Methods and systems for detection and estimation of mosquito noise
Aspects of the present invention are related to systems and methods for detection and estimation of mosquito noise in an image or in a video sequence. A mosquito-noise estimate, at a pixel location, may be computed based on an image ring parameter, a statistical value of noise estimates in a first region associated with the pixel location, a maximum activity value in a second region associated with the pixel location and a noise estimate at the pixel location.
1. A method for estimating mosquito noise at a first pixel location in an image, said method comprising:
computing a ring parameter for the image;
receiving a noise estimate for the first pixel;
computing a maximum of respective activity values for a second region including the first pixel location; and
estimating mosquito noise at the first pixel location as the maximum of:
the noise estimate for the first pixel, and
a product of the ring parameter, the maximum activity value, and a multiplicative constant.
2. The method as described in claim 1 , wherein computing the ring parameter for the image includes computing the ring parameter for the image based on an additive-noise estimate for the image and a compression-noise estimate for the image.
3. The method as described in claim 1 , wherein the noise estimate for the first pixel is the average of noise estimates for each pixel included in a first region including the first pixel.
4. The method as described in claim 3 , wherein the first region is a square region centered on the first pixel location.
5. The method as described in claim 1 , wherein the second region is a square region centered on the first pixel location.
6. The method as described in claim 1 , wherein computing the maximum of respective activity values for the second region comprises computing an activity value corresponding to each pixel location in the second region and wherein computing an activity value corresponding to a pixel location includes computing a standard-deviation value of a plurality of pixel values in a corresponding region including the pixel location.
7. The method as described in claim 1 , wherein the image is an image selected from the group consisting of a luminance image, an R-channel image, a G-channel image, a B-channel image and a chrominance image.
8. The method as described in claim 1 , wherein the multiplicative constant is based on the maximum activity value.
9. The method as described in claim 1 , wherein the multiplicative constant is a first constant value when the maximum activity value meets a first criterion and the multiplicative constant is a second constant value when the maximum activity value does not meet the first criterion.
10. The method as described in claim 1 , wherein:
the multiplicative constant is a first constant value when the maximum activity value is less than a first threshold value;
the multiplicative constant is a second constant value when the maximum activity value is greater than a second threshold value; and
the multiplicative constant is a third constant value when the maximum activity value is between the first threshold value and the second threshold value.
11. The method as described in claim 1 , wherein:
the multiplicative constant is a first constant value when the maximum activity value is less than a first threshold value;
the multiplicative constant is a second constant value when the maximum activity value is greater than a second threshold value; and
the multiplicative constant varies smoothly between the first constant value and the second constant value when the maximum activity value is between the first threshold value and the second threshold value.
12. The method as described in claim 1 , wherein:
the multiplicative constant is a first constant value when the maximum activity value is less than a first threshold value;
the multiplicative constant is a second constant value when the maximum activity value is greater than a second threshold value; and
the multiplicative constant varies smoothly between the first constant value and a third constant value when the maximum activity value is between the first threshold value and the second threshold value, wherein the third constant value is greater than the first constant value and less than the second constant value.
13. A non-transitory computer-readable medium encoded with a computer program code for implementing a method for estimating additive noise in an image, said method comprising:
computing a ring parameter for the image;
receiving a noise estimate for the first pixel;
computing a maximum of respective activity values for a second region including the first pixel location; and
estimating mosquito noise at the first pixel location as the maximum of:
the noise estimate for the first pixel, and
a product of the ring parameter, the maximum activity value, and a multiplicative constant.
14. The non-transitory computer-readable medium as described in claim 13 , wherein computing the ring parameter for the image includes computing the ring parameter for the image based on an additive-noise estimate for the image and a compression-noise estimate for the image.
15. A non-transitory computer-readable medium as described in claim 13 , wherein the multiplicative constant is based on the maximum activity value.
16. The non-transitory computer-readable medium as described in claim 13 , wherein the noise estimate for the first pixel is the average of noise estimates for each pixel included in a first region including the first pixel.
17. The non-transitory computer-readable medium as described in claim 16 , wherein computing the maximum of respective activity values for the second region comprises computing an activity value corresponding to each pixel location in the second region and wherein computing an activity value corresponding to a pixel location includes computing a standard-deviation value of a plurality of pixel values in a corresponding region including the pixel location.
18. A method for estimating mosquito noise for a pixel of interest in an image, the method comprising:
computing a ring parameter for the image;
for each pixel included in an region including the pixel of interest, computing a respective activity value;
computing a maximum of the respective activity values; and
estimating mosquito noise for the pixel of interest as the product of the ring parameter, the maximum of the respective activity values, and a multiplicative constant.
19. The method of claim 18 , wherein computing a respective activity value includes computing a standard-deviation value of pixel values in a respective N×N region corresponding to a pixel.
20. The method of claim 19 , wherein a respective region respective N×N region corresponding to the pixel is centered on a location of the pixel.
21. The method of claim 20 , wherein computing the ring parameter for the image includes computing the ring parameter for the image based on an additive-noise estimate for the image and a compression-noise estimate for the image.
22. The method of claim 20 , wherein the multiplicative constant is a first constant value when the maximum activity value meets a first criterion.