Method and device for reducing noise in a component of a picture
View Patent ↗The present principles relate to a method and a device for reducing noise in a component of a picture. The method includes obtaining a low-pass filtered component by low-pass filtering the component of the picture, for a current pixel in the component of the picture. For at least one current neighboring pixel of the current pixel, computing a distance, relative to the current neighboring pixel, between the value of the current pixel in the low-pass filtered component and the value of the current neighboring pixel in the low-pass filtered component. When the distances relative to the at least one neighboring pixels of the current pixel are lower than a first threshold, modifying the value of the current pixel in said component of the picture according to a linear combination of the value of the current pixel in the component of the picture and the value of the current pixel in the low-pass filtered component. Otherwise the value of the current pixel in the at least one component of the picture remains unchanged.
1. A method for reducing noise in a component of a picture, characterized in that the method comprises:
obtaining a low-pass filtered component by low-pass filtering said component of the picture;
for a current pixel in said component of the picture:
for at least one current neighboring pixel of the current pixel, computing a distance, relative to the current neighboring pixel, between the value of the current pixel in the low-pass filtered component and the value of the current neighboring pixel in the low-pass filtered component;
when the distances relative to said at least one neighboring pixels of the current pixel are lower than a first threshold, modifying the value of the current pixel in said component of the picture according to a linear combination of the value of the current pixel in said component of the picture and the value of the current pixel in the low-pass filtered component,
otherwise the value of the current pixel in said at least one component of the picture remains unchanged;
computing a distance, relative to the current pixel, between the value of the current pixel in said component of the picture and the value of the current pixel in the low-pass filtered component; and
when said distance, relative to the current pixel, is higher than a third threshold:
checking if said distances are lower than the first threshold; and
when said distances are lower than the first threshold, modifying the value of the current pixel in said component of the picture according to said linear combination, otherwise the value of the current pixel in said at least one component of the picture remains unchanged.
2. The method of claim 1 , wherein the modifying of the value of the current pixel in said component of the picture comprises:
replacing the value of the current pixel in said component of the picture by said linear combination.
3. The method of claim 1 , wherein the modifying of the value of the current pixel in said component of the picture comprises:
if at least one of the distance relative to a neighboring pixel is greater than a second threshold, said second threshold being lower than said first threshold, replacing the value of the current pixel in said component of the picture by said linear combination weighted by a coefficient that depends on both the value of the current pixel in said component, the value of the current pixel in the low-pass filtered component and the first and second thresholds;
otherwise replacing the value of the current pixel in said component of the picture by said linear combination.
4. The method of claim 1 , wherein said linear combination is defined by the value of the current pixel in said component of the picture and a difference between the value of the current pixel in said component of the picture and the value of the current pixel in the low-pass filtered component.
5. The method of claim 4 , wherein said linear combination is defined by a difference between the value of the current pixel in said component of the picture and a product of the difference between the value of the current pixel in said component of the picture and the value of the current pixel in the low-pass filtered component by an attenuation function of the value of the current pixel in said component of the picture.
6. The method of claim 5 , wherein said attenuation function of the value of the current pixel in said component of the picture is a piecewise-linear function.
7. The method of claim 5 , wherein the values of said attenuation function are obtained from a look-up table.
8. A device for reducing noise in a component of a picture, characterized in that the device comprises a processor configured to:
obtain a low-pass filtered component by low-pass filtering said component of the picture;
for a current pixel in said component of the picture:
for at least one current neighboring pixel of the current pixel, compute a distance, relative to the current neighboring pixel, between the value of the current pixel in the low-pass filtered component and the value of the current neighboring pixel in the low-pass filtered component;
when the distances relative to said at least one neighboring pixels of the current pixel are lower than a first threshold, modify the value of the current pixel in said component of the picture according to a linear combination of the value of the current pixel in said component of the picture and the value of the current pixel in the low-pass filtered component,
otherwise the value of the current pixel in said at least one component of the picture remains unchanged;
computing a distance, relative to the current pixel, between the value of the current pixel in said component of the picture and the value of the current pixel in the low-pass filtered component; and
when said distance, relative to the current pixel, is higher than a third threshold:
checking if said distances are lower than the first threshold; and
when said distances are lower than the first threshold, modifying the value of the current pixel in said component of the picture according to said linear combination, otherwise the value of the current pixel in said at least one component of the picture remains unchanged.
9. A non-transitory storage medium having stored therein instructions for causing a processor to perform at least the steps of the encoding method according to claim 1 .
10. A non-transitory storage medium having stored therein instructions for causing a processor to perform at least the steps of the encoding method according to claim 7 .