Detecting printing effects

Patent metadata

Field	Value
Publication number	US-9704236-B2
Application number	US-201114345389-A
Country	US
Kind code	B2
Filing date	Sep 27, 2011
Priority date	Sep 27, 2011
Publication date	Jul 11, 2017
Grant date	Jul 11, 2017

Abstract

A method for detecting printing defects includes, with a physical computing system, creating a matched reference image from a digital image, colors of said matched reference image being closer matched to a scanned image of said digital image. The method further includes, with said physical computing system, determining a color difference between corresponding points on said matched reference image and said scanned image; and with said physical computing system, identifying points that have said color distance greater than a predefined threshold as potential defects.

First claim

What is claimed is:

1. A method for detecting printing defects, the method comprising:
with a physical computing system, creating a matched reference image from a digital image, colors of said matched reference image being closer matched to a scanned image of said digital image;
with said physical computing system, determining a color difference between corresponding points on said matched reference image and said scanned image; and
with said physical computing system, identifying points that have said color difference greater than a predefined threshold as potential defects,
wherein forming said matched reference image comprises:
with said physical computing system, for a range of color values for a color channel, determining an average color difference between corresponding pixels within said digital image and said scanned image for said range; and
with said physical computing system, adding said average color difference to color values of said digital image within said range to form said matched reference image.

2. The method of claim 1 , wherein if a number of pixels within said range of color values for said color channel is less than a predefined threshold, pixels within said range are omitted from said matched reference image.

3. The method of claim 1 , wherein determining a color difference between said matched reference image and said scanned image comprises forming a similarity image, said similarity image comprising a grayscale depiction of color differences between corresponding points between said scanned image and said matched reference image.

4. The method of claim 3 , wherein a similarity measure used to determine said color differences comprises features for being tolerant to at least one of: spatial misregistration and scanner color instabilities.

5. The method of claim 4 , further comprising, determining whether an anomalous region of said similarity image corresponds to a defect by comparing colors inside of a boundary of a region on said scanned image with colors outside of said boundary of said region, wherein said region on said scanned image corresponds to said anomalous region of said similarity image.

6. The method of claim 1 , wherein said physical computing system is integrated with a printing system used to print said digital image.

7. The method of claim 4 , wherein said printing system comprises a scanning system to create said scanned image.

8. The method of claim 1 , wherein said scanned image is created after a said digital image has been printed.

9. The method of claim 2 , wherein said matched reference image indicates omitted regions which were ignored during a matching process.

10. The method of claim 3 , wherein forming a similarity image comprises measuring a luminance similarity between two corresponding patches from said matched reference image and said scanned image.

11. The method of claim 9 , further comprising wherein forming a similarity image further comprises determining a context aware luminance.

12. The method of claim 5 , wherein said anomalous region is not a defect when the coloring in said inside of said regions is similar to a coloring of said outside of said regions.

13. A computing system comprising:
at least one processor;
a memory communicatively coupled to the at least one processor, the memory comprising computer executable code that, when executed by the at least one processor, causes the at least one processor to:
create a matched reference image from a digital image, colors of said matched reference image being closer matched to a scanned image of said digital image;
determine a color difference between corresponding points on said matched reference image and said scanned image, wherein to determine said color difference between said matched reference image and said scanned image, said computer executable code causes said processor to form a similarity image, said similarity image comprising a grayscale depiction of color differences between corresponding points between said scanned image and said matched reference image; and
identify points that have said color difference greater than a predefined threshold as potential defects.

14. The system of claim 13 , wherein to form said matched reference image, said computer executable code causes said processor to:
for a range of color values for a color channel, determine an average color difference between corresponding pixels within said digital image and said scanned image for said range; and
add said average color difference to color values of said digital image within said range to form said matched reference image.

15. The system of claim 14 , wherein if a number of pixels within said range of color values for said color channel is less than a predefined threshold, pixels within said range are omitted from said matched reference image.

16. The system of claim 13 , wherein:
darker shades of gray indicate more dissimilar regions; and
lighter shades of gray indicate more similar regions.

17. The system of claim 16 , wherein color differences are computed for each color channel and combined to form said similarity image.

18. The system of claim 17 , wherein said computer executable code further causes said processor to determine whether an anomalous region of said similarity image corresponds to a defect by comparing colors inside of a boundary of a region on said scanned image with colors outside of said boundary of said region, wherein said region on said scanned image corresponds to said anomalous region of said similarity image.

19. A computer program product for detecting image defects, said computer program product comprising:
a non-transitory computer readable storage medium having computer readable code embodied therewith, said computer readable program code comprising:
computer readable program code to define a number of color clusters of a digital image, said color clusters including a range of color values from color channels;
computer readable program code to create a matched reference image from said digital image, by adjusting said digital image coloring of pixels within said color duster, wherein said matched reference image includes omitted regions corresponding to regions of said digital image that have less than a predefined threshold number of pixels to use for matching;
computer readable program code to create a stability map for each color value range of each color channel;
computer readable program code to determine a color difference between corresponding points on said matched reference image and a scanned image;
computer readable program code to form a similarity image based on a determined color difference;
computer readable program code to identify points that have said color difference greater than a predefined threshold as potential defects, which potential defects are manifest in said similarity image as an anomalous region; and
computer readable program code to validate said anomalous region by comparing colors inside and outside of regions of said scanned image that correspond to said anomalous region of said matched reference image.

Assignees

• Kogan Hadas
• Hewlett Packard Development Co Lp

Inventors

• Kogan Hadas

Classifications

• H04N1/00047

  using an image not specifically designed for the purpose · CPC title

• H04N1/0005

  in service, i.e. during normal operation · CPC title

• H04N1/00045

  using a reference pattern designed for the purpose, e.g. a test chart · CPC title

• H04N1/00068

  Calculating or estimating · CPC title

• H04N1/00015

  Reproducing apparatus · CPC title