Noise reduction apparatus, noise reduction method, and computer-readable storage medium

• US 9,299,131 B2
• Filed: 08/21/2012
• Issued: 03/29/2016
• Est. Priority Date: 08/22/2011
• Status: Active Grant

• Current Assignee: NEC Corporation
• Original Assignee: NEC Corporation
• Inventors: Toda, Masato
• Primary Examiner(s): Alavi, Amir
• Application Number:
  US14/239,577
• Publication Number:
  US 20140205202A1
• Time in Patent Office: 1,316 Days
• US Class Current: 1/1
• CPC Class Codes: G06T 2207/20016 Hierarchical, coarse-to-fine, multiscale or multiresolution image processing; Pyramid transform
• CPC Class Codes: G06T 2207/20076 Probabilistic image processing
• CPC Class Codes: G06T 5/70 Denoising; Smoothing
• CPC Class Codes: H04N 1/409 Edge or detail enhancement; Noise or error suppression

Abstract

The present invention is directed to a noise reduction method, comprising: for each of multi-layer regions each containing a pixel of interest and having a successively reducing area, calculating a pixel statistic value of pixels in that region; for each of successive layers, correcting the pixel statistic value for a region at a current layer using a corrected pixel statistic value for a region at a preceding layer having a greater area than that of the region at the current layer; and correcting the pixel of interest using a corrected pixel statistic value for a region with a smallest area.

Claims

1. A noise reduction method, comprising:

for each of multi-layer regions each containing a pixel of interest and having a successively reducing area, calculating a pixel statistic value of pixels in that region;

for each of successive layers, correcting the pixel statistic value for a region at a current layer using a corrected pixel statistic value for a region at a preceding layer having a greater area than that of the region at the current layer; and

correcting said pixel of interest using a corrected pixel statistic value for a region with a smallest area.

2. The noise reduction method according to claim 1, comprising:

calculating edge information for a region at each layer; and

correcting the pixel statistic value in the region at the current layer using the corrected pixel statistic value at a preceding layer and edge information at the preceding layer.

3. The noise reduction method according to claim 2, comprising:

weighting to the corrected pixel statistic value at the preceding layer according to the edge information at the preceding layer.

4. The noise reduction method according to claim 3, wherein:

in a case that the edge information for the region at the preceding layer exceeds a predetermined threshold, no correction is applied to the pixel statistic value at the current layer.

5. The noise reduction method according to claim 1, comprising:

modifying a parameter in the correcting function for correcting the pixel statistic value on a layer-by-layer basis.

6. The noise reduction method according to claim 5, comprising:

modifying the parameter in the correcting function for correcting the pixel statistic value according to the amount of change in the pixel value due to a noise in the region at the current layer.

7. The noise reduction method according to claim 5, comprising:

modifying the parameter in the correcting function for correcting the pixel statistic value on a layer-by-layer basis based on the edge information.

8. The noise reduction method according to claim 1, wherein:

the function for correcting the pixel statistic value is a function having such a property as to generate an output value coming closer to zero for an input value for the function closer to zero, and an output value coming closer to the input value for a greater absolute value of the input value.

9. The noise reduction method according to claim 1, wherein:

said pixel statistic value is a spatial average value of pixels in a region at each layer.

10. The noise reduction method according to claim 9, wherein:

said spatial average value is any one of an arithmetic mean value, a geometric mean value, and a weighted average value of pixels in a region at each layer.

11. A noise reduction apparatus, comprising:

a memory storing instructions;

one or more processors configured to execute the instructions to;

for each of multi-layer regions each containing a pixel of interest and having a successively reducing area, calculate a pixel statistic value of pixels in that region, and

for each of successive layers, correct the pixel statistic value for a region at a current layer using a corrected pixel statistic value for a region at a preceding layer having a greater area than that of the region at the current layer, and correct said pixel of interest using a corrected pixel statistic value for a region with a smallest area.

12. The noise reduction apparatus according to claim 11, wherein the one or more processors are further configured to execute the instructions to:

calculate edge information for a region at each layer, and

correct the pixel statistic value in the region at the current layer using the corrected pixel statistic value at a preceding layer and edge information at the preceding layer.

13. The noise reduction apparatus according to claim 12, wherein the one or more processors are further configured to execute the instructions to:

apply weighting to the corrected pixel statistic value at the preceding layer according to the edge information at the preceding layer.

14. The noise reduction apparatus according to claim 13, wherein the one or more processors are further configured to execute the instructions to:

in a case that the edge information for the region at the preceding layer exceeds a predetermined threshold, apply no correction to the pixel statistic value at the current layer.

15. The noise reduction apparatus according to claim 11, wherein the one or more processors are further configured to execute the instructions to:

modify a parameter in the correcting function for correcting the pixel statistic value on a layer-by-layer basis.

16. The noise reduction apparatus according to claim 15, wherein the one or more processors are further configured to execute the instructions to:

modify the parameter in the correcting function for correcting the pixel statistic value according to the amount of change in the pixel value due to a noise in the region at the current layer.

17. The noise reduction apparatus according to claim 15, wherein the one or more processors are further configured to execute the instructions to:

modify the parameter in the correcting function for correcting the pixel statistic value on a layer-by-layer basis based on the edge information.

18. The noise reduction apparatus according to claim 11, wherein the one or more processors are further configured to execute the instructions to:

correct said pixel statistic value by a function having such a property as to generate an output value coming closer to zero for an input value closer to zero, and an output value coming closer to the input value for a greater absolute value of the input value.

19. The noise reduction apparatus according to claim 11, wherein the one or more processors are further configured to execute the instructions to:

calculate a spatial average value of pixels in a region at each layer as said pixel statistic value.

20. A non-transitory computer readable storage medium storing program that, when executed by a computer, cause the computer to:

for each of multi-layer regions each containing a pixel of interest and having a successively reducing area, calculating a pixel statistic value of pixels in that region;

for each of successive layers, correcting the pixel statistic value for a region at a current layer using a corrected pixel statistic value for a region at a preceding layer having a greater area than that of the region at the current layer; and

correcting said pixel of interest using a corrected pixel statistic value for a region with a smallest area.