Printing control interface system and method with handwriting discrimination capability

• US 6,940,617 B2
• Filed: 02/09/2001
• Issued: 09/06/2005
• Est. Priority Date: 02/09/2001
• Status: Expired due to Fees

• Current Assignee: Matsushita Electric Industrial Company Limited (Panasonic Holdings Corporation)
• Original Assignee: Matsushita Electric Industrial Company Limited (Panasonic Holdings Corporation)
• Inventors: Ma, Yue, Guo, Jinhong Katherine
• Primary Examiner(s): POON, KING Y
• Application Number:
  US09/781,529
• Publication Number:
  US 20020135786A1
• Time in Patent Office: 1,670 Days
• Field of Search: 382/186, 382/187, 382/176, 382/177, 382/179, 382/218, 396/313, 715/541, 358 11- 19, 358 111- 118, 358/2.1, 358/448, 358/452, 358/462
• US Class Current: 358/1.15
• CPC Class Codes: G06V 30/2455 Discrimination between machine-print, hand-print and cursive writing
• CPC Class Codes: H04N 1/38 Circuits or arrangements for blanking or otherwise eliminating unwanted parts of pictures H04N1/387 takes precedence
• CPC Class Codes: H04N 1/40062 Discrimination between different image types, e.g. two-tone, continuous tone

Abstract

A system and method for identifying and separating handwritten text from machine printed text in a document having a combination of handwritten text and machine printed text. In one embodiment, the present invention is installed as an intelligent printing control interface in a conventional computer and allows any document generated by any application to be selectively processed to remove any handwritten text prior to printing of the document, under the control of the user. In an alternative embodiment, the present invention is installed as an intelligent printing control interface in a conventional digital copy machine, which allows any document being copied to be selectively processed to remove any handwritten text prior to printing of the document, under the control of the user. The present invention employs a handwriting separation method which uses Hidden Markov Models to identify regions as either handwritten text or machine printed text.

Claims

1. A system for printing documents, comprising:

means for generating digital documents having handwritten text and machine printed text and for generating a request to print at least one of said digital documents;

an intelligent printing control interface coupled to receive and selectively process said request to remove said handwritten text, said intelligent printing control interface comprising a virtual printer driver for receiving said request, a printing control panel for determining whether to process said request, and a handwriting separation module for selectively processing said request, under control of said printing control panel, to remove said handwritten text; and

means coupled to said intelligent printing control interface for printing said selectively processed request;

wherein said handwriting separation module processes said request by segmenting the digital image into a plurality of text blocks, computing an observation sequence for each of said text blocks, computing a first probability for each of said text blocks that said observation sequence would appear in a predetermined Hidden Markov Model for machine printed text, computing a second probability for each of said text blocks that said observation sequence would appear in a predetermined Hidden Markov Model for handwritten text, comparing said first probability and said second probability for each of said text blocks, marking said text block as machine printed text if said first probability is greater than said second probability, marking said text block as handwritten text if said second probability is greater than said first probability, and removing said marked handwritten text.

2. A system for copying documents, comprising:

a scanner for generating a digital document having handwritten text and machine printed text and for generating a request to print said digital document;

an intelligent printing control interface coupled to receive and selectively processes said request to remove said handwritten text, said intelligent printing control interface comprising a virtual printer driver for receiving said request, a printing control panel for determining whether to process said request, and a handwriting separation module for selectively processing said request, under control of said printing control panel, to remove said handwritten text; and

a printer coupled to said intelligent printing control interface for printing said selectively processed request;

wherein said handwriting separation module processes said request by segmenting the digital image into a plurality of text blocks, computing an observation sequence for each of said text blocks, computing a first probability for each of said text blocks that said observation sequence would appear in a predetermined Hidden Markov Model for machine printed text, computing a second probability for each of said text blocks that said observation sequence would appear in a predetermined Hidden Markov Model for handwritten text, comparing said first probability and said second probability for each of said text blocks, marking said text block as machine printed text if said first probability is greater than said second probability, marking said text block as handwritten text if said second probability is greater than said first probability, and removing said marked handwritten text.

3. A method of separating handwritten text from machine printed text in a digital image containing a mixture of handwritten text and machine printed text, comprising the steps of:

A. segmenting the digital image into a plurality of text blocks;

B. computing an observation sequence for each of said text blocks;

C. computing a first probability for each of said text blocks that said observation sequence would appear in a predetermined Hidden Markov Model for machine printed text;

D. computing a second probability for each of said text blocks that said observation sequence would appear in a predetermined Hidden Markov Model for handwritten text;

E. comparing said first probability and said second probability for each of said text blocks;

F. marking said text block as machine printed text if said first probability is greater than said second probability; and

G. marking said text block as handwritten text if said second probability is greater than said first probability.

4. The method of claim 3, further comprising the step of post-processing the digital image after each of said text blocks are marked as either machine printed text or handwritten text to correct single errors.

5. The method of claim 3, wherein said segmenting step comprises the steps of:

generating connecting components within the digital image; and

grouping said connected components into a plurality of text blocks.

6. The method of claim 5, wherein said grouping step comprises the steps of:

determining if each pair of neighboring connected components are on the same line;

for a given pair of neighboring connected components on the same line, determining if said pair of neighboring connected components are adjacent to each other;

for a given pair of neighboring connected components adjacent to each other, determining if said pair of neighboring connected components have the same height; and

if said pair of neighboring connected components have the same height, merging said pair of neighboring connected components together into a single text block.

7. The method of claim 6, wherein said step of determining if said pair of neighboring connected components are on the same line comprises the steps of:

identifying minimum vertical coordinates for each of said pair of connected components;

determining if the difference between said minimum vertical coordinates of said neighboring connected components is less than a first predetermined threshold determining; and

when the difference between said minimum vertical coordinates of said neighboring connected components is less than a first predetermined threshold determining, marking said pair of connected components as being on the same line.

8. The method of claim 6, wherein said step of determining if said pair of neighboring connected components are adjacent to each other further comprises the steps of:

identifying maximum and minimum horizontal coordinates for each of said neighboring connected components;

determining if the difference between said maximum horizontal coordinate of a first of said neighboring connected components and said minimum horizontal coordinate of a second of said neighboring connected components is less than a second predetermined threshold;

determining if the difference between said maximum horizontal coordinate of said second of said neighboring connected components and said minimum horizontal coordinate of said first of said neighboring connected components is less than said second predetermined threshold; and

when the difference between said maximum horizontal coordinate of said first of said neighboring connected components and said minimum horizontal coordinate of said second of said neighboring connected components is less than said second predetermined threshold or when the difference between said maximum horizontal coordinate of said second of said neighboring connected components and said minimum horizontal coordinate of said first of said neighboring connected components is less than said second predetermined threshold, marking said pair of connected components as being adjacent to each other.

9. The method of claim 6, wherein said step of determining if said pair of neighboring connected components have the same height further comprises the steps of:

identifying maximum and minimum vertical coordinates for each of said neighboring connected components;

determining the difference between the maximum and minimum vertical coordinate of a first of said neighboring connected components less the difference between the maximum and minimum coordinates of a second of said neighboring connected components is less than a third predetermined threshold; and

when the difference between the maximum and minimum vertical coordinate of said first of said neighboring connected components less the difference between the maximum and minimum coordinates of said second of said neighboring connected components is less than a third predetermined threshold, marking said pair of connected components as having the same height.

10. A method for copying and printing documents, comprising the steps of:

generating a digital image of a document having both handwritten text and machine printed text;

generating a request to print said digital image;

processing said request to selectively remove said handwritten text, said step of processing said request further comprises the steps of;

A. segmenting said digital image into a plurality of text blocks;

B. computing an observation sequence for each of said text blocks;

C. computing a first probability for each of said text blocks that said observation sequence would appear in a predetermined Hidden Markov Model for machine printed text;

D. computing a second probability for each of said text blocks that said observation sequence would appear in a predetermined Hidden Markov Model for handwritten text;

E. comparing said first probability and said second probability for each of said text blocks;

F. marking said text block as machine printed text if said first probability is greater than said second probability; and

G. marking said text block as handwritten text if said second probability is greater than said first probability; and

transmitting said selectively processed request to a printer for printing thereof.