Texture masking for video quality measurement
View Patent ↗A particular implementation decomposes an image into a structure component and a texture component. An edge strength map is calculated for the structure component, and a texture strength map is calculated for the texture component. Using the edge strength and the texture strength, texture masking weights are calculated. The stronger the texture strength is, or the weaker the edge strength is, the more distortion can be tolerated by human eyes, and thus, the smaller the texture masking weight is. The local distortions are then weighted by the texture masking weights to generate an overall distortion level or an overall quality metric.
1. A method for estimating video quality, comprising:
accessing image data having at least one image region;
decomposing the image region into a structure component and a texture component;
determining an edge strength for the structure component of the image region;
determining a texture masking weight in response to the edge strength,
wherein the texture masking weight is set such that no texture masking property is used when the edge strength exceeds a threshold,
wherein the texture masking weight is determined responsive to a texture strength for the texture component of the image region when the edge strength does not exceed the threshold, and
wherein the texture masking weight is set such that no texture masking property is used when the texture strength of the texture component of the image region does not exceed a second threshold; and
determining a quality metric in response to a local distortion of the image region weighted by the texture masking weight.
2. The method of claim 1 , wherein a bilateral filter is used to decompose the image region into the structure component and the texture component.
3. The method of claim 1 , wherein the texture strength is determined as a variance of the texture component of the image region.
4. The method of claim 1 , wherein the image data comprises a plurality of image regions, the decomposing, determining the edge strength and determining the texture masking weight comprise decomposing, determining respective edge strengths and determining respective texture masking weights for the plurality of image regions, and wherein the quality metric is determined in response to a combination of the weighted local distortions for the plurality of image regions.
5. An apparatus for estimating video quality, comprising at least one memory and one or more processors, wherein the one or more processors are configured to:
access image data having at least one image region;
decompose an image region into a structure component and a texture component;
determine an edge strength for the structure component of the image region;
determine a texture masking weight in response to the edge strength,
wherein the texture masking weight is set such that no texture masking is used when the edge strength exceeds a threshold,
wherein the texture masking weight is determined responsive to a texture strength for the texture component of the image region when the edge strength does not exceed the threshold, and
wherein the texture masking weight is set such that no texture masking property is used when the texture strength of the texture component of the image region does not exceed a second threshold; and
determine a quality metric in response to a local distortion of the image region weighted by the texture masking weight.
6. The apparatus of claim 5 , wherein a bilateral filter is used to decompose the image region into the structure component and the texture component.
7. The apparatus of claim 5 , wherein the texture strength is determined as a variance of the texture component of the image region.
8. The apparatus of claim 5 , wherein the image data comprises a plurality of image regions, the one or more processors are configured to respectively decompose, determine respective edge strengths and determine respective texture masking weights for the plurality of image regions, and determine the quality metric in response to a combination of the weighted local distortions for the plurality of image regions.
9. A non-transitory processor readable medium having stored thereupon instructions for causing one or more processors to collectively perform:
decomposing an image region into a structure component and a texture component;
determining an edge strength for the structure component of the image region;
determining a texture masking weight in response to the edge strength,
wherein the texture masking weight is set such that no texture masking is used when the edge strength exceeds a threshold,
wherein the texture masking weight is determined responsive to a texture strength for the texture component of the image region when the edge strength does not exceed the threshold, and
wherein the texture masking weight is set such that no texture masking property is used when the texture strength of the texture component of the image region does not exceed a second threshold; and
determining a quality metric in response to a local distortion of the image region weighted by the texture masking weight.