Method and apparatus for character recognition based upon the frequency of occurrence of said characters

Claim 3

The method of claim 1 further comprising the step of sequentially applying to any image data not recognized by the application of the second set of discriminatory logic tests, at least one additional set of discriminatory tests, each additional set of discriminatory logic tests being applied independently of the other sets of discriminatory logic tests for recognizing only the unrecognized image data that belongs to an additional subset of characters and recognizing only the image data that belongs to each additional subset of characters, each succeeding additional subset of characters having a lower frequency of occurrence than each preceding subset of characters.

Claim 4

The method of claim 1 wherein the application of each of the sets of a discriminatory logic tests comprises applying a discriminatory binary tree logic and sequentially applying a verifier logic.

Claim 7

The method of claim 4 wherein the application of the discrete sets of discriminatory logic tests of the first stage of discriminatory tests further includes: sequentially applying to the unrecognized image data a second set of discriminatory logic tests for recognizing only the unrecognized image data that belongs to a second subset of characters, the second subset of characters having a lower frequency of occurrence than the first subset of characters and containing at least some characters not in the first subset of characters.

Claim 8

Apparatus for processing electric image data for recognizing unknown characters of a known character set based upon the frequency of occurrence of said characters, said apparatus comprising: means for converting the information representing the unknown characters into electric image data; means for storing the image data of unknown characters; means for segregating into at least two independent and discrete sets discriminatory logic tests for recognizing the stored image data, the first set of logic tests for recognizing the data representing a first subset of the known character set, the first subset being characters having the highest frequency of occurrence and containing less than all of the characters of the complete character set, and the second set of logic tests for recognizing a second subset of the known character set, the second subset being characters having a lower frequency of occurrence than said first subset and containing at least some characters not in the first subset of characters; means for applying the first set of discriminatory logic tests to the stored image data for recognizing only the image data that belongs to the first subset of characters; and means for sequentially applying the second set of discriminatory logic tests to the image data not recognized by the application of the first set of discriminatory logic tests, the second set of logic tests being applied independently of the first set of logic tests for recognizing only the image data that belongs to the second subset of characters.

Claim 9

Apparatus for processing electric image data for recognizing unknown characters of a known character set based upon the frequency of occurrence of said characters, said apparatus comprising: means for converting the information representing the unknown characters into electric image data; means for storing the image data of unknown characters; means for segregating into at least two independent and discrete sets discriminatory logic tests for recognizing the stored image data, the first set of logic tests for recognizing the data representing a first subset of the known character set, the first subset being characters having the highest frequency of occurrence and containing less than all of the characters of the complete character set, and the second set of logic tests for recognizing a second subset of the known character set, the second subset being characters having a lower frequency of occurrence than said first subset and containing at least some characters not in the first subset of characters; means for applying the first set of discriminatory logic tests to the stored image data for recognizing only the image data that belongs to the first subset of characters; and means for sequentially applying the second set of discriminatory logic tests to the image data not recognized by the application of the first set of discriminatory logic tests, the second set of logic tests being applied independently of the first set of logic tests for recognizing only the image data that belongs to the second subset of characters.

Claim 10

The apparatus of claim 6 further comprising means for applying at least one additional set of discriminatory tests to any unrecognized data from the application of the second set of discriminatory logic tests, each additional set of discrimatory logic tests being applied independently of the other sets of discriminatory logic tests for recognizing only the unrecognized image data that belongs to an additional subset of characters and recognizing only the image data that belongs to each additional subset of characters, each succeeding additional subset of characters, having a lower frequency of occurrence than each preceding subset of characters.

Claim 11

Apparatus for processing electric image data for recognizing unknown characters of a known character set based upon the frequency of occurrence of said characters, said apparatus comprising: means for converting the information representing the unknown characters into electric image data; means for storing the image data of unknown characters; means for segregating into at least two independent and discrete stages discriminatory logic tests for recognizing the data representing all characters of the known character set, each stage including multiple discrete sets of discriminatory logic tests for recognizing the data representing discrete subsets of the known character; means for applying the first stage of discriminatory logic tests to the stored image data for recognizing all the image data as characters and generating recognized character data and unrecognized character data, wherein the means for applying the discrete sets of discriminatory logic tests of the first stage of discriminatory tests includes means for applying the first set of discriminatory logic tests for recognizing only the image data that belongs to the first subset of characters, the first subset of characters having the highest frequency of occurrence and containing less than all of the characters of the complete character set; and means for sequentially applying a second stage of discriminatory logic tests independently of the first stage of discriminatory logic tests to only the unrecognized character data, the means for applying the second stage of discriminatory logic tests including means for sequentially applying discrete sets of independent discriminatory logic tests different from the discriminatory logic tests applied in the first stage of discriminatory logic tests.

Claim 12

Apparatus for processing electric image data for recognizing unknown characters of a known character set based upon the frequency of occurrence of said characters, said apparatus comprising: means for converting the information representing the unknown characters into electric image data; means for storing the image data of unknown characters; means for segregating into at least two independent and discrete stages discriminatory logic tests for recognizing the data representing all characters of the known character set, each stage including multiple discrete sets of discriminatory logic tests for recognizing the data representing discrete subsets of the known character; means for applying the first stage of discriminatory logic tests to the stored image data for recognizing all the image data as characters and generating recognized character data and unrecognized character data, wherein the means for applying the discrete sets of discriminatory logic tests of the first stage of discriminatory tests includes means for applying the first set of discriminatory logic tests for recognizing only the image data that belongs to the first subset of characters, the first subset of characters having the highest frequency of occurrence and containing less than all of the characters of the complete character set; and means for sequentially applying a second stage of discriminatory logic tests independently of the first stage of discriminatory logic tests to only the unrecognized character data, the means for applying the second stage of discriminatory logic tests including means for sequentially applying discrete sets of independent discriminatory logic tests different from the discriminatory logic tests applied in the first stage of discriminatory logic tests.

Claim 13

The apparatus of claim 8 wherein the means for applying the discrete sets of discriminatory logic tests of the first stage of discriminatory tests further includes: means for sequentially applying to the unrecognized image data a second set of discriminatory logic tests for recognizing only the unrecognized image data that belongs to a second subset of characters, the second subset of characters having a lower frequency of occurrence than the first subset of characters and containing at least some characters not in the first subset of characters.