Noise reduction method, medium, and system

• US 20070296871A1
• Filed: 06/04/2007
• Published: 12/27/2007
• Est. Priority Date: 06/22/2006
• Status: Active Grant

• Current Assignee: Samsung Electronics Co. Ltd.
• Original Assignee: Samsung Electronics Co. Ltd.
• Inventors: Lee, Seong-deok, Yoo, Young-Jin
• Granted Patent: US 8,467,003 B2
• Time in Patent Office: Days
• US Class Current: 348/701
• CPC Class Codes: G06T 2207/10024 Color image
• CPC Class Codes: G06T 2207/20012 Locally adaptive
• CPC Class Codes: G06T 5/20 using local operators
• CPC Class Codes: G06T 5/70 Denoising; Smoothing
• CPC Class Codes: H04N 23/81 for suppressing or minimising disturbance in the image signal generation

Abstract

A noise reduction method, medium, and system. The noise reduction method includes calculating a noise level of an input image and removing noise from a central pixel within a window of a predetermined size in the input image using a weight determined based on a difference between signal intensities of the central pixel and a plurality of adjacent pixels within the window and the calculated noise level.

Claims

1. A noise reduction method, comprising:

calculating a noise level of an input image; and

controlling a removing of noise from a central pixel within a window of a predetermined size in the input image using a weight determined based on a difference between signal intensities of the central pixel and a plurality of adjacent pixels within the window and the calculated noise level.

2. The noise reduction method of claim 1, wherein the calculated noise level comprises at least one of a first noise level based on an auto gain control (AGC) value of the input image when photographed and a second noise level based on a signal intensity of the input image.

3. The noise reduction method of claim 1, wherein the removal of the noise from the central pixel comprises:

calculating weights to be assigned to the plurality of adjacent pixels based on differences between the signal intensities and the calculated noise level;

compensating for the calculated weights according to an edge direction of the central pixel; and

outputting a mean of weighted sums between the signal intensities of the central pixel and the plurality of adjacent pixels as a final signal intensity of the central pixel,

wherein the mean of the weighted sums is calculated using the calculated weights.

4. The noise reduction method of claim 3, wherein the calculation of the weights comprises calculating the weights using a fuzzy membership function whose resultant values have a Gaussian distribution according to the differences between the signal intensities of the central pixel and the plurality of adjacent pixels.

5. The noise reduction method of claim 3, wherein the compensation of the calculated weights comprises:

assigning predetermined mask values to the plurality of adjacent pixels according to the edge direction of the central pixel; and

providing the compensated weights based on respective calculated weights of the plurality of adjacent pixels and the mask values.

6. The noise reduction method of claim 5, wherein higher mask values are assigned to at least one of the plurality of adjacent pixels existing in the edge direction of the central pixel from the central pixel than mask values assigned to remaining adjacent pixels of the plurality of adjacent pixels.

7. The noise reduction method of claim 5, wherein the providing of the compensated weights implements a bit shift operation.

8. The noise reduction method of claim 3, wherein the mean of the weighted sums is calculated using a fuzzy spatial filter.

9. The noise reduction method of claim 3, wherein the outputting of the mean of the weighted sums comprises outputting a mean of signal intensities of the plurality of adjacent pixels as the final signal intensity of the central pixel when impulse noise is determined to be contained in the central pixel.

10. The noise reduction method of claim 9, wherein the outputting of the mean of the weighted sums comprises determining whether the impulse noise is contained in the central pixel using a rank-ordered absolute differences statistic (ROAD) calculated for the window.

11. The noise reduction method of claim 10, wherein the determination of whether the impulse noise is contained in the central pixel comprises determining that the impulse noise is contained in the central pixel when the ROAD is greater than a first threshold value.

12. The noise reduction method of claim 10, wherein the determination of whether the impulse noise is contained in the central pixel comprises determining that the impulse noise is not contained in the central pixel when the ROAD is less than the first threshold value but greater than a second threshold value and a variance of the signal intensities of the central pixel and the plurality of adjacent pixels is greater than a third threshold value.

13. The noise reduction method of claim 10, wherein the determination of whether the impulse noise is contained in the central pixel comprises determining that the impulse noise is contained in the central pixel if all differences between the signal intensities of the central pixel and the plurality of adjacent pixels are greater than a fourth threshold value when the ROAD is less than the first threshold value but greater than the second threshold value and the variance of the signal intensities of the central pixel and the plurality of adjacent pixels is less than the third threshold value.

14. The noise reduction method of claim 13, wherein the determination that the impulse noise is contained in the central pixel is made when an illuminance value of the input image is less than a predetermined threshold value.

15. The noise reduction method of claim 1, wherein the noise level is separately calculated for each optical channel of the input image, and the removal of the noise from the central pixel comprises removing noise from respective central pixels for each optical channel of the input image.

16. The noise reduction method of claim 1, wherein the noise reduction method is implemented by a photographing system.

17. At least one medium comprising computer readable code to control at least one processing element to implement the method of claim 1.

18. A system with noise reduction, comprising:

a noise level calculation unit to calculate a noise level of an input image; and

a filtering unit to control a removing of noise from a central pixel within a window of a predetermined size in the input image using a weight determined based on a difference between signal intensities of the central pixel and plurality of adjacent pixels within the window and the calculated noise level.

19. The system of claim 18, wherein the calculated noise level comprises at least one of a first noise level based on an AGC value of the input image when photographed and a second noise level based on a signal intensity of the input image.

20. The system of claim 18, further comprising:

a weight calculation unit to calculate weights to be assigned to the plurality of adjacent pixels based on differences between the signal intensities and the calculated noise level; and

a weight compensation unit to compensate for the calculated weights according to an edge direction of the central pixel,

wherein the filtering unit calculates and outputs a mean of weighted sums of the signal intensities of the central pixel and the plurality of adjacent pixels using the calculated weights, with the output mean being a final signal intensity of the central pixel.

21. The system of claim 20, wherein the weight calculation unit calculates the weights using a fuzzy membership function whose resultant values have a Gaussian distribution according to the differences between the signal intensities of the central pixel and the plurality of adjacent pixels.

22. The system of claim 20, wherein the weight compensation unit assigns predetermined mask values to the plurality of adjacent pixels according to the edge direction of the central pixel and provides the compensated weights based on the respective calculated weights of the plurality of adjacent pixels and the mask values.

23. The system of claim 22, wherein higher mask values are assigned to at least one of the plurality of adjacent pixels existing in the edge direction of the central pixel from the central pixel than mask values assigned to remaining adjacent pixels of the plurality of adjacent pixels.

24. The system of claim 22, wherein the providing of the compensated weights is implemented through a bit shift operation.

25. The system of claim 20, wherein the filtering unit calculates the mean of the weighed sums using a fuzzy spatial filter.

26. The system of claim 20, wherein the filtering unit outputs a mean of signal intensities of the plurality of adjacent pixels as the final signal intensity of the central pixel when impulse noise is determined to be contained in the central pixel.

27. The system of claim 26, further comprising an impulse noise determination unit which determines whether the impulse noise is contained in the central pixel using a ROAD calculated for the window.

28. The system of claim 27, wherein the impulse noise determination unit determines that the impulse noise is contained in the central pixel when the ROAD is greater than a first threshold value.

29. The system of claim 27, wherein the impulse noise determination unit determines that the impulse noise is not contained in the central pixel when the ROAD is less than the first threshold value but greater than a second threshold value and a variance of the signal intensities of the central pixel and the plurality of adjacent pixels is greater than a third threshold value.

30. The system of claim 27, wherein the impulse noise determination unit determines that the impulse noise is contained in the central pixel if all differences between the signal intensities of the central pixel and the plurality of adjacent pixels are greater than a fourth threshold value when the ROAD is less than the first threshold value but greater than the second threshold value and the variance of the signal intensities of the central pixel and the plurality of adjacent pixels is less than the third threshold value.

31. The system of claim 30, wherein the impulse noise determination unit determines that the impulse noise is contained in the central pixel when an illuminance value of the input image is less than a predetermined threshold value.

32. The system of claim 18, wherein the noise level is separately calculated for each optical channel of the input image, and the filtering unit removes the noise from respective central pixels for each optical channel of the input image.

33. A photographing system, comprising:

a photographing unit to photograph a subject as an input image; and

a noise reduction unit to calculate a noise level of the input image and control a removal of noise from a central pixel within a window of a predetermined size in the input image using a weight determined based on a difference between signal intensities of the central pixel and a plurality of adjacent pixels within the window and the calculated noise level.