Detection of streaks in images

1. A method for identifying a streak in an image, the method comprising:

determining a cross-sectional pixel profile of an image;

determining a peak in the cross-sectional pixel profile of the image;

determining a strength of the peak in the cross-sectional pixel profile of the image;

determining a rotation angle of the image;

calculating a rotated image based on a rotation of the image by the rotation angle;

determining a cross-sectional pixel profile of the rotated image;

determining a peak in the cross-sectional pixel profile of the rotated image corresponding to the peak in the cross-sectional pixel profile of the image;

determining a strength of the peak in the cross-sectional pixel profile of the rotated image; and

identifying a streak in the image based on the strength of the peak in the cross-sectional pixel profile of the image and the strength of the peak in the cross-sectional pixel profile of the rotated image.

2. The method according to claim 1 , wherein identifying the streak includes identifying the streak as an image capturing streak in response to the strength of the peak in the cross-sectional pixel profile of the rotated image being smaller than the strength of the peak in the cross-sectional pixel profile of the image.

3. The method according to claim 1 , wherein identifying the streak includes identifying the streak as an image printing streak in response to the strength of the peak in the cross-sectional pixel profile of the rotated image being larger than the strength of the peak in the cross-sectional pixel profile of the image.

4. The method according to claim 1 , wherein identifying the streak includes applying a local spatial standard deviation on the cross-sectional pixel profile of the rotated image to calculate a new pixel profile and identifying the streak as a wiper streak in response the strength of the peak in the new pixel profile being larger than the strength of the peak in the cross-sectional pixel profile of the image.

5. The method according to claim 1 , further comprising capturing, by an image capture device, the image from a printout.

6. The method according to claim 5 , further comprising printing the printout by a printing press.

7. The method according to claim 1 , wherein the rotation angle is greater than 0° and less than 30°.

8. The method according to claim 1 , wherein determining the rotation angle includes analyzing at least two fiducial marks included in the image.

9. The method according to claim 1 , further comprising broadening streaks by applying a morphological filter to at least one of the image and the rotated image.

10. A printing press comprises an image capture device; wherein

the printing press includes at least one processor and a memory, and

the memory comprises executable instructions that when executed by the at least one processor cause the at least one processor to

generate, by the printing press, a printout;

cause the image capture device to capture an image of the printout;

determine a cross-sectional pixel profile of the image;

determine a peak in the cross-sectional pixel profile of the image;

determine a strength of the peak in the cross-sectional pixel profile of the image;

determine a rotation angle of the image;

calculate a rotated image by rotating the image according to the rotation angle;

determine a cross-sectional pixel profile of the rotated image;

determine a peak in the cross-sectional pixel profile of the rotated image corresponding to the peak in the cross-sectional pixel profile of the image;

determine a strength of the peak in the cross-sectional pixel profile of the rotated image; and

identify a streak in the image based on the strength of the peak in the cross-sectional pixel profile of the image and the strength of the peak in the cross-sectional pixel profile of the rotated image.

11. A non-transitory machine-readable storage medium encoded with instructions executable by a processor, the machine-readable storage medium comprising:

instructions to determine a rotation angle of an image;

instructions to calculate a rotated image by rotating the image according to the rotation angle;

instructions to determine a cross-sectional pixel profile of the rotated image;

instructions to determine a cross-sectional pixel profile of the image;

instructions to determine a peak in the cross-sectional pixel profile of the rotated image;

instructions to determine a strength of the peak in the cross-sectional pixel profile of the rotated image;

instructions to determine a peak in the cross-sectional pixel profile of the image, wherein the peak in the cross-sectional pixel profile of the image corresponds to the peak in the cross-sectional pixel profile of the rotated image;

instructions to determine a strength of the peak in the cross-sectional pixel profile of the image; and

instructions to identify a streak in the image based on the strength of the peak in the cross-sectional pixel profile of the image and the strength of the peak in the cross-sectional pixel profile of the rotated image.

12. The medium according to claim 11 , wherein the instructions to identify the streak include instructions to identify the streak as an image capturing streak in response to the strength of the peak in the cross-sectional pixel profile of the rotated image being smaller than the strength of the peak in the cross-sectional pixel profile of the image.

13. The medium according to claim 11 , wherein the instructions to identify the streak include instructions to identify the streak as an image printing streak in response to the strength of the peak in the cross-sectional pixel profile of the rotated image being larger than the strength of the peak in the cross-sectional pixel profile of the image.