Image processing method

• US 20090290067A1
• Filed: 07/24/2009
• Published: 11/26/2009
• Est. Priority Date: 02/02/2007
• Status: Active Grant

• Current Assignee: Nikon Corporation
• Original Assignee: Nikon Corporation
• Inventors: Ishiga, Kenichi
• Granted Patent: US 8,369,640 B2
• Time in Patent Office: Days
• US Class Current: 348/607
• CPC Class Codes: G06T 2207/20064 Wavelet transform [DWT]
• CPC Class Codes: G06T 2207/20192 Edge enhancement; Edge preservation
• CPC Class Codes: G06T 5/002 Denoising; Smoothing noise processing or correction adapted to be used in an image pickup device containing and electronic image sensor H04N23/81, H04N25/60 - H04N25/67
• CPC Class Codes: G06T 5/50 using two or more images, e.g. averaging or subtraction
• CPC Class Codes: G06T 5/70 Denoising; Smoothing
• CPC Class Codes: G06T 7/13 Edge detection
• CPC Class Codes: H04N 1/409 Edge or detail enhancement; Noise or error suppression

Abstract

An image processing method for removing a noise component contained in an original image, includes: separating an original image into a temporary noise-free image and a temporary noise component; extracting an edge component in the temporary noise-free image by executing edge detection on the temporary noise-free image; determining a fine edge component in the original image contained in the temporary noise component based upon a level of the extracted edge component; extracting an actual noise component by excluding the fine edge component from the temporary noise component; and removing noise from the original image based upon the extracted actual noise component.

Claims

1. An image processing method for removing a noise component contained in an original image, comprising:

separating an original image into a temporary noise-free image and a temporary noise component;

extracting an edge component in the temporary noise-free image by executing edge detection on the temporary noise-free image;

determining a fine edge component in the original image contained in the temporary noise component based upon a level of the extracted edge component;

extracting an actual noise component by excluding the fine edge component from the temporary noise component; and

removing noise from the original image based upon the extracted actual noise component.

2. An image processing method according to claim 1, wherein:

a probability of the fine edge component being present in the temporary noise component is estimated to be higher when the level of the edge component is greater relative to a noise fluctuation index value determined uniquely in correspondence to the original image.

3. An image processing method according to claim 2, wherein:

a mixing ratio with which the fine edge component is present is statistically estimated based upon a Gaussian distribution probability assuming a ratio of the edge component and the noise fluctuation index value as an argument.

4. An image processing method according to claim 3, wherein:

the fine edge component is excluded through attenuation by multiplying the temporary noise component by the Gaussian distribution probability.

5. An image processing method according to claim 1, wherein:

the original image is separated into the temporary noise-free image having undergone adaptive smoothing and the temporary noise component representing a residue thereof by setting a noise fluctuation index value determined uniquely in correspondence to the original image and comparing a signal fluctuation width detected in an original image structure with the noise fluctuation index value.

6. An image processing method according to claim 5, wherein:

the temporary noise-free image having undergone adaptive smoothing is generated through weighted averaging of signal values at pixels near a target pixel, which fluctuate over a narrow range equal to or less than a range indicated by the noise fluctuation value, the weighted averaging being selectively executed according to a signal differential value; and

the temporary noise component is obtained by determining a difference between the temporary noise-free image and the original image.

7. An image processing method according to claim 5, wherein:

the temporary noise component is extracted through a signal conversion processing phase in which only a component fluctuating over a small range equal to or less than a range indicated by the noise fluctuation index value in a differential signal representing a difference between the original image and a uniformly smooth image, is allowed to pass through; and

the temporary noise-free image is generated by subtracting the temporary noise component from the original image.

8. An image processing method according to claim 1, wherein:

an image referencing range over which the edge detection is executed is set in correspondence to an image referencing range set for generation of the temporary noise-free image.

9. An image processing method according to claim 1, wherein:

the original image is a band-limited subband image in multiresolution representation.

10. An image processing method according to claim 1, wherein:

the original image is an original image expressed in an actual space.

11. An image processing method according to claim 1, wherein:

processing on the original image is executed individually on a luminance plane and a chrominance plane.

18. An image processing method according to claim 2, wherein:

a uniform index value is used as the noise fluctuation index value irrespective of a brightness level of the original image by converting the original image to an image in uniform noise space or an image in a uniform color•

uniform noise space.

12. An image processing method for removing a noise component contained in an original image, comprising:

generating a plurality of band-limited images assuming sequentially lower resolution levels by filtering an original image;

separating each band-limited image to a temporary noise-free image and a temporary noise component;

extracting an edge component in the temporary noise-free image in correspondence to each band-limited image through edge detection executed on the temporary noise-free image;

synthesizing temporary noise component signals in the band-limited images having been separated, so as to generate a synthesized temporary noise component assuming a resolution level matching a resolution level of the original image;

synthesizing edge component signals in the band-limited images having been extracted, so as to generate a synthesized edge component assuming the resolution level matching the resolution level of the original image;

determining, based upon the synthesized edge component, a quantity of an edge component in the original image contained in the synthesized temporary noise component;

extracting an actual noise component by excluding the edge component in the original image from the synthesized temporary noise component; and

removing noise from the original image based upon the extracted actual noise component.

13. An image processing method according to claim 12, wherein:

a probability of the edge component having been mixed in during temporary noise removal is estimated to be higher when a level of the synthesized edge component is greater relative to a noise fluctuation index value determined uniquely in correspondence to the original image.

14. An image processing method according to claim 13, wherein:

a mixing ratio with which the edge component has been mixed in during the temporary noise removal is statistically estimated based upon a Gaussian distribution probability assuming a ratio of the synthesized edge component and the noise fluctuation index value as an argument.

15. An image processing method according to claim 14, wherein:

the edge component having been mixed in during the temporary noise removal is excluded through attenuation by multiplying the synthesized temporary noise component by the Gaussian distribution probability.

16. An image processing method according to claim 12, wherein:

the temporary noise removal and synthesis of the temporary noise component are both executed by using band-limited images that correspond to low-frequency images and high-frequency images respectively.

17. An image processing method according to claim 12, wherein:

extraction of the edge component and synthesis of the extracted edge component are both executed by using band-limited images that correspond to only high-frequency images.

19. An image processing method for removing a noise component contained in an original image, comprising:

extracting a noise component contained in an original image;

generating a plurality of band-limited images assuming sequentially lower resolution levels by filtering the original image;

extracting an edge component in each of the band-limited images by executing edge detection on the band-limited image;

synthesizing edge component signals extracted from the band-limited images to generate an synthesized edge component assuming a resolution level matching a resolution level of the original image; and

removing noise from the original image based upon the extracted noise component and the synthesized edge component.

20. An image processing method according to claim 19, wherein:

a quantity of an edge component present in the extracted noise component is estimated based upon a level of the synthesized edge component;

the edge component present in the extracted noise component is excluded from the extracted noise component based upon estimation results; and

noise is removed from the original image based upon the noise component from which the edge component has been excluded.

21. An image processing method according to claim 19, wherein:

a noise component is extracted from each of the band-limited images; and

noise component signals extracted from the band-limited images are synthesized to generate an synthesized noise component assuming a resolution level matching the resolution level of the original image and the noise component contained in the original image is extracted.

22. An image processing method according to claim 19, wherein:

a temporary noise-free image is generated in correspondence to each of the band-limited images prior to edge component extraction;

a residual edge component remaining following temporary noise removal is extracted in correspondence to each band-limited image by executing the edge detection on the band-limited image having undergone the temporary noise removal; and

residual edge component signals having been extracted from the band-limited images are synthesized as the edge component so as to generate an synthesized edge component assuming a resolution level matching the resolution of the original image.