Method and system for parameter generation for digital noise reduction based on bitstream properties

1. A method for image processing, the method comprising:

receiving at a host processor, a plurality of picture level parameters generated by a video decoder, wherein said plurality of picture level parameters comprises an average quantization parameter of a current picture and an average quantization parameter of a reference picture;

calculating an adjusted average quantization parameter as a weighted sum of said average quantization parameter of said current picture and said average quantization parameter of said reference picture; and

controlling by said host processor, a digital noise reduction (DNR) module that processes input video based on said calculated adjusted average quantization parameter.

2. The method according to claim 1 , wherein said plurality of picture level parameters comprises one or more of: a reference picture flag, total number of intra macroblocks, total number of inter macroblocks, total number of macroblocks, a sum of low order matrix elements of an intra quantization matrix in zig-zag scan order, a sum of high order matrix elements of said intra quantization matrix in zig-zag scan order, a sum of low order matrix elements of an inter quantization matrix in zig-zag scan order, and/or a sum of high order matrix elements of said inter quantization matrix in zig-zag scan order.

3. The method according to claim 2 , comprising calculating said average quantization parameter of said current picture as a ratio of a sum of quantization parameters of each said macroblock to said total number of said macroblocks.

4. The method according to claim 3 , comprising setting said quantization parameter of skipped macroblocks equal to a quantization parameter of non-skipped macroblocks.

5. The method according to claim 3 , wherein said average quantization parameter of said reference picture, for an intra (I) picture is set equal to said average quantization parameter of said current picture.

6. The method according to claim 3 , wherein said average quantization parameter is:

if said reference picture is a predicted (P) picture, equal to one or both of: said average quantization parameter of said current picture of most recent reference P picture and/or said average quantization parameter of said current picture of most recent reference intra (I) picture; and

if said reference picture is a bidirectional (B) picture, equal to an arithmetic average of said average quantization parameter of said current picture values for two most recently decoded reference pictures.

7. The method according to claim 3 , wherein said weighted sum is one or more of: a function of ratio of said total number of said intra macroblocks to said total number of said macroblocks, a function of ratio of said total number of said inter macroblocks to said total number of said macroblocks, a function of said sum of said low order matrix elements, and/or a function of said sum of said high order matrix elements.

8. The method according to claim 1 , comprising controlling said digital noise reduction (DNR) module by said host processor via received user controlled information and at least a portion of said received plurality of picture level parameters.

9. The method according to claim 1 , comprising reducing one or both of: mosquito noise artifacts and/or block noise artifacts based on at least a portion of said generated plurality of picture level parameters.

10. A machine-readable storage having stored thereon, a computer program having at least one code section for image processing, the at least one code section being executable by a machine for causing the machine to perform steps comprising:

receiving at a host processor, a plurality of picture level parameters generated by a video decoder, wherein said plurality of picture level parameters comprises an average quantization parameter of a current picture and an average quantization parameter of a reference picture;

calculating an adjusted average quantization parameter as a weighted sum of said average quantization parameter of said current picture and said average quantization parameter of said reference picture; and

controlling by said host processor, a digital noise reduction (DNR) module that processes input video based on said calculated adjusted average quantization parameter.

11. The machine-readable storage according to claim 10 , wherein said plurality of picture level parameters comprises one or more of: a reference picture flag, total number of intra macroblocks, total number of inter macroblocks, total number of macroblocks, a sum of low order matrix elements of an intra quantization matrix in zig-zag scan order, a sum of high order matrix elements of said intra quantization matrix in zig-zag scan order, a sum of low order matrix elements of an inter quantization matrix in zig-zag scan order, and/or a sum of high order matrix elements of said inter quantization matrix in zig-zag scan order.

12. The machine-readable storage according to claim 11 , wherein said at least one code section comprises code for calculating said average quantization parameter of said current picture as a ratio of a sum of quantization parameters of each said macroblock to said total number of said macroblocks.

13. The machine-readable storage according to claim 12 , wherein said at least one code section comprises code for setting said quantization parameter of skipped macroblocks equal to a quantization parameter of non-skipped macroblocks.

14. The machine-readable storage according to claim 12 , wherein said average quantization parameter of said reference picture, for an intra (I) picture is set equal to said average quantization parameter of said current picture.

15. The machine-readable storage according to claim 12 , wherein said average quantization parameter is:

if said reference picture is a predicted (P) picture, equal to one or both of: said average quantization parameter of said current picture of most recent reference P picture and/or said average quantization parameter of said current picture of most recent reference intra (I) picture; and

if said reference picture is a bidirectional (B) picture, equal to an arithmetic average of said average quantization parameter of said current picture values for two most recently decoded reference pictures.

16. The machine-readable storage according to claim 12 , wherein said weighted sum is one or more of: a function of ratio of said total number of said intra macroblocks to said total number of said macroblocks, a function of ratio of said total number of said inter macroblocks to said total number of said macroblocks, a function of said sum of said low order matrix elements, and/or a function of said sum of said high order matrix elements.

17. The machine-readable storage according to claim 10 , wherein said at least one code section comprises code for controlling said digital noise reduction (DNR) module by said host processor via received user controlled information and at least a portion of said received plurality of picture level parameters.

18. The machine-readable storage according to claim 10 , wherein said at least one code section comprises code for reducing one or both of: mosquito noise artifacts and/or block noise artifacts based on at least a portion of said generated plurality of picture level parameters.

19. A system for image processing, the system comprising:

one or more processors that are operable to receive a plurality of picture level parameters generated by a video decoder, wherein said plurality of picture level parameters comprises an average quantization parameter of a current picture and an average quantization parameter of a reference picture;

said one or more processors are operable to calculate an adjusted average quantization parameter as a weighted sum of said average quantization parameter of said current picture and said average quantization parameter of said reference picture; and

said one or more processors are operable to control a digital noise reduction (DNR) module that processes input video based on said calculated adjusted average quantization parameter.

20. The system according to claim 19 , wherein said plurality of picture level parameters comprises one or more of: a reference picture flag, total number of intra macroblocks, total number of inter macroblocks, total number of macroblocks, a sum of low order matrix elements of an intra quantization matrix in zig-zag scan order, a sum of high order matrix elements of said intra quantization matrix in zig-zag scan order, a sum of low order matrix elements of an inter quantization matrix in zig-zag scan order, and/or a sum of high order matrix elements of said inter quantization matrix in zig-zag scan order.

21. The system according to claim 20 , wherein said one or more processors are operable to calculate said average quantization parameter of said current picture as a ratio of a sum of quantization parameters of each said macroblock to said total number of said macroblocks.

22. The system according to claim 21 , wherein said one or more processors are operable to set said quantization parameter of skipped macroblocks equal to a quantization parameter of non-skipped macroblocks.

23. The system according to claim 21 , wherein said one or more processors are operable to set said average quantization parameter of said reference picture for an intra (I) picture to be equal to said average quantization parameter of said current picture.

24. The system according to claim 21 , wherein said average quantization parameter is:

if said reference picture is a predicted (P) picture, equal to one or both of: said average quantization parameter of said current picture of most recent reference P picture and/or said average quantization parameter of said current picture of most recent reference intra (I) picture; and

if said reference picture is a bidirectional (B) picture, equal to an arithmetic average of said average quantization parameter of said current picture values for two most recently decoded reference pictures.

25. The system according to claim 21 , wherein said weighted sum is one or more of: a function of ratio of said total number of said intra macroblocks to said total number of said macroblocks, a function of ratio of said total number of said inter macroblocks to said total number of said macroblocks, a function of said sum of said low order matrix elements, and/or a function of said sum of said high order matrix elements.

26. The system according to claim 19 , wherein said one or more processors are operable to control said digital noise reduction (DNR) module via received user controlled information and at least a portion of said received plurality of picture level parameters.

27. The system according to claim 19 , wherein said digital noise reduction module (DNR) reduces one or both of: mosquito noise artifacts and/or block noise artifacts based on at least a portion of said generated plurality of picture level parameters.