System and method for flicker detection in digital imaging
View Patent ↗A system and method for detecting flicker in a digital image. The system and method detects flicker by performing a frame-to-frame comparison of image data. The frame-to-frame comparison of image data is based on dividing each image into multiple columns, and performing a weighted sum of the pixel luminance in each column using a pattern of alternating sign weights. The weighted sum of luminance for each column in a first frame is compared to the weighted sum of luminance for that column in a second frame to determine if there is a simultaneous luminance variation in those columns. For example, if the difference between weighted sums of luminance between frames exceeds a threshold value in a predetermined number of columns then there is a simultaneous luminance variation in those columns, and a horizontal pattern of flicker has been detected in the image data.
1. A method of detecting flicker in a digital camera, the method comprising the steps of:
receiving image data, the image data including a plurality of pixels arranged in a plurality of rows;
performing a first weighted sum of pixel luminance in a first plurality of columns defined in the image data using a first pattern of alternating sign weights with a summation mechanism, where the performing of the first weighted sum comprises multiplying each of a first plurality of pixels in the first plurality of columns with either a positive sign weight from the first pattern of alternating sign weights or a negative sign weight from the first pattern of alternating sign weights depending on the pixel's row position in the first pattern of alternating sign weights;
performing a second weighted sum of pixel luminance in a second plurality of columns defined in the image data using a second pattern of alternating sign weights with the summation mechanism, where the performing of the second weighted sum comprises multiplying each of a second plurality of pixels in the second plurality of columns with either a positive sign weight from the second pattern of alternating sign weights or a negative sign weight from the second pattern of alternating sign weights depending on the pixel's row position in the second pattern of alternating sign weight; and
comparing the first weighted sum of pixel luminance and the second weighted sum of pixel luminance with a comparator to determine if flicker is present in the image data.
2. The method of claim 1 wherein the first pattern of alternating sign weights and the second pattern of alternating sign weights comprise identical patterns.
3. The method of claim 1 wherein the first plurality of columns and the second plurality of columns comprise identical plurality of columns.
4. The method of claim 1 wherein wherein a positive sign weight in the first pattern of alternating sign weight comprises a +1 and a negative sign weight in the first pattern of alternating sign weight comprises a −1.
5. The method of claim 1 wherein the step of performing a first weighted sum of pixel luminance in a first plurality of columns defined in the image data using a first pattern of alternating sign weights comprises performing the first weighted sum on a first digital image and wherein the step of performing a second weighted sum of pixel luminance in a second plurality of columns defined in the image data using a second pattern of alternating sign weights comprises performing the second weighted sum on a second digital image.
6. The method of claim 1 wherein the step of comparing the first weighted sum of pixel luminance and the second weighted sum of pixel luminance to determine if flicker is present in the image data comprises determining a difference between the first weighted sum and the second weighted sum and comparing the difference to a threshold value.
7. The method of claim 6 wherein the threshold value is determined at least in part by an overall luminance in the image data.
8. The method of claim 1 further comprising the steps of:
performing a third weighted sum of pixel luminance in a third plurality of columns defined in the image data using a third pattern of alternating sign weights with the summation mechanism, wherein the third pattern of alternating sign weights is out of phase with the first pattern of alternating sign weights, where the performing of the third weighted sum comprises multiplying each of a third plurality of pixels in the third plurality of columns with either a positive sign weight from the third pattern of alternating sign weights or a negative sign weight from the third pattern of alternating sign weights depending on the pixel's row position in the third pattern of alternating sign weights;
performing a fourth weighted sum of pixel luminance in a fourth plurality of columns defined in the image data using a fourth pattern of alternating sign weights with the summation mechanism, wherein the fourth pattern of alternating sign weights is out of phase with the second pattern of alternating sign weights, where the performing of the fourth weighted sum comprises multiplying each of a fourth plurality of pixels in the fourth plurality of columns with either a positive sign weight from the fourth pattern of alternating sign weights or a negative sign weight from the fourth pattern of alternating sign weights depending on the pixel's row position in the fourth pattern of alternating sign weights; and
comparing the third weighted sum of pixel luminance and the fourth weighted sum of pixel luminance with the comparator to determine if flicker is present in the image data.
9. The method of claim 8 wherein the third pattern of alternating sign weights is out of phase with the first pattern of alternating sign weights by substantially 90 degrees and wherein the fourth pattern of alternating sign weights is out of phase with the second pattern of alternating sign weights by substantially 90 degrees.
10. The method of claim 8 wherein the steps of comparing the first weighted sum of pixel luminance and the second weighted sum of pixel luminance to determine if flicker is present in the image data and comparing the third weighted sum of pixel luminance and the fourth weighted sum of pixel luminance to determine if flicker is present in the image data comprises determining a first difference between the first weighted sum and the second weighted sum and comparing the first difference to a first threshold value and determining a second difference between the third weighted sum and the fourth weighted sum and comparing the second difference to a second threshold value.
11. The method of claim 10 wherein flicker is detected if either the first difference exceeds the first threshold value or the second difference exceeds the second threshold value.
12. The method of claim 10 wherein the first threshold value and the second threshold value comprise identical threshold values.
13. The method of claim 1 wherein the first plurality of columns and second plurality of columns each comprise four columns.
14. The method of claim 1 wherein the step of comparing the first weighted sum of pixel luminance and the second weighted sum of pixel luminance to determine if flicker is present comprises determining a difference between the first weighted sum and the second weighted sum for each of the first plurality of columns.
15. The method of claim 14 wherein flicker is determined to be present only if the difference between the first weighted sum and the second weighted sum for each of the first plurality of columns has the same sign.
16. A method of detecting flicker in a digital camera, the method comprising the steps of:
receiving a first digital image, the first digital image including a plurality of pixels arranged in a plurality of rows;
defining a first plurality of columns in the first digital image;
performing a first weighted sum of pixel luminance in each of the first plurality of columns in the first digital image using a first pattern of alternating sign weights with a summation mechanism, where the performing of the first weighted sum comprises multiplying each of a first plurality of pixels in the first plurality of columns with either a positive sign weight from the first pattern of alternating sign weights or a negative sign weight from the first pattern of alternating sign weights depending on the pixel's row position in the first pattern of alternating sign weights;
receiving a second digital image, the second digital image including a plurality of pixels arranged in a plurality of rows;
defining a second plurality of columns in the second digital image;
performing a second weighted sum of pixel luminance in each of the second plurality of columns in the second digital image using a second pattern of alternating sign weights with the summation mechanism, where the performing of the second weighted sum comprises multiplying each of a second plurality of pixels in the second plurality of columns with either a positive sign weight from the second pattern of alternating sign weights or a negative sign weight from the second pattern of alternating sign weights depending on the pixel's row position in the second pattern of alternating sign weights;
determining a difference between the first weighted sum of pixel luminance in each of the first plurality of columns and the second weighted sum of pixel luminance in each of the second plurality of columns with a comparator to determine if flicker is present.
17. The method of claim 16 wherein the step of determining a difference between the first weighted sum of pixel luminance in each of the first plurality of columns and the second weighted sum of pixel luminance in each of the second plurality of columns to determine if flicker is present determines that flicker is present only if the difference between each of the first plurality of columns and second plurality of columns has the same sign.
18. A method of detecting flicker in a digital camera, the method comprising the steps:
receiving a first digital image, the first digital image including a plurality of pixels arranged in a plurality of rows;
defining a first plurality of columns in the first digital image;
performing a first weighted sum of pixel luminance in each of the first plurality of columns in the first digital image using a first pattern of alternating sign weights with a summation mechanism, where the performing of the first weighted sum comprises multiplying each of a first plurality of pixels in the first plurality of columns with either a positive sign weight from the first pattern of alternating sign weights or a negative sign weight from the first pattern of alternating sign weights depending on the pixel's row position in the first pattern of alternating sign weights;
receiving a second digital image, the second digital image including a plurality of pixels arranged in a plurality of rows;
defining a second plurality of columns in the second digital image;
performing a second weighted sum of pixel luminance in each of the second plurality of columns in the second digital image using a second pattern of alternating sign weights with the summation mechanism, where the performing of the second weighted sum comprises multiplying each of a second plurality of pixels in the second plurality of columns with either a positive sign weight from the second pattern of alternating sign weights or a negative sign weight from the second pattern of alternating sign weights depending on the pixel's row position in the second pattern of alternating sign weights;
determining a difference between the first weighted sum of pixel luminance in each of the first plurality of columns and the second weighted sum of pixel luminance in each of the second plurality of columns with a comparator;
defining a third plurality of columns in the first digital image;
performing a third weighted sum of pixel luminance in each of the third plurality of columns in the first digital image using a third pattern of alternating sign weights with the summation mechanism, the third pattern of alternating sign weights substantially 90 degrees out of phase with the first pattern of alternating sign weights, where the performing of the third weighted sum comprises multiplying each of a third plurality of pixels in the third plurality of columns with either a positive sign weight from the third pattern of alternating sign weights or a negative sign weight from the third pattern of alternating sign weights depending on the pixel's row position in the third pattern of alternating sign weights;
defining a fourth plurality of columns in the second digital image;
performing a fourth weighted sum of pixel luminance in each of the fourth plurality of columns in the second digital image using a fourth pattern of alternating sign weights with the summation mechanism, the fourth pattern of alternating sign weights substantially 90 degrees out of phase with the second pattern of alternating sign weights, where the performing of the fourth weighted sum comprises multiplying each of a fourth plurality of pixels in the fourth plurality of columns with either a positive sign weight from the fourth pattern of alternating sign weights or a negative sign weight from the fourth pattern of alternating sign weights depending on the pixel's row position in the fourth pattern of alternating sign weight;
determining a difference between the third weighted sum of pixel luminance in each of the third plurality of columns and the fourth weighted sum of pixel luminance in each of the fourth plurality of columns with the comparator; and
determining that flicker is present if either the difference between the first weighted sum of pixel luminance in each of the first plurality of columns and the second weighted sum of pixel luminance in each of the second plurality of columns is greater than a first threshold value or if the difference between the third weighted sum of pixel luminance in each of the third plurality of columns and the fourth weighted sum of pixel luminance in each of the fourth plurality of columns is greater than a second threshold value.
19. A processor tangibly embodying a program of instructions to perform method steps for detecting flicker in a digital camera, the method comprising the steps of:
receiving image data, the image data including a plurality of pixels arranged in a plurality of rows;
performing a first weighted sum of pixel luminance in a first plurality of columns defined in the image data using a first pattern of alternating sign weights with a summation mechanism, where the performing of the first weighted sum comprises multiplying each of a first plurality of pixels in the first plurality of columns with either a positive sign weight from the first pattern of alternating sign weights or a negative sign weight from the first pattern of alternating sign weights depending on the pixel's row position in the first pattern of alternating sign weights;
performing a second weighted sum of pixel luminance in a second plurality of columns defined in the image data using a second pattern of alternating sign weights with the summation mechanism, where the performing of the second weighted sum comprises multiplying each of a second plurality of pixels in the second plurality of columns with either a positive sign weight from the second pattern of alternating sign weights or a negative sign weight from the second pattern of alternating sign weights depending on the pixel's row position in the second pattern of alternating sign weights; and
comparing the first weighted sum of pixel luminance and the second weighted sum of pixel luminance with a comparator to determine if flicker is present in the image data.
20. The processor of claim 19 wherein the first pattern of alternating sign weights and the second pattern of alternating sign weights comprise identical patterns.
21. The processor of claim 19 wherein the first plurality of columns and the second plurality of columns comprise identical plurality of columns.
22. The processor of claim 19 wherein the step of performing a first weighted sum of pixel luminance in a first plurality of columns defined in the image data using a first pattern of alternating sign weights comprises performing the first weighted sum on a first digital image and wherein the step of performing a second weighted sum of pixel luminance in a second plurality of columns defined in the image data using a second pattern of alternating sign weights comprises performing the second weighted sum on a second digital image.
23. The processor of claim 19 wherein a positive sign weight in the first pattern of alternating sign weights comprises a +1 and a negative sign weight in the first pattern of alternating sign weights comprises a −1.
24. The processor of claim 19 wherein the step of comparing the first weighted sum of pixel luminance and the second weighted sum of pixel luminance to determine if flicker is present in the image data comprises determining a difference between the first weighted sum and the second weighted sum and comparing the difference to a threshold value.
25. The processor of claim 24 wherein the threshold value is determined at least in part by an overall luminance in the image data.
26. The processor of claim 19 further comprising the steps of:
performing a third weighted sum of pixel luminance in a third plurality of columns defined in the image data using a third pattern of alternating sign weights with the summation mechanism, wherein the third pattern of alternating sign weights is out of phase with the first pattern of alternating sign weights, where the performing of the third weighted sum comprises multiplying each of a third plurality of pixels in the third plurality of columns with either a positive sign weight from the third pattern of alternating sign weights or a negative sign weight from the third pattern of alternating sign weights depending on the pixel's row position in the third pattern of alternating sign weights;
performing a fourth weighted sum of pixel luminance in a fourth plurality of columns defined in the image data using a fourth pattern of alternating sign weights with the summation mechanism, wherein the fourth pattern of alternating sign weights is out of phase with the second pattern of alternating sign weights, where the performing of the fourth weighted sum comprises multiplying each of a fourth plurality of pixels in the fourth plurality of columns with either a positive sign weight from the fourth pattern of alternating sign weights or a negative sign weight form the fourth pattern of alternating sign weights depending on the pixel's row position in the fourth pattern of alternating sign weights; and
comparing the third weighted sum of pixel luminance and the fourth weighted sum of pixel luminance with a comparator to determine if flicker is present in the image data.
27. The processor of claim 26 wherein the third pattern of alternating sign weights is out of phase with the first pattern of alternating sign weights by substantially 90 degrees and wherein the fourth pattern of alternating sign weights is out of phase with the second pattern of alternating sign weights by substantially 90 degrees.
28. The processor of claim 26 wherein the steps of comparing the first weighted sum of pixel luminance and the second weighted sum of pixel luminance to determine if flicker is present in the image data and comparing the third weighted sum of pixel luminance and the fourth weighted sum of pixel luminance to determine if flicker is present in the image data comprises determining a first difference between the first weighted sum and the second weighted sum and comparing the first difference to a first threshold value and determining a second difference between the third weighted sum and the fourth weighted sum and comparing the second difference to a second threshold value.
29. The processor of claim 28 wherein the first threshold value and the second threshold value comprise identical threshold values.
30. The processor of claim 28 wherein flicker is detected if either the first difference exceeds the first threshold value or the second difference exceeds the second threshold value.
31. The processor of claim 19 wherein the first plurality of columns and second plurality of columns each comprise four columns.
32. The processor of claim 19 wherein the step of comparing the first weighted sum of pixel luminance and the second weighted sum of pixel luminance to determine if flicker is present comprises determining a difference between the first weighted sum and the second weighted sum for each of the first plurality of columns.
33. The processor of claim 32 wherein flicker is determined to be present only if the difference between the first weighted sum and the second weighted sum for each of the first plurality of columns has the same sign.