Electronic color dropout utilizing spatial context to enhance accuracy

• US 20040042657A1
• Filed: 09/04/2002
• Published: 03/04/2004
• Est. Priority Date: 09/04/2002
• Status: Active Grant

• Current Assignee: Kodak Alaris Inc (KPP (no. 2) Trustees Ltd.)
• Original Assignee: Eastman Kodak Company
• Inventors: Barton, Mark A.
• Granted Patent: US 7,020,328 B2
• Time in Patent Office: Days
• US Class Current: 382/173
• CPC Class Codes: G06V 10/143 Sensing or illuminating at different wavelengths

Abstract

A method for electronic color dropout utilizing spatial context to enhance accuracy comprises scanning a document to produce a color digital image. A thresholding apparatus is applied to the color digital image to produce a bitonal image. A region of the color digital image is selected corresponding to a feature of the bitonal image. An average color for the region is computed and compared to a predetermined dropout color. If the average color matches the predetermined dropout color, the region is erased from the bitonal image. Once it is determined that all features of the bitonal image have been examined, the process is complete. If all the features of the bitonal image have not been examined a region of the color digital image is selected corresponding to a feature of the bitonal image.

Claims

1. A method for electronic color dropout utilizing spatial context to enhance accuracy comprising:

a) scanning a document to produce a color digital image;

b) applying a thresholding process to said color digital image to produce a bitonal image;

c) select a region of said color digital image corresponding to a feature of said bitonal image;

d) compute an average color for said region;

e) compared said average color of said region to one or more predetermined dropout colors;

f) if said average color matches any of said predetermined dropout colors, erase said region from said bitonal image;

g) determine if all features of said bitonal image have been examined; and

h) if all features of said bitonal image have not been examined go to step c) if all features have been examined, the process is complete.

2. A method for electronic color dropout as in claim 1 wherein said thresholding process is an adaptive thresholding process.

3. A method for electronic color dropout as in claim 1 wherein said matching of said average color and said predetermined dropout color is within a predetermined tolerance.

4. A method for electronic color dropout utilizing spatial context to enhance accuracy comprising:

a) scanning a document to produce a color digital image;

b) applying a thresholding process to said color digital image to produce a bitonal image;

c) utilizing a contour following algorithm applied to said bitonal image to select a region of said color digital image corresponding to a feature of said bitonal image;

d) compute an average color for said region;

e) compare said average color of said region to one or more predetermined dropout colors;

f) if said average color matches any of said predetermined dropout colors, erase said region from said bitonal image;

g) determine if all regions of said bitonal image have been examined; and

h) if all features of said bitonal image have not been examined go to step c) and if all regions have been examined, the process is complete.

5. A method for electronic color dropout as in claim 4 wherein said thresholding process is an adaptive thresholding process.

6. A method for electronic color dropout as in claim 4 wherein said matching of said average color and said predetermined dropout color is within a predetermined tolerance.

7. A method for electronic color dropout utilizing spatial context to enhance accuracy comprising:

a) scanning a document to produce a color digital image;

b) applying a thresholding process to said color digital image to produce a bitonal image;

c) establishing an evaluation window surrounding a pixel of said bitonal image;

d) locating the connected pixels within said evaluation window of said bitonal image, to select a region of said color digital image, corresponding to the connected pixels of said bitonal image;

c) compute an average color for said region;

f) compared said average color of said region to one or more predetermined dropout colors;

g) if said average color matches any of said predetermined dropout colors, erase said region from said bitonal image; and

h) repeat steps c) through f), for each pixel in said color digital image.

8. A method for electronic color dropout as in claim 7 wherein said thresholding process is an adaptive thresholding process.

9. A method for electronic color dropout as in claim 7 wherein said matching of said average color and said predetermined dropout color is within a predetermined tolerance.

10. A method for electronic color dropout as in claim 7 said computation of an average color for said region is accomplished on a pixel by pixel basis.