Method and device for estimating noise level of dark reference rows of an image sensor

1. A method, comprising:

sequentially computing, for each of a plurality of pairs of sets of dark reference pixels of an image sensor, absolute differences in dark signal levels of the plurality of pairs of sets of dark reference pixels;

accumulating the absolute differences in dark signal levels; and

estimating a noise level of the dark reference pixels of the image sensor based on the accumulated absolute differences.

2. The method of claim 1 wherein a first pair of sets of dark reference pixels of the plurality of pairs of sets of dark reference pixels consists of two dark reference pixels.

3. The method of claim 2 wherein the two dark reference pixels are adjacent.

4. The method of claim 3 , comprising:

storing an absolute value of a darkness level of a first pixel of the two dark reference pixels;

computing an absolute difference between a darkness level of a second pixel of the two dark reference pixels and the stored absolute value of the darkness level of the first pixel of the two dark reference pixels; and

initializing the accumulated absolute differences to the computed absolute difference between the darkness level of the second pixel of the two dark reference pixels and the stored darkness level of the first pixel of the two dark reference pixels.

5. The method of claim 4 wherein a second pair of sets of dark reference pixels of the plurality of pairs of sets of dark reference pixels consists of the second reference pixel of the first pair of sets of dark reference pixels and a third dark reference pixel, the third dark reference pixel being adjacent to the second reference pixel of the first pair of sets of dark reference pixels.

6. The method of claim 5 , comprising:

storing an absolute value of a darkness level of the second pixel of the two dark reference pixels;

computing an absolute difference between a darkness level of the third dark reference pixel and the stored absolute value of the darkness level of the second pixel of the two dark reference pixels; and

updating the accumulated absolute differences based on the computed absolute difference between the darkness level of the third dark reference pixel and the stored absolute value of the darkness level of the second pixel of the two dark reference pixels.

7. The method of claim 1 wherein a first pair of sets of dark reference pixels of the plurality of pairs of sets of dark reference pixels comprises a first group of two or more dark reference pixels and a second group of two or more dark reference pixels.

8. The method of claim 7 wherein the second group of dark reference pixels is adjacent to the first group of dark reference pixels.

9. The method of claim 8 wherein a second pair of sets of dark reference pixels comprises the second group of dark reference pixels and a third group of two or more dark reference pixels, the third group of dark reference pixels being adjacent to the second group of dark reference pixels.

10. The method of claim 1 wherein the dark reference pixels of the image sensor are organized into a plurality of pixel channels and the method comprises estimating respective noise levels of the plurality of pixel channels, wherein the estimated noise level of the dark reference pixels of the image sensor is based on the estimated respective noise levels of the plurality of pixel channels.

11. The method of claim 1 wherein, after each computing of absolute differences in dark signal levels of a pair of sets of dark reference pixels, the accumulated absolute differences are updated.

12. The method of claim 1 , comprising determining an average of the accumulated absolute differences in dark signal levels.

13. An image sensor, comprising:

a plurality of active pixels;

a plurality of dark reference pixels; and

processing circuitry coupled to the plurality of dark reference pixels, wherein the processing circuitry, in operation:

sequentially computes, for each of a plurality of pairs of sets of dark reference pixels of the plurality of dark reference pixels, absolute differences in dark signal levels of the plurality of pairs of sets of dark reference pixels;

accumulates the absolute differences in dark signal levels; and

estimates a noise level of the dark reference pixels of the image sensor based on the accumulated absolute differences.

14. The image sensor of claim 13 wherein a first pair of sets of dark reference pixels of the plurality of pairs of sets of dark reference pixels consists of two dark reference pixels.

15. The image sensor of claim 14 wherein the two dark reference pixels are adjacent.

16. The image sensor of claim 13 , wherein the processing circuitry, in operation:

stores an absolute value of a darkness level associated with a first set of a first pair of sets of dark reference pixels;

computes an absolute difference between a darkness level associated with a second set of the first pair of sets of dark reference pixels and the stored absolute value associated with of the darkness level of the first set of dark reference pixels; and

initializes the accumulated absolute differences to the computed absolute difference between the darkness level associated with the second set of the first pair of sets of dark reference pixels and the stored darkness level of the first set of dark reference pixels.

17. The image sensor of claim 13 wherein a first pair of sets of dark reference pixels of the plurality of pairs of sets of dark reference pixels comprises a first group of two or more dark reference pixels and a second group of two or more dark reference pixels.

18. The image sensor of claim 17 wherein the second group of dark reference pixels is adjacent to the first group of dark reference pixels.

19. The image sensor of claim 18 wherein a second pair of sets of dark reference pixels comprises the second group of dark reference pixels and a third group of two or more dark reference pixels, the third group of dark reference pixels being adjacent to the second group of dark reference pixels.

20. The image sensor of claim 17 wherein the processing circuitry, in operation:

stores a sum of dark signal levels of the pixels of the first group of dark reference pixels; and

computes an absolute difference between a sum of dark signal levels of the pixels of the second group of dark reference pixels and the stored sum of the dark signal levels of the first group of dark reference pixels.

21. The image sensor of claim 13 wherein the dark reference pixels of the image sensor are organized into a plurality of pixel channels and the processing circuitry, in operation, estimates respective noise levels of the plurality of pixel channels, and the estimated noise level of the dark reference pixels of the image sensor is based on the estimated respective noise levels of the plurality of pixel channels.

22. The image sensor of claim 13 wherein, after each computing of absolute differences in dark signal levels of a pair of sets of dark reference pixels, the processing circuitry, in operation, updates the accumulated absolute differences.

23. The image sensor of claim 13 wherein the processing circuitry, in operation, determines an average of the accumulated absolute differences in dark signal levels.

24. The image sensor of claim 13 wherein the image sensor is a color image sensor.

25. A system, comprising:

an array of pixels including a plurality of active pixels and a plurality of dark reference pixels; and

processing circuitry coupled to the array of pixels, wherein the processing circuitry, in operation:

sequentially computes, for each of a plurality of pairs of sets of dark reference pixels of the plurality of dark reference pixels, absolute differences in dark signal levels of the plurality of pairs of sets of dark reference pixels;

accumulates the absolute differences in dark signal levels; and

estimates a noise level of the dark reference pixels of the array of pixels based on the accumulated absolute differences.

26. The system of claim 25 wherein a first pair of sets of dark reference pixels of the plurality of pairs of sets of dark reference pixels consists of two dark reference pixels.

27. The system of claim 25 wherein a first pair of sets of dark reference pixels of the plurality of pairs of sets of dark reference pixels comprises a first group of two or more dark reference pixels and a second group of two or more dark reference pixels.

28. The system of claim 25 wherein the dark reference pixels of the array of pixels are organized into a plurality of pixel channels and the processing circuitry, in operation, estimates respective noise levels of the plurality of pixel channels, and the estimated noise level of the dark reference pixels of the array of pixels is based on the estimated respective noise levels of the plurality of pixel channels.

29. The system of claim 25 , comprising a back-up camera having an integrated circuit including the array of pixels and the processing circuitry.

30. The system of claim 25 , comprising a mobile phone having an integrated circuit including the array of pixels and the processing circuitry.