Defective pixel fixing

Patent metadata

Field	Value
Publication number	US-9779490-B2
Application number	US-201414571591-A
Country	US
Kind code	B2
Filing date	Dec 16, 2014
Priority date	Dec 18, 2013
Publication date	Oct 3, 2017
Grant date	Oct 3, 2017

Abstract

Methods of fixing defective pixels are described wherein a predicted value for a target pixel in a target color channel is determined based on the values of nearby pixels, wherein the target pixel value can be selectively replaced with the predicted value. The predicted value is determined by determining a candidate value for each of a plurality of directions using: (i) a gradient of pixel values in one color channel along the respective direction and (ii) a pixel value of a pixel in the target color channel which is aligned with the target pixel along the respective direction. Using gradients can provide better predicted values than averaging nearby pixel values since rates of change of pixel values are taken into account. The median of the candidate values may be used in order to reduce the impact of other defective pixels on the predicted value for the target pixel.

First claim

The invention claimed is:

1. A method of applying defective pixel fixing to image data, the image data comprising pixel values for a regular array of pixels in a plurality of colour channels, wherein the method comprises, for a target pixel in a target colour channel of said regular array of pixels, the target pixel having a given pixel value:
for each of a plurality of directions which intercept the target pixel:
(a) determining a gradient of pixel values in one of said colour channels along said direction; and
(b) determining a candidate value for the target pixel using: (i) the determined gradient, and (ii) a pixel value of a pixel in the target colour channel which is aligned with the target pixel along said direction;

determining the median of the plurality of candidate values determined for the respective plurality of directions, wherein the determined median is used as a predicted value for the target pixel;
determining whether the target pixel is characteristic of a defective pixel based on the similarity of the given value of the target pixel and the predicted value; and
selectively replacing the given value of the target pixel with the predicted value based on the determination as to whether the target pixel is characteristic of a defective pixel.

2. The method of claim 1 wherein said step of selectively replacing the given value of the target pixel with the predicted value comprises:
calculating the difference between the given value of the target pixel and the predicted value; and
comparing the calculated difference with a limit, wherein if the calculated difference is greater than the limit then the given value of the target pixel is replaced with the predicted value, and if the calculated difference is less than the limit then the given value of the target pixel is not replaced with the predicted value.

3. The method of claim 2 further comprising determining the limit based on a measure of the variation of the gradients determined for the respective plurality of directions.

4. The method of claim 3 further comprising calculating said measure of the variation of the gradients by calculating a median of the absolute differences between the absolute values of the gradients and the median of the absolute values of the gradients.

5. The method of claim 4 wherein the limit is determined according to the equation:
limit=threshold+MAD×weight,

where MAD is the median of the absolute differences between the absolute values of the gradients and the median of the absolute values of the gradients, and where threshold and weight are variables which can be adapted to thereby adapt the limit.

6. The method of claim 5 further comprising analysing the results of applying defective pixel fixing to image data of a first frame of a sequence of frames and adapting the values of the threshold and weight variables based on said analysis in order to adapt the limit for use in applying defective pixel fixing to values of a second frame of the sequence.

7. The method of claim 1 wherein, for each of said directions, said pixel in the target colour channel which is aligned with the target pixel is the closest pixel to the target pixel in the target colour channel along the respective direction.

8. The method of claim 1 wherein, for each of said directions, said pixel values in one of said colour channels along the direction for which a gradient is determined are pixel values of the closest pixels in said one of said colour channels to the target pixel along the respective direction.

9. The method of claim 1 wherein said plurality of directions comprises an odd number of directions.

10. The method of claim 1 wherein the regular array of pixels in a plurality of colour channels comprises a repeating pattern of 2×2 blocks, wherein each 2×2 block comprises an arrangement of pixels from the plurality of colour channels of said image data.

11. A defective pixel fixing unit configured to apply defective pixel fixing to image data, the image data comprising pixel values for a regular array of pixels in a plurality of colour channels, wherein one of the pixels is a target pixel in a target colour channel of said regular array of pixels, the target pixel having a given pixel value, wherein the defective pixel fixing unit comprises:
a candidate module configured to, for each of a plurality of directions which intercept the target pixel:
(a) determine a gradient of pixel values in one of said colour channels along said direction; and
(b) determine a candidate value for the target pixel using: (i) the determined gradient, and (ii) a pixel value of a pixel in the target colour channel which is aligned with the target pixel along said direction;

a prediction module configured to determine a predicted value for the target pixel by determining the median of the plurality of candidate values determined for the respective plurality of directions; and
a selection module configured to: (i) determine whether the target pixel is characteristic of a defective pixel based on the similarity of the given value of the target pixel and the predicted value, and (ii) selectively replace the given value of the target pixel with the predicted value based on the determination as to whether the target pixel is characteristic of a defective pixel.

12. The defective pixel fixing unit of claim 11 wherein the selection module is configured to:
calculate the difference between the given value of the target pixel and the predicted value; and
compare the calculated difference with a limit to determine whether to replace the given value of the target pixel with the predicted value.

13. The defective pixel fixing unit of claim 11 wherein the candidate module is configured to determine a candidate value for the target pixel by using the determined gradient to extrapolate from the pixel value of the pixel in the target colour channel which is aligned with the target pixel along said direction to the position of the target pixel.

14. The defective pixel fixing unit of claim 11 wherein the candidate module is further configured to scale the determined gradient as part of determining a candidate value for the target pixel.

15. The defective pixel fixing unit of claim 11 wherein the selection module is further configured to receive a defective pixel map indicating pixels which are likely to be defective, wherein the selection module is configured to selectively replace the given value of the target pixel with the predicted value is based on the defective pixel map.

16. The defective pixel fixing unit of claim 11 configured to receive a window of pixel values of the image data in parallel, said window of pixel values comprising a plurality of rows of pixel values from the image data, and wherein the defective pixel fixing unit further comprises a plurality of parallel processing modules configured to apply defective pixel fixing in parallel to respective target pixels from a respective plurality of rows of pixels in the window.

17. The defective pixel fixing unit of claim 16 wherein each of the processing modules is configured to use a sub-block of pixel values from the window for applying the defective pixel fixing for a respective target pixel, wherein each of the processing modules is configured to:
center the sub-block of pixel values on the target pixel; and
crop the sub-block if, when centered on the target pixel, the sub-block of pixel values extends outside of the window, such that the sub-block does not extend outside of the window, wherein the directions from the target pixel in which the sub-b

Assignees

• Imagination Tech Ltd

Inventors

• Bishop Michael

Classifications

• H04N25/134

  based on three different wavelength filter elements · CPC title

• H04N25/135

  based on four or more different wavelength filter elements · CPC title

• H04N25/683

  by defect estimation performed on the scene signal, e.g. real time or on the fly detection · CPC title

• H04N25/68

  applied to defects · CPC title

• H04N5/367

  Electricity · mapped topic