METHOD AND SYSTEM FOR AUTOMATIC WINDOW CLASSIFICATION IN A DIGITAL REPROGRAPHIC SYSTEM

• US 20080049238A1
• Filed: 08/28/2006
• Published: 02/28/2008
• Est. Priority Date: 08/28/2006
• Status: Abandoned Application

• Current Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Original Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Inventors: Tse, Francis Kapo, Li, Xing, Nagarajan, Ramesh, McCandlish, Peter D.
• Application Number:
  US11/467,584
• Publication Number:
  US 20080049238A1
• Time in Patent Office: Days
• US Class Current: 358/1.9
• CPC Class Codes: G06V 10/28 Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns

Abstract

A method and system for processing an image in a digital reprographic system that performs a first segmentation process upon a contone digital image to identify one of a plurality of image data types for each pixel in the contone digital image, generating tags, binarizing the contone digital image and storing the binary image as an intermediate format. This binary image is later reconstructed to generate a reconstructed contone digital image; and performing a second segmentation process upon the reconstructed contone digital image to identify areas of the reconstructed contone digital image with common image data types.

Claims

1. A method for processing an image in a digital reprographic system, comprising:

(a) performing a first segmentation process upon a contone digital image to identify one of a plurality of image data types for each pixel in the contone digital image;

(b) generating tags, the tags coding the identity of the image data type of each pixel of the contone digital image;

(c) binarizing the contone digital image;

(d) reconstructing the binarized digital image to generate a reconstructed contone digital image; and

(e) performing a second segmentation process upon the reconstructed contone digital image to identify areas of the reconstructed contone digital image with common image data types.

2. The method as claimed in claim 1, further comprising:

(f) processing the reconstructed contone digital image based upon the tags.

3. The method as claimed in claim 1, wherein the plurality of image data types includes text and line art.

4. The method as claimed in claim 1, wherein the plurality of image data types includes contone images.

5. The method as claimed in claim 1, wherein the plurality of image data types includes halftones.

6. The method as claimed in claim 1, wherein the binarization of the contone image utilizes a halftone matrix.

7. The method as claimed in claim 1, wherein the binarization of the contone image utilizes error diffusion.

8. The method as claimed in claim 1, further comprising:

(f) converting a color contone digital image to a YCC color space before segmenting the color contone digital image.

9. The method as claimed in claim 1, further comprising:

(f) converting a color contone digital image to a L*a*b* color space before segmenting the color contone digital image.

10. The method as claimed in claim 1, wherein reconstruction of the binarized image maintains sharpness of edges based on the tag data stream.

11. The method as claimed in claim 1, wherein reconstruction of the binarized image utilizes a pattern identifier which is based on the pattern of pixels in a neighborhood around each pixel to choose one of a plurality of contone values.

12. A system for image processing in a reprographic system comprising:

a source of contone digital images;

a first segmentation module, operatively connected to said source of digital page images, to generate tags, the tags encoding the identity of the image data type of each pixel of the contone digital image;

a binary encoding module, operatively connected to said first segmentation module, said binary encoding module transforming the contone digital image into a high resolution binary image;

a decoding module, operatively connected to said binary encoding module, to convert the high resolution binary image into a second contone image; and

a second segmentation module to identify areas of the second contone image with common image data types.

13. The system as claimed claim 12, further comprising an image processing module, operatively connected to said second segmentation module, to image process the second contone image based upon the tags.

14. The system as claimed claim 12, wherein said binary encoding module utilizes a halftone matrix.

15. The system as claimed claim 12, wherein said binary encoding module utilizes error diffusion.

16. The system as claimed claim 12, wherein the plurality of image data types includes text and line art.

17. The system as claimed claim 12, wherein the plurality of image data types includes contone images.

18. The system as claimed claim 12, wherein the plurality of image data types includes halftones.

19. The system as claimed in claim 12, wherein the decoding module includes a two dimensional digital filter to reconstruct the second contone image.

20. The system as claimed in claim 12, wherein the decoding module includes a pattern generator to generate a pattern identifier which is used to choose one of a plurality of contone values.