Image defect detection

1. A computer implemented method of detecting a defect in a printed image, the method comprising:

receiving a target image comprising digital image data representing a scan of the printed image;

receiving a reference image comprising digital image data representing a reference of the printed image;

calculating a structural dissimilarity measure, D, associated with a target pixel located in the target image and a reference pixel located in the reference image; and,

determining on the basis of the structural dissimilarity measure whether a defect is present at the target pixel,

wherein the structural dissimilarity measure is calculated using a structural measure, s, and a contrast measure, c; the structural measure calculated using a spatial cross-correlation associated with a target region, {right arrow over (x)}, containing the target pixel and a reference region, {right arrow over (y)}, containing the reference pixel, and the contrast measure calculated using a standard deviation associated with the target region, and a standard deviation associated with the reference region, and

wherein the structural dissimilarity measure is independent of a luminance measure.

2. A method according to claim 1 , wherein the spatial cross-correlation is a spatial intensity cross-correlation associated with the target region and the reference region.

3. A method according to claim 1 , wherein the structural dissimilarity measure is calculated according to:

D|[ 1 −c ( {right arrow over (x)},{right arrow over (y)} )]× s ( {right arrow over (x)},{right arrow over (y)} )|.

4. A method according to claim 3 wherein the structural measure is calculated as:

s

⁡

(

x

,

y

)

=

2

⁢

σ

xy

+

C

3

σ

x

⁢

σ

y

+

C

3

,

where C 3 is a constant and:

σ

x

=

(

1

N

-

1

⁢

∑

i

=

1

N

⁢

(

x

i

-

μ

x

)

2

)

1

2

and

σ

xy

=

(

1

N

-

1

⁢

∑

i

=

1

N

⁢

(

x

i

-

μ

x

)

⁢

(

y

i

-

μ

y

)

)

1

2

.

5. A method according to claim 4 , wherein the contrast measure is calculated as:

c

(

x

->

,

y

->

)

=

2

⁢

σ

x

⁢

σ

y

+

C

2

σ

x

2

+

σ

y

2

+

C

2

,

(

6

)

where C 2 is a constant.

6. A method according to claim 5 , wherein the method further comprises registration of the target region with the reference region.

7. A method according to claim 6 , wherein registration is performed on the basis of a structural similarity measure, S, between the target region and the reference region.

8. A method according to claim 7 , wherein registration comprises calculating the structural similarity measure, S, associated with the target region and each of a plurality of reference regions within a window of predetermined size in the reference image, and selecting the target region as a best matched reference region on the basis of the calculated structural similarity measure.

9. A method according to claim 7 , wherein the structural similarity measure is calculated according to:

S ( {right arrow over (x)},{right arrow over (y)} )= f ( l ( {right arrow over (x)},{right arrow over (y)} ), c ( {right arrow over (x)},{right arrow over (y)} ), s ( {right arrow over (x)},{right arrow over (y)} ))

where/is a luminance measure calculated according to:

l

(

x

->

,

y

->

)

=

2

⁢

μ

x

⁢

μ

y

+

C

1

μ

x

2

+

μ

y

2

+

C

1

,

where C 1 is a constant.

10. A method according to claim 9 , wherein the method further comprises generating a defect map on the basis of a plurality of structural dissimilarity measures calculated for a plurality of pixels in the target image.

11. A method according to claim 10 , wherein the defect map is generated by applying a predetermined threshold to the plurality of structural dissimilarity measures to produce a binary map.

12. A method according to claim 11 , wherein the method further comprises filtering the defect map to remove visually imperceptible defects and/or systematic errors.

13. A method according to claim 11 , wherein the method further comprises performing a projection in the horizontal and vertical directions of the defect map and triggering a defect warning upon a determination that more than a predetermined pixels are ‘on’ in a row or column.

14. A defect detection apparatus to detect a defect in a printed image, the apparatus to:

receive a target image comprising digital image data representing a scan of the printed image;

receive a reference image comprising digital image data representing a reference of the printed image;

calculate a structural dissimilarity measure, D, associated with a target pixel located in the target image and a reference pixel located in the reference image; and,

determine on the basis of the structural dissimilarity measure whether a defect is present at the target pixel,

wherein the structural dissimilarity measure is calculated using a structural measure, s, and a contrast measure, c; the structural measure is calculated using a spatial cross-correlation associated with a target region, {right arrow over (x)}, containing the target pixel and a reference region, {right arrow over (y)}, containing the reference pixel, and the contrast measure is calculated using a standard deviation associated with the target region, and a standard deviation associated with the reference region and

wherein the structural dissimilarity measure is independent of a luminance measure.

15. A system for detecting to detect a defect in a printed image, the system comprising:

a defect detection apparatus to:

receive a target image comprising digital image data representing a scan of the printed image;

receive a reference image comprising digital image data representing a reference of the printed image;

calculate a structural dissimilarity measure, D, associated with a target pixel located in the target image and a reference pixel located in the reference image; and,

determine on the basis of the structural dissimilarity measure whether a defect is present at the target pixel,

wherein the structural dissimilarity measure is calculated using a structural measure, s, and a contrast measure, c; the structural measure calculated using a spatial cross-correlation associated with a target region, {right arrow over (x)}, containing the target pixel and a reference region, {right arrow over (y)}, containing the reference pixel, and the contrast measure calculated using a standard deviation associated with the target region, and a standard deviation associated with a reference region and

wherein the structural dissimilarity measure is independent of a luminance measure; and,

a scanning apparatus to scan the printed image to a target image and to send the target image to the defect detection apparatus.

16. A non-transitory computer readable medium containing instructions stored thereon for enacting a method of detecting a defect in a printed image, the instructions to cause a computer to:

receive a target image comprising digital image data representing a scan of the printed image;

receive a reference image comprising digital image data representing a reference of the printed image;

calculate a structural dissimilarity measure, D, associated with a target pixel located in the target image and a reference pixel located in the reference image; and,

determine on the basis of the structural dissimilarity measure whether a defect is present at the target pixel,

wherein the structural dissimilarity measure is calculated using a structural measure, s, and a contrast measure, c; the structural measure is calculated using a spatial cross-correlation associated with the target region, {right arrow over (x)}, containing the target pixel and a reference region, {right arrow over (y)}, containing the reference pixel, and the contrast measure is calculated using a standard deviation associated with the target region, and a standard deviation associated with the reference region and

wherein the structural dissimilarity measure is independent of a luminance measure.

17. A defect detection apparatus according to claim 14 , wherein the structural dissimilarity measure is calculated according to:

D=|[ 1 −c ( {right arrow over (x)},{right arrow over (y)} )]× s ( {right arrow over (x)},{right arrow over (y)} )|.

18. A system according to claim 15 , wherein the structural dissimilarity measure is calculated according to:

D=|[ 1 −c ( {right arrow over (x)},{right arrow over (y)} )]× s ( {right arrow over (x)},{right arrow over (y)} )|.

19. A non-transitory computer readable medium according to claim 16 , wherein the structural dissimilarity measure is calculated according to:

D=|[ 1 −c ( {right arrow over (x)},{right arrow over (y)} )]× s ( {right arrow over (x)},{right arrow over (y)} )|.