Data processing using digitized images

• US 5,396,588 A
• Filed: 07/03/1990
• Issued: 03/07/1995
• Est. Priority Date: 07/03/1990
• Status: Expired due to Term

• Current Assignee: Horst Froessl
• Original Assignee: Horst Froessl
• Inventors: Froessl, Horst
• Primary Examiner(s): Zimmerman, Mark K.
• Assistant Examiner(s): FEILD, JOSEPH H
• Application Number:
  US07/547,190
• Time in Patent Office: 1,708 Days
• Field of Search: 395/145, 395/150, 395/600, 382/11, 382/48, 382/69, 364/419, 364/963, 364/419.19
• US Class Current: 715/234
• CPC Class Codes: G06F 16/5846 using extracted text
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/245 Font recognition

Abstract

A method of manipulating information is disclosed in which the data is stored as a digitized image and is retained in image form for various data processing manipulations. A font table is formed having a matrix of fonts correlated with characters and symbols in code form such as ASCII. Desired material in the stored documents is located using pattern-match searching with a parallel processor search engine.

Claims

1. A method of data processing comprising

storing digitized images of document contents including text in at least one font,

establishing in non-volatile memory a font table including code values of alphanumeric characters and symbols and images of characters and symbols in each font used in the document text, each character and symbol in each font being correlated with the code value for the character or symbol,

selecting and locating digitized images of selected portions of stored documents to be manipulated by data processing, the step of locating including

defining a search word which appears in the stored documents,

entering code representing the search word,

compiling from the font table a plurality of image words constructed of a sequence of images of characters which are the same characters as the search word and in image in a plurality of fonts,

comparing the image words with the stored digitized images until a match is found, and

copying into a memory the stored image of the document in which the match is found;

and

processing the selected digitized images.

2. A method according to claim 1 wherein the step of selecting includes

selecting a plurality of words from the stored text and copying stored images of the selected words into a temporary store,

sequentially evaluating alphabetical positions of images of successive characters in each of the plurality of words, and

arranging the images of the words in alphabetical order correlated with the alphabetical positions of the successive character images.

3. A method according to claim 2 wherein the stored image copies are in at least two different fonts.

4. A method according to claim 1 wherein the document has a plurality of lines of imprinted characters in a font and the step of storing includes

digitizing successive lines of imprinted characters in a document to form digitized images of the document contents,

comparing selected characters of the digitized images with the font table to determine whether a match exists between an existing font in the table and a font in which the document is imprinted, and,

when no match is found to exist, adding images of the selected characters to the font table correlated with the code values of those characters.

5. A method according to claim 4 and including correlating fonts used in documents emanating from a specific source with an identification of that source.

6. A method according to claim 1 wherein said digitized images are stored in a mass data image store, and wherein the step of locating includes

retrieving into volatile memory a portion of said digitized images,

establishing at least one pattern representative of a pattern in the image store which is to be located, and

comparing the at least one pattern with the digitized images stored in the volatile memory.

7. A method of data processing comprising

storing digitized images of document contents including text in at least one font,

establishing in non-volatile memory a font table including code values of alphanumeric characters and symbols and images of characters and symbols in each font used in the document text, each character and symbol in each font being correlated with the code value for the character or symbol,

selecting and locating digitized images of selected portions of stored documents to be manipulated by data processing, the step of locating each image including

forming a table of word images selected and copied from stored documents for use as search words,

displaying the search word images,

selecting a search word image which appears in the stored documents from the display of word images,

comparing each selected search word image with the stored digitized images until a match is found, and

copying the stored image of the document containing the word into volatile memory;

and

processing the selected digitized images.

8. A method according to claim 7 wherein the step of selecting includes selecting a plurality of images of a search word, each image being in a different one of a plurality of fonts,

and wherein the method further includes

providing a plurality of data processors for performing comparisons of images, and

assigning a different data processor to each font of said plurality of fonts for parallel comparison.