating enclosures in said smoothed transformation, said enclosures defining spatial boundaries;
sizing said smoothed transformation, resulting in selected erased enclosures;
restoring all erased enclosures in said sized transformation;
normalizing said enclosure-restored transformation so that stray marks are erased, to form a normalized image;
thinning said normalized image;
storing said thinned image in a sparse space matrix;
locating significant features in said thinned image;
quantizing location of said significant features; and
performing database searches using said quantized values as search keys to identify ciphers associated with said search keys.
1 Assignment
0 Petitions
Accused Products
Abstract
Pattern recognition, particularly character recognition, is effected by categorizing a cipher by a unique description using three feature sets identifying the type and location of lines which form the cipher, the location of the endpoints of each of the lines, and the location of the intersections of the lines. In the process according to the invention, lines are located by expanding or contracting the original image of the cipher to fit into a sparse space pixel matrix. The image is then thinned to define lines and so as not to obliterate lines which give the cipher its uniqueness. Once the cipher is thinned, fuzzy logic is used to "locate", that is, define a set of coordinates for, each line relative to an x-y coordinate system. (As each line is located, it is removed from the matrix image to facilitate locating the remaining lines.) After all the lines are located, the endpoints of each line are located and the intersect points are located. These three feature sets are encoded, resulting in three values, such as 64-bit integers. These three integers are then used as search keys to query a relational database which contains the integer sets for known characters. If a direct match is not found, a matching algorithm uses partial matches to extrapolate the correct character. Because the matrix is sparse, the likelihood of identifying the correct character is enhanced.
-
Citations
24 Claims
-
1. A method for recognizing ciphers, said method comprising the steps of:
-
inputting a digital representation of a cipher, along with a bits-to-pixel ratio of said digital representation, the number of pixels per row of said digital representation, the number of rows in said digital representation, and an indicator which represents a block boundary condition of said digital representation; transforming said digital representation so that each individual pixel of image information is represented by an addressable data block; smoothing said transformation; locating enclosures in said smoothed transformation, said enclosures defining spatial boundaries; sizing said smoothed transformation, resulting in selected erased enclosures; restoring all erased enclosures in said sized transformation; normalizing said enclosure-restored transformation so that stray marks are erased, to form a normalized image; thinning said normalized image; storing said thinned image in a sparse space matrix; locating significant features in said thinned image; quantizing location of said significant features; and performing database searches using said quantized values as search keys to identify ciphers associated with said search keys. - View Dependent Claims (3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23)
-
-
2. A method for recognizing ciphers, said method comprising the steps of:
-
inputting a digital representation of a cipher, along with a bits to pixel ratio of said digital representation, the number of pixels per row of said digital representation, the number of rows in said digital representation, and an indicator which represents a block boundary condition of said digital representation; transforming said digital representation so that one pixel of image information is represented by a block; smoothing said transformation wherein said smoothing step comprises the steps of; testing the nearest eight neighbors of each "off" pixel; testing the nearest forty-eight neighbors of said "off" pixel if said nearest eight neighbors are all "on"; and turning said pixel on if any of the said nearest forty-eight neighbors are "off"; locating enclosures in said smoothed transformation, said enclosures defining spatial boundaries; sizing said smoothed transformation, resulting in selected erased enclosures; restoring all erased enclosures in said sized transformation; normalizing said enclosure-restored transformation so that stray marks are erased, to form a normalized image; thinning said normalized image; storing said thinned image in a sparse space matrix; locating significant features in said thinned image; quantizing location of said significant features; and performing database searches using said quantized values as search keys to identify ciphers associated with said search keys.
-
-
24. A method for recognizing ciphers, said method comprising the steps of;
-
inputting a digital representation of a cipher, along with a bits to pixel ratio of said digital representation, the number of pixels per row of said digital representation, the number of rows in said digital representation, and an indicator which represents a block boundary condition of said digital representation; transforming said digital representation so that one pixel of image information is represented by a block; smoothing said transformation wherein said smoothing step comprises the steps of; testing adjacent neighbors of each "off" pixel; testing next nearest neighbors of said "off" pixel if said adjacent neighbors are all "on"; and turning said pixel on if any of the said next nearest neighbors are "off"; locating enclosures in said smoothed transformation, said enclosures defining spatial boundaries; sizing said smoothed transformation, resulting in selected erased enclosures; restoring all erased enclosures in said sized transformation; normalizing said enclosure-restored transformation so that stray marks are erased, to form a normalized image; thinning said normalized image; storing said thinned image in a sparse space matrix; locating significant features in said thinned image; quantizing location of said significant features; and performing database searches using said quantized values as search keys to identify ciphers associated with said search keys.
-
Specification