Transform artifact reduction process
First Claim
1. A transform artifact reduction method for decoded video pixel data, said method comprising:
- digitally noise reducing a block of pixels to obtain a digitally noise reduced block of pixels; and
filtering said digitally noise reduced block of pixels using a spatially-variant filter wherein said filtering further comprisesadaptively identifying each pixel in said digitally noise reduced block of pixels as one of an edge pixel and a non-edge pixel;
assigning a pixel in said digitally noise reduced block of pixels one classification in a plurality of edge classifications using said edge pixel and said non-edge pixel identifications for pixels in a window about said pixel, said assigning said pixel in said digitally noise reduced block of pixels one classification in said plurality of edge classifications using said edge pixel and said non-edge pixel classifications for pixels in a window about said pixel further comprisinggenerating an edge map for said digitally noise reduced block of pixels wherein in said edge map, an edge flag for each pixel in said digitally noise reduced block of pixels is configured to identify each pixel as one of said edge pixel and said non-edge pixel; and
adaptively filtering each pixel in said digitally noise reduced block of pixels based upon said assigned classification to reduce transform artifacts in a video image.
4 Assignments
0 Petitions
Accused Products
Abstract
A post-processor for a decoded video sequence includes a digital noise reduction unit and an artifact reduction unit which significantly reduce blocking artifacts and mosquito noise in a video image. The post-processor uses both temporal and edge characteristics of the video image to enhance the displayed image. The post-processor operates on a current frame of pixel data using information from the immediately preceding post-processed frame that is stored in a frame memory of the post-processor. The post-processor first identifies texture and fine detail areas in the image. The post-processor uses artifact reduction only on portions of the image that are not part of an edge, and are not part of a texture or fine detail area. Since artifact reduction is not utilized on these areas, the post-processed image is not softened in regions where it is easily noticed by the human eye.
-
Citations
111 Claims
-
1. A transform artifact reduction method for decoded video pixel data, said method comprising:
-
digitally noise reducing a block of pixels to obtain a digitally noise reduced block of pixels; and filtering said digitally noise reduced block of pixels using a spatially-variant filter wherein said filtering further comprises adaptively identifying each pixel in said digitally noise reduced block of pixels as one of an edge pixel and a non-edge pixel; assigning a pixel in said digitally noise reduced block of pixels one classification in a plurality of edge classifications using said edge pixel and said non-edge pixel identifications for pixels in a window about said pixel, said assigning said pixel in said digitally noise reduced block of pixels one classification in said plurality of edge classifications using said edge pixel and said non-edge pixel classifications for pixels in a window about said pixel further comprising generating an edge map for said digitally noise reduced block of pixels wherein in said edge map, an edge flag for each pixel in said digitally noise reduced block of pixels is configured to identify each pixel as one of said edge pixel and said non-edge pixel; and adaptively filtering each pixel in said digitally noise reduced block of pixels based upon said assigned classification to reduce transform artifacts in a video image. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37)
-
-
38. A transform artifact reduction method for decoded video pixel data, said method comprising:
-
digitally noise reducing a block of pixels to obtain a digitally noise reduced block of pixels; setting a replenished block flag to a predetermined state upon determining that said digitally noise reduced block of pixels is a replenished block; identifying each pixel in said digitally noise reduced block of pixels as one of an edge pixel and not edge pixel; classifying each pixel that is identified as said not edge pixel as one of a texture pixel and not texture pixel; setting an edge block flag to a predetermined state upon said digitally noise reduced block of pixels having a predetermined number of pixels classified as edge pixels; setting a texture block flag to a predetermined state upon said digitally noise reduced block of pixels having a predetermined number of pixels classified as texture pixels; and filtering said digitally noise reduced block of pixels using a spatially-variant filter when any one of said block texture flag, said block edge flag, and said block replenish flag is set to said predetermined state. - View Dependent Claims (39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76)
-
-
77. An edge detection method for transform artifact reduction comprising:
-
comparing a pixel gradient for a pixel with a luminance threshold; setting an edge flag for said pixel in an edge map upon said pixel gradient being greater than said threshold; repeating said comparing and setting processes for each pixel in a block of pixels. - View Dependent Claims (78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98)
-
-
99. A spatially-variant filtering method to reduce transform artifacts comprising:
-
assigning a pixel in a block of pixels one classification in a plurality of edge classifications using edge flags in an edge map for pixels in a window about said pixel wherein an edge flag for a pixel is set in said edge map to indicate said pixel is in an edge of a video image; and filtering each pixel in said block of pixels based upon said assigned classification to reduce transform artifacts in said video image. - View Dependent Claims (100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111)
-
Specification