Character recognition method

Claim 2

A character recognition method comprising the steps of: (1) acquiring an image from a solid state TV camera, said image comprising a two dimensional array of pixels, said pixels having grey scale values associated therewith; (2) locating a character within said image by calculating intensity gradients and determining when said intensity gradients exceed predetermined values; (3) determining the size of said character measured in pixels; (4) reducing the number of pixels associated with said character by a first determining the number of pixels per dot required to fit said character into a smaller dot matrix and then; (5) combining the pixels associated with each dot to form a matrix of dot data points having fewer dots than pixels; (6) generating a grey scale histogram for said dot data points; (7) using said histogram to determine a threshhold for the grey scale values associated with the dot data points; and (8) assigning binary values to said dot data points based on said threshhold to form a binary representation of said character.

Claim 3

A method according to claim 1 comprising the further step of: (9) displaying the digital representation of said character.

Claim 4

A method according to claim 1 comprising the further steps of: (9) displaying the digital representation of said character; and (10) assigning an ASCII character to the displayed character.

Claim 5

A method according to claim 1 comprising the further steps of: (9) displaying the digital representation of said character; (10) assigning an ASCII character to the displayed character; and (11) storing the displayed character.

Claim 6

A method according to claim 1 comprising the further steps of: (9) displaying the digital representation of said character; (10) assigning an ASCII character to the displayed character; (11) storing the displayed character; and (12) repeating steps (2) through (11) to build up a library comprising a plurality of characters.

Claim 7

A method according to claim 1 comprising the further step of: (9) comparing the digital representation of said character with digital representations of characters stored in a library.

Claim 8

A method according to claim 1 comprising the further steps of: (9) comparing the digital representation of said character with digital representations of characters stored in a library; and (10) repeating steps (2) through (9).

Claim 12

A method according to claim 10 wherein said 8 by 10 array comprises 10 rows having 8 dots per row.

Claim 13

A method according to claim 11 wherein said digital representation of said character comprises ten 8 bit bytes.

Claim 16

A character recognition method comprising the steps of: (1) acguiring an image from a solid state TV, camera, said image comprising a two dimensional array of pixels, said pixels having grey scale values associated therewith; (2) locating a character within said image by calculating intensity gradients and determining when said intensity gradients exceed predetermined values; (3) scanning the area in which said character is located to determine the left, right, top and bottom most edges of the character; (4) determining the size of said character measured in pixels by differencing the left and right and top and bottom most edges, respectively; (5) reducing the number of pixels associated with said character by first determining the number of pixels per dot required to fit said character into a smaller dot matrix and then; (6) combining the pizels associated with each dot to form a matrix of dot data points having fewer dots than pixels; (7) generating a grey scale histogram for said dot data points; (8) using said histogram to determine a threshhold for the grey scale values associated with the dot data points; and (9) assigning binary values to said dot data points based on said threshhold to form a binary representation of said character.

Claim 17

A character recognition method comprising the steps of: (1) acguiring an image from a solid state TV, camera, said image comprising a two dimensional array of pixels, said pixels having grey scale values associated therewith; (2) locating a character within said image by calculating intensity gradients and determining when said intensity gradients exceed predetermined values; (3) scanning the area in which said character is located to determine the left, right, top and bottom most edges of the character; (4) determining the size of said character measured in pixels by differencing the left and right and top and bottom most edges, respectively; (5) reducing the number of pixels associated with said character by first determining the number of pixels per dot required to fit said character into a smaller dot matrix and then; (6) combining the pizels associated with each dot to form a matrix of dot data points having fewer dots than pixels; (7) generating a grey scale histogram for said dot data points; (8) using said histogram to determine a threshhold for the grey scale values associated with the dot data points; and (9) assigning binary values to said dot data points based on said threshhold to form a binary representation of said character.

Claim 18

A method according to claim 15 wherein the pixels are combined by averaging the grey scale values of the pixels associated with each dot.

Claim 19

A method according to claim 16 wherein said dot matrix comprises an 8 by 10 array of 80 dots.

Claim 20

A method according to claim 17 wherein said 8 by 10 array comprises 10 rows having 8 dots per row.

Claim 21

A method according to claim 18 wherein said digital representation of said character comprises ten 8 bit byty.