Anisotropic denoising method
In accordance with an embodiment of the invention, an anisotropic denoising method is provided that removes sensor noise from a digital image while retaining edges, lines, and details in the image. In one embodiment, the method removes noise from a pixel of interest based on the detected type of image environment in which the pixel is situated. If the pixel is situated in an edge/line image environment, then denoising of the pixel is increased such that relatively stronger denoising of the pixel occurs along the edge or line feature. If the pixel is situated in a detail image environment, then denoising of the pixel is decreased such that relatively less denoising of the pixel occurs so as to preserve the details in the image. In one embodiment, detection of the type of image environment is accomplished by performing simple arithmetic operations using only pixels in a 9 pixel by 9 pixel matrix of pixels in which the pixel of interest is situated. As a result, improved image environment sensitive noise reduction is achieved that requires a relatively low gate count in hardware implementations.
1. A machine-implemented method comprising:
determining an initial denoised pixel value for a color component of a particular pixel based at least in part upon a value of the color component of the particular pixel and values of the color component of a set of other pixels;
for at least a first pixel in the set of other pixels, determining whether the first pixel is to be used in denoising the particular pixel;
in response to determining that the first pixel is to be used to denoise the particular pixel, determining a weight value to be assigned to the first pixel;
determining the initial denoised pixel value for the particular pixel based at least in part upon the weight value and a value of the color component for the first pixel;
determining a type of image environment in which the particular pixel is situated, the type of image environment being one of a set comprising (a) edge/line and (b) detail;
based at least in part upon the type of image environment, determining an adjusted denoising power value for the color component of the particular pixel; and
based at least in part upon the initial denoised pixel value and the adjusted denoising power value, determining a final denoised pixel value for the color component of the particular pixel.
2. The method of claim 1 , further comprising:
based at least in part upon the adjusted denoising power value, determining a final denoised pixel value for one or more additional pixels above the particular pixel and one or more additional pixels below the particular pixel.
3. The method of claim 2 , wherein the particular pixel and the additional pixels above and below the particular pixel are all part of a matrix of pixels, and wherein the final denoised pixel value for the particular pixel and the final denoised pixel values for the additional pixels above and below the particular pixel are all determined by processing pixel values in the same matrix of pixels.
4. The method of claim 3 , wherein determining whether the first pixel is to be used in denoising the particular pixel comprises:
determining a difference indicator value that indicates how different the first pixel is to the particular pixel; and
determining whether the difference indicator value is within a range of values that indicates that the first pixel is sufficiently similar to the particular pixel to be used in denoising the particular pixel.
5. The method of claim 4 , wherein the range of values is determined based at least in part upon a difference threshold value, and wherein the difference threshold value is determined based at least in part upon the value of the particular pixel.
6. The method of claim 4 , wherein determining a weight value to be assigned to the first pixel comprises:
determining which sub range of the range of values the difference indicator value falls within;
assigning a first weight value to the first pixel if the difference indicator value falls within a first sub range; and
assigning a second and different weight value to the first pixel if the difference indicator value falls within a second and different sub range.
7. The method of claim 1 , wherein the set of other pixels consists of four pixels, two vertically aligned with the particular pixel, and two horizontally aligned with the particular pixel.
8. The method of claim 1 , wherein determining the type of image environment in which the particular pixel is situated comprises determining a score value that indicates a likelihood that the particular pixel is situated on an edge or line feature of an image.
9. The method of claim 1 , wherein determining the type of image environment in which the particular pixel is situated comprises determining a score value that indicates an amount of detail surrounding the particular pixel in an image.
10. The method of claim 1 , wherein determining the adjusted denoising power value comprises:
determining an initial denoising power value; and
determining the adjusted denoising power value based at least in part upon the initial denoising power value.
11. The method of claim 10 , wherein determining the initial denoising power value comprises:
determining a first denoising power value; and
based at least in part a radial distance between the particular pixel and a center of an image to which the particular pixel belongs, adjusting the first denoising power value as a function of the radial distance to produce the initial denoising power value.
12. The method of claim 10 , wherein determining the initial denoising power value comprises:
determining a signal level estimate based at least in part upon the value of the particular pixel and one or more values of one or more pixels neighboring the particular pixel; and
determining the initial denoising power value based at least in part upon the signal level estimate.
13. The method of claim 1 , wherein determining the adjusted denoising power value comprises:
if the type of image environment is (a) edge/line, determining the adjusted denoising power value based at least in part upon a score value that indicates a likelihood that the particular pixel is situated on an edge or line feature of an image;
if the type of image environment is (b) detail, determining the adjusted denoising power value based at least in part upon a score value that indicates an amount of detail surrounding the particular pixel in an image.
14. A non-transitory computer-readable medium storing processor executable instructions for performing the method of claim 1 .
15. A device comprising one or more logic elements configured for:
determining an initial denoised pixel value for a color component of a particular pixel based at least in part upon a value of the color component of the particular pixel and values of the color component of a set of other pixels;
for at least a first pixel in the set of other pixels, determining whether the first pixel is to be used in denoising the particular pixel;
in response to determining that the first pixel is to be used to denoise the particular pixel, determining a weight value to be assigned to the first pixel;
determining the initial denoised pixel value for the particular pixel based at least in part upon the weight value and a value of the color component for the first pixel;
determining a type of image environment in which the particular pixel is situated, the type of image environment being one of a set comprising (a) edge/line and (b) detail;
based at least in part upon the type of image environment, determining an adjusted denoising power value for the color component of the particular pixel; and
based at least in part upon the initial denoised pixel value and the adjusted denoising power value, determining a final denoised pixel value for the color component of the particular pixel.
16. The device of claim 15 , further comprising one or more logic elements configured for:
based at least in part upon the adjusted denoising power value, determining a final denoised pixel value for one or more additional pixels above the particular pixel and one or more additional pixels below the particular pixel.
17. The device of claim 16 , wherein the particular pixel and the additional pixels above and below the particular pixel are all part of a matrix of pixels, and wherein the final denoised pixel value for the particular pixel and the final denoised pixel values for the additional pixels above and below the particular pixel are all determined by processing pixel values in the same matrix of pixels.
18. The device of claim 16 , wherein determining whether the first pixel is to be used in denoising the particular pixel comprises:
determining a difference indicator value that indicates how different the first pixel is to the particular pixel; and
determining whether the difference indicator value is within a range of values that indicates that the first pixel is sufficiently similar to the particular pixel to be used in denoising the particular pixel.
19. The device of claim 18 , wherein the range of values is determined based at least in part upon a difference threshold value, and wherein the difference threshold value is determined based at least in part upon the value of the particular pixel.
20. The device of claim 18 , wherein determining a weight value to be assigned to the first pixel comprises:
determining which sub range of the range of values the difference indicator value falls within;
assigning a first weight value to the first pixel if the difference indicator value falls within a first sub range; and
assigning a second and different weight value to the first pixel if the difference indicator value falls within a second and different sub range.
21. The device of claim 15 , wherein the set of other pixels consists of four pixels, two vertically aligned with the particular pixel, and two horizontally aligned with the particular pixel.
22. The device of claim 15 , wherein determining the type of image environment in which the particular pixel is situated comprises determining a score value that indicates a likelihood that the particular pixel is situated on an edge or line feature of an image.
23. The device of claim 15 , wherein determining the type of image environment in which the particular pixel is situated comprises determining a score value that indicates an amount of detail surrounding the particular pixel in an image.
24. The device of claim 15 , wherein determining the adjusted denoising power value comprises:
determining an initial denoising power value; and
determining the adjusted denoising power value based at least in part upon the initial denoising power value.
25. The device of claim 24 , wherein determining the initial denoising power value comprises:
determining a first denoising power value; and
based at least in part a radial distance between the particular pixel and a center of an image to which the particular pixel belongs, adjusting the first denoising power value as a function of the radial distance to produce the initial denoising power value.
26. The device of claim 24 , wherein determining the initial denoising power value comprises:
determining a signal level estimate based at least in part upon the value of the particular pixel and one or more values of one or more pixels neighboring the particular pixel; and
determining the initial denoising power value based at least in part upon the signal level estimate.
27. The device of claim 15 , wherein determining the adjusted denoising power value comprises:
if the type of image environment is (a) edge/line, determining the adjusted denoising power value based at least in part upon a score value that indicates a likelihood that the particular pixel is situated on an edge or line feature of an image;
if the type of image environment is (b) detail, determining the adjusted denoising power value based at least in part upon a score value that indicates an amount of detail surrounding the particular pixel in an image.