Automatic parameter estimation for adaptive pixel-based filtering
One particular automatic parameter estimation method and apparatus estimates low level filtering parameters from one or more user controlled high-level filtering parameters. The high level filtering parameters are strength and quality, where strength indicates how much noise reduction will be performed, and quality indicates a tolerance which controls the balance between filtering uniformity and loss of detail. The low level filtering parameters that can be estimated include the spatial neighborhood and/or temporal neighborhood size from which pixel candidates are selected, and thresholds used to verify the “goodness” of the spatially or temporally predicted candidate pixels. More generally, a criterion for filtering digital image data is accessed, and a value is determined for a parameter for use in filtering digital image data, the value being determined based on whether the value results in the criterion being satisfied for at least a portion of a digital image.
1. A method comprising:
accessing a criterion for filtering digital image data, wherein the criterion is indicative of filter strength for a pixel in the digital image data;
accessing a second criterion for filtering the digital image data, wherein the second criterion is indicative of a percentage of pixels in the digital image data that have the filter strength indicated by the criterion; and
determining a value for a parameter for use in filtering digital image data, the value being determined based on whether the value results in at least one of the criterion and the second criterion being satisfied for at least a portion of the digital image data.
2. The method of claim 1 , wherein determining the value comprises determining a value for which the criterion is satisfied for pixels in the portion of the digital image data.
3. The method of claim 1 , wherein the criterion is related to a measure of one or more performance characteristics of a filter for filtering the digital image data.
4. The method of claim 3 , wherein the performance characteristics are related to one or more of filter strength, filter quality, and filter complexity.
5. The method of claim 1 , wherein accessing the criterion comprises receiving the criterion from a user.
6. The method of claim 1 , wherein the portion of the digital image data is determined by:
accessing the digital image data;
evaluating pixel values of the accessed digital image data; and
selecting the portion from the accessed digital image data based on the evaluation.
7. The method of claim 6 , wherein said evaluating pixel values comprises determining a measure of homogeneity of the portion of the accessed digital image data.
8. The method of claim 6 , wherein said evaluating pixel values comprises determining a variance of one or more blocks of pixels, resulting in one or more variances, and said selecting the portion comprises:
comparing the one or more variances to a threshold;
selecting at least one of the one or more blocks based on the comparing; and
selecting the portion that includes pixels in the one or more selected blocks.
9. The method of claim 1 , wherein said determining of said value further comprises determining a size of a spatial neighborhood based on whether the spatial neighborhood size allows the criterion to be satisfied.
10. The method of claim 1 , wherein said determining of said value further comprises determining the size of a temporal neighborhood based on whether the temporal neighborhood size allows the criterion to be satisfied.
11. The method of claim 1 , wherein said determining of said value further comprises determining the size of a spatial neighborhood and the size of a temporal neighborhood based on whether the spatial neighborhood size and temporal neighborhood size together allow the criterion to be satisfied.
12. The method of claim 1 , wherein said determining of said value further comprises:
determining, for the portion, a value for a pixel-difference-threshold such that both of the two criteria are satisfied for the portion of the digital image data.
13. The method of claim 1 , wherein the criterion indicates a number of pixels by which a given pixel is to be averaged.
14. The method of claim 1 , wherein the parameter is related to a threshold on a measure of the goodness of a candidate pixel.
15. The method of claim 14 , wherein the measure of the goodness of a candidate pixel is based on the difference between at least one current pixel and at least one candidate pixel.
16. An apparatus comprising:
a processor configured to receive digital image data and further comprising:
a criteria input receiver configured to access at least one criterion for filtering the digital image data, wherein the criterion is indicative of filter strength for a pixel in the digital image data, and configured to access a second criterion for filtering the digital image data, wherein the second criterion is indicative of a percentage of pixels in the digital image data that have the filter strength indicated by the criterion; and
a parameter determination unit for determining a parameter value for use in filtering the digital image data, said parameter value being determined based on whether the parameter value results in at least one of the at least one criterion and the second criterion being satisfied for at least a portion of the digital image data.
17. The apparatus of claim 16 , wherein the parameter determination unit is configured to determine the parameter value for which the at least one criterion is satisfied for pixels in the portion of the digital image data.
18. The apparatus of claim 16 , wherein the determined parameter value for use in filtering comprises at least one of spatial neighborhood size, temporal neighborhood size, and a threshold for verifying goodness of candidate pixels.
19. The apparatus of claim 16 , wherein the processor is integrated into an encoder and the determined parameters are transmitted with the digital image data.
20. The apparatus of claim 16 , wherein the processor is integrated into a decoder and parameter determination is performed on the received digital image data.
21. An apparatus comprising:
means for accessing a criterion for filtering digital image data, wherein the criterion is indicative of filter strength for a pixel in the digital image data, and for accessing a second criterion for filtering the digital image data, wherein the second criterion is indicative of a percentage of pixels that have the filter strength indicated by the criterion; and
means for determining a value for a parameter for use in filtering the digital image data, the value being determined based on whether the value results in at least one of the criterion and the second criterion being satisfied for at least a portion of the digital image data.
22. A non-transitory program storage device having an application program tangibly embodied thereon, the application program including instructions for performing at least the following:
accessing a criterion for filtering digital image data, wherein the criterion is indicative of filter strength for a pixel in the digital image data;
accessing a second criterion for filtering the digital image data, wherein the second criterion is indicative of a percentage of pixels in the digital image data that have the filter strength indicated by the criterion; and
determining a value for a parameter for use in filtering digital image data, the value being determined based on whether the value results in at least one of the criterion and the second criterion being satisfied for at least a portion of the digital image data.
23. The non-transitory program storage device of claim 22 , wherein the application program further includes instructions for performing at least the following:
accessing the digital image;
evaluating pixel values of the accessed digital image; and
selecting the portion from the accessed digital image based on the evaluation,
wherein determining the value comprises determining the value based on whether the value results in the criterion being satisfied for the selected portion of the accessed digital image data.
24. A method comprising:
filtering digital image data using a filter that includes a parameter having a value determined based on whether the value results in at least one of a user-supplied criterion and a non-user-supplied criterion being satisfied for at least a portion of the digital image data,
wherein said non-user-supplied criterion is established based on the user-supplied criterion, and said non-user-supplied criterion is different from said user-supplied criterion.
25. The method defined in claim 24 , wherein user-supplied criterion and the non-user-supplied criterion are each related to a measure of one or more performance characteristics of a filter for filtering the digital image data.
26. The method of claim 25 , wherein the performance characteristics are related to one or more of filter strength, filter quality, and filter complexity.