Image processing based on automatic image-type detection

• US 9,055,251 B2
• Filed: 09/24/2013
• Issued: 06/09/2015
• Est. Priority Date: 09/24/2013
• Status: Active Grant

• Current Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Original Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Inventors: Palanivel, Dhevendra Alagan, Gopalakrishnan, Sainarayanan, Li, Xing, Cuciurean-Zapan, Clara
• Primary Examiner(s): Bayat, Ali
• Application Number:
  US14/034,690
• Publication Number:
  US 20150085320A1
• Time in Patent Office: 623 Days
• Field of Search: 382/232, 382/233, 382/244, 382/245, 382/246, 382/247, 382/248, 382/250, 382/251, 382/254, 382/270, 382/274
• US Class Current: 1/1
• CPC Class Codes: G06F 3/1208 resulting in improved quality of the output result, e.g. print layout, colours, workflows, print preview
• CPC Class Codes: G06F 3/126 Job scheduling, e.g. queuing, determine appropriate device
• CPC Class Codes: H04N 1/40062 Discrimination between different image types, e.g. two-tone, continuous tone

Abstract

Methods and systems divide an image into image blocks, determine a number of high-variation blocks within valid image blocks, determine a page variance measure by dividing the number of high-variation blocks by the total number of valid image blocks, and classify the image as comprising a halftone and/or text image if the page variance measure exceeds a halftone threshold. Such methods and systems classify the image as being a continuous tone image if the page variance measure does not exceed the halftone threshold and a text index is below a text index threshold, and classify the image as being a continuous tone and text image if the page variance measure does not exceed the halftone threshold and the text index is not below the text index threshold. Further, such methods and systems select an image processing mode based on such image classifications.

Claims

1. A method comprising:

receiving an image into a computerized device;

automatically determining an image-type of said image; and

automatically selecting an image processing mode based on said image-type of said image,

said automatically determining said image-type comprising;

automatically dividing said image into image blocks, using said computerized device;

automatically determining a block average by determining an average pixel value for each of said image blocks, using said computerized device;

automatically determining a squared error for each pixel within each of said image blocks by calculating a squared difference between a pixel value and said block average for each said pixel in each of said image blocks, using said computerized device;

automatically determining a sum of squared errors value for each of said image blocks, using said computerized device;

automatically removing ones of said image blocks determined to be a white background block to produce valid image blocks of said image, using said computerized device;

automatically determining a number of high-variation blocks within said valid image blocks by counting ones of said valid image blocks that have said sum of squared errors value exceeding a block variation threshold, using said computerized device;

automatically determining a page variance measure by dividing said number of high-variation blocks by a total number of said valid image blocks, using said computerized device;

automatically classifying said image as comprising a halftone and text image based on said page variance measure exceeding a halftone threshold, using said computerized device;

automatically determining a text index by counting a number of said valid image blocks that have said sum of squared errors value exceeding a text threshold, using said computerized device;

automatically classifying said image as comprising a continuous tone image based on said page variance measure not exceeding said halftone threshold and said text index being below a text index threshold, using said computerized device; and

automatically classifying said image as comprising a continuous tone and text image based on said page variance measure not exceeding said halftone threshold and said text index not being below said text index threshold, using said computerized device,

said automatically selecting said image processing mode comprising automatically selecting said image processing mode based on said image being one of said halftone and text image, said continuous tone and text image, and said continuous tone image.

2. The method according to claim 1, said determining said squared error for each pixel comprising multiplying said difference between a pixel value and said block average by itself.

3. The method according to claim 1, said dividing said image into image blocks comprising assigning rectangular geometric regions to said image.

4. The method according to claim 1, further comprising filtering all colors except gray from said image before performing said dividing said image into image blocks.

5. The method according to claim 1, said pixel value comprising a darkness measure for each said pixel.

6. The method according to claim 1, said pixel value comprising a darkness measure for each said pixel.

7. A method comprising:

receiving an image into a computerized device;

automatically dividing said image into image blocks, using said computerized device;

automatically determining a block average by determining an average pixel value for each of said image blocks, using said computerized device;

automatically determining a squared error for each pixel within each of said image blocks by calculating a squared difference between a pixel value and said block average for each said pixel in each of said image blocks, using said computerized device;

automatically determining a sum of squared errors value for each of said image blocks by summing, for each of said image blocks, said squared error for all pixels within each of said image blocks, using said computerized device;

automatically determining whether each of said image blocks comprises a white background block of said image by determining if said block average for each of said image blocks exceeds a white background threshold, using said computerized device;

automatically removing ones of said image blocks determined to be said white background block to produce valid image blocks of said image, using said computerized device;

automatically determining a number of high-variation blocks within said valid image blocks by counting ones of said valid image blocks that have said sum of squared errors value exceeding a block variation threshold, using said computerized device;

automatically determining a page variance measure by dividing said number of high-variation blocks by a total number of said valid image blocks, using said computerized device;

automatically classifying said image as comprising a halftone and text image based on said page variance measure exceeding a halftone threshold, using said computerized device;

automatically determining a text index by counting a number of said valid image blocks that have said sum of squared errors value exceeding a text threshold, using said computerized device;

automatically classifying said image as comprising a continuous tone image based on said page variance measure not exceeding said halftone threshold and said text index being below a text index threshold, using said computerized device;

automatically classifying said image as comprising a continuous tone and text image based on said page variance measure not exceeding said halftone threshold and said text index not being below said text index threshold, using said computerized device; and

automatically selecting an image processing mode based on said image being one of said halftone and text image, said continuous tone and text image, and said continuous tone image.

8. The method according to claim 7, said determining said block average, said determining said squared error, said determining said sum of squared errors, and said determining whether each of said image blocks comprises a white background block, being performed in parallel processing that individually evaluates all of said image blocks simultaneously.

9. The method according to claim 7, said determining said squared error for each pixel comprising multiplying said difference between a pixel value and said block average by itself.

10. The method according to claim 7, said dividing said image into image blocks comprising assigning rectangular geometric regions to said image.

11. The method according to claim 7, further comprising filtering all colors except gray from said image before performing said dividing said image into image blocks.

12. The method according to claim 7, said pixel value comprising a darkness measure for each said pixel.

13. A method comprising:

receiving an image into a computerized device;

automatically dividing said image into image blocks, using said computerized device;

automatically determining a block average by determining an average pixel value for each of said image blocks, using said computerized device;

automatically determining a squared error for each pixel within each of said image blocks by calculating a squared difference between a pixel value and said block average for each said pixel in each of said image blocks, using said computerized device;

automatically determining a sum of squared errors value for each of said image blocks by summing, for each of said image blocks, said squared error for all pixels within each of said image blocks, using said computerized device;

automatically determining whether each of said image blocks comprises a white background block of said image by determining if said block average for each of said image blocks exceeds a white background threshold, using said computerized device;

automatically removing ones of said image blocks determined to be said white background block to produce valid image blocks of said image, using said computerized device;

automatically determining a number of high-variation blocks within said valid image blocks by counting ones of said valid image blocks that have said sum of squared errors value exceeding a block variation threshold, using said computerized device;

automatically determining a page variance measure by dividing said number of high-variation blocks by a total number of said valid image blocks, using said computerized device;

automatically classifying said image as comprising a halftone and text image based on said page variance measure exceeding a halftone threshold, using said computerized device;

automatically determining a text index by counting a number of said valid image blocks that have said sum of squared errors value exceeding a text threshold, using said computerized device;

automatically classifying said image as comprising a continuous tone image based on said page variance measure not exceeding said halftone threshold and said text index being below a text index threshold, using said computerized device;

automatically classifying said image as comprising a continuous tone and text image based on said page variance measure not exceeding said halftone threshold and said text index not being below said text index threshold, using said computerized device;

automatically selecting a tile-based parallel error diffusion image processing mode based on said image being said halftone and text image;

automatically selecting a sequential error diffusion image processing mode based on said image being said continuous tone and text image; and

automatically selecting a clustered dot halftone image processing mode based on said image being said continuous tone image.

14. The method according to claim 13, said determining said block average, said determining said squared error, said determining said sum of squared errors, and said determining whether each of said image blocks comprises a white background block, being performed in parallel processing that individually evaluates all of said image blocks simultaneously.

15. The method according to claim 13, said determining said squared error for each pixel comprising multiplying said difference between a pixel value and said block average by itself.

16. The method according to claim 13, said dividing said image into image blocks comprising assigning rectangular geometric regions to said image.

17. The method according to claim 13, further comprising filtering all colors except gray from said image before performing said dividing said image into image blocks.

18. The method according to claim 13, said pixel value comprising a darkness measure for each said pixel.

19. An image processing device comprising:

an input and output device receiving an image; and

a processor operatively connected to said input and output device, said processor;

automatically dividing said image into image blocks,

automatically determining a block average by determining an average pixel value for each of said image blocks,

automatically determining a squared error for each pixel within each of said image blocks by calculating a squared difference between a pixel value and said block average for each said pixel in each of said image blocks,

automatically determining a sum of squared errors value for each of said image blocks,

automatically removing ones of said image blocks determined to a white background block to produce valid image blocks of said image,

automatically determining a number of high-variation blocks within said valid image blocks by counting ones of said valid image blocks that have said sum of squared errors value exceeding a block variation threshold,

automatically determining a page variance measure by dividing said number of high-variation blocks by a total number of said valid image blocks,

automatically classifying said image as comprising a halftone and text image based on said page variance measure exceeding a halftone threshold,

automatically determining a text index by counting a number of said valid image blocks that have said sum of squared errors value exceeding a text threshold,

automatically classifying said image as comprising a continuous tone image based on said page variance measure not exceeding said halftone threshold and said text index being below a text index threshold,

automatically classifying said image as comprising a continuous tone and text image based on said page variance measure not exceeding said halftone threshold and said text index not being below said text index threshold, and

automatically selecting an image processing mode based on said image being one of said halftone and text image, said continuous tone and text image, and said continuous tone image.

20. The image processing device according to claim 19, said determining said block average, said determining said squared error, said determining said sum of squared errors, and said determining whether each of said image blocks comprises a white background block, being performed in parallel processing that individually evaluates all of said image blocks simultaneously.

21. The image processing device according to claim 19, said determining said squared error for each pixel comprising multiplying said difference between a pixel value and said block average by itself.

22. The image processing device according to claim 19, said dividing said image into image blocks comprising assigning rectangular geometric regions to said image.

23. The image processing device according to claim 19, said processor filtering all colors except gray from said image before performing said dividing said image into image blocks.