Loop filter and loop filtering method

Claim 2

A loop filtering method for reducing blocking effect and ringing noise of image data, the loop filtering method comprising the steps of: (a) generating a binary edge map by comparing a value obtained by operating each pixel of the image data using a predetermined one-dimensional gradient operator with a predetermined threshold value; (b) applying a filter window with a predetermined size to the generated binary edge map to determine whether edge information exists within the filter window; (c) filtering the pixel value of the corresponding filter window by a pixel using predetermined first weights to generate a new pixel value if it is determined in the step (b) that edge information does not exist; and (d) filtering the pixel value of the corresponding filter window by a pixel using predetermined second weights to generate a new pixel value if it is determined in the step (b) that edge information exists, wherein the filtering is not performed in the step (d) if the pixel at the center of the filter window is edge information.

Claim 3

A loop filtering method as claimed in claim 1, wherein the filter window is 3×3 in size.

Claim 4

A loop filtering method as claimed in claim 1, wherein the one-dimensional gradient operator used in the step (a) includes: a horizontal gradient operator of 1×2 size, having a weight of (1,-1); and a vertical gradient operator of 2×1 size, having a weight of (1,-1).

Claim 5

A loop filtering method as claimed in claim 3, wherein the step (a) includes the sub-steps of: (a-1) receiving image data in a predetermined block unit; (a-2) performing a gradient operation with respect to each pixel of the input image block using the one-dimensional gradient operator; and (a-3) generating the binary edge map information through a comparison with a predetermined first threshold value if the gradient operated pixel belongs to a predetermined area near a block boundary, or through a comparison with a predetermined second threshold value, if otherwise.

Claim 6

A loop filtering method as claimed in claim 1, wherein said threshold values comprise a first threshold value that is greater than a second threshold value.

Claim 7

A loop filtering method as claimed in claim 1, wherein the one-dimensional horizontal and vertical gradient operator is applied with respect to each pixel of an image block, and an OR operation is performed with respect to each binary value obtained through one-dimensioal horizontal and vertical gradient operations.

Claim 8

A loop filtering method as claimed in claim 1, wherein the block is of 16×16 size and the loop filtering is performed in a block unit of 8×8 size.

Claim 9

A loop filtering method as claimed in claim 1, wherein said predetermined one-dimensioal gradient operator uses at least two predetermined threshold values.

Claim 10

A loop filtering method for reducing a blocking effect and ringing noise of image data, the loop filtering method comprising the steps of: (a) generating a binary edge map by comparing a value obtained by operating each pixel of the image data using a predetermined one-dimensional gradient operator with a predetermined threshold value by: (a-1) receiving image data in a predetermined block unit; (a-2) performing a gradient operation with respect to each pixel of the input image block using the one-dimensional gradient operator; and (a-3) generating the binary edge map information through a comparison with a predetermined first threshold value if the gradient operated pixel belongs to a predetermined area near a block boundary, or through a comparison with a predetermined second threshold value, if otherwise: (b) applying a filter window with a predetermined size to the generated binary edge map to check whether an edge information exists within the filter window; (c) filtering the pixel value of the corresponding filter window by a pixel using predetermined first weights to generate a new pixel value if it is determined in the step (b) that the edge information does not exist; and (d) filtering the pixel value of the corresponding filter window by a pixel using predetermined second weights to generate a new pixel value if it is determined in the step (b) that the edge information exists, wherein the filtering is not performed in the step (d) if the pixel at the center of the filter window is edge information.

Claim 11

A loop filtering method for reducing a blocking effect and ringing noise of image data, the loop filtering method comprising the steps of: (a) generating a binary edge map by comparing a value obtained by operating each pixel of the image data using a predetermined one-dimensional gradient operator with a predetermined threshold value by: (a-1) receiving image data in a predetermined block unit; (a-2) performing a gradient operation with respect to each pixel of the input image block using the one-dimensional gradient operator; and (a-3) generating the binary edge map information through a comparison with a predetermined first threshold value if the gradient operated pixel belongs to a predetermined area near a block boundary, or through a comparison with a predetermined second threshold value, if otherwise: (b) applying a filter window with a predetermined size to the generated binary edge map to check whether an edge information exists within the filter window; (c) filtering the pixel value of the corresponding filter window by a pixel using predetermined first weights to generate a new pixel value if it is determined in the step (b) that the edge information does not exist; and (d) filtering the pixel value of the corresponding filter window by a pixel using predetermined second weights to generate a new pixel value if it is determined in the step (b) that the edge information exists, wherein the filtering is not performed in the step (d) if the pixel at the center of the filter window is edge information.

Claim 12

A loop filtering method as claimed in claim 9, wherein the first threshold value is greater than the second threshold value in the step (a-3).

Claim 13

A loop filtering method as claimed in claim 9, wherein the one-dimensional horizontal and vertical gradient operations are performed with respect to each pixel of the image block in the step (a-2), and an OR operation is performed with respect to each binary value obtained through the one-dimensional horizontal and vertical gradient operations in the step (a-3).

Claim 14

A loop filtering method as claimed in claim 9, wherein the block is of 16×16 size and the loop filtering is performed in a block unit of 8×8 size.

Claim 15

A loop filter comprising: an image storing unit for temporarily storing image data; a gradient operator for performing an one-dimensional gradient operation in horizontal and vertical directions using a one-dimensional gradient operator to find out edge pixels by receiving the image data in a block unit having a predetermined size from the image storing unit; an area determining unit for determining whether the pixels which have been horizontal or vertical one-dimensional gradient operated belong to a boundary area or a global area; a boundary area binary edge information generator for generating edge information if a value obtained by the one-dimensional gradient operation is greater than a predetermined first threshold value, or non-edge information, if otherwise, when it is determined by the area determining unit that the image data passed through the horizontal or vertical one-dimensional gradient operator belongs to the boundary area; a global area binary edge information generator for generating edge information if a value obtained by the one-dimensional gradient operation is greater than a predetermined second threshold value, or non-edge information, if otherwise, when it is determined by the area determining unit that the image data passed through the horizontal or vertical one-dimensioal gradient operator belongs to the global area; an OR-processing unit for performing an OR operation to generate binary edge map information with respect to information output via the area determining unit and the boundary area binary edge information generator, and information output via the area determining unit and the global area binary edge information generator, after the vertical and horizontal one-dimensioal gradient operations are performed with respect to a pixel; a filter determiner for storing the binary edge map information output from the OR-processing unit and classifying the input image data into one of an edge area including at least one piece of edge information and a homogeneous area without having edge information according to the binary edge map information; an average filter for performing a predetermined average filtering with respect to the pixels of a filter window classified into the homogeneous area by the filter determiner; and a weighted filter for performing a predetermined weighted filtering with respect to the pixels of a filter window classified into the edge area by the filter determiner.

Claim 16

A loop filter comprising: an image storing unit for temporarily storing image data; a gradient operator for performing an one-dimensional gradient operation in horizontal and vertical directions using a one-dimensional gradient operator to find out edge pixels by receiving the image data in a block unit having a predetermined size from the image storing unit; an area determining unit for determining whether the pixels which have been horizontal or vertical one-dimensional gradient operated belong to a boundary area or a global area; a boundary area binary edge information generator for generating edge information if a value obtained by the one-dimensional gradient operation is greater than a predetermined first threshold value, or non-edge information, if otherwise, when it is determined by the area determining unit that the image data passed through the horizontal or vertical one-dimensional gradient operator belongs to the boundary area; a global area binary edge information generator for generating edge information if a value obtained by the one-dimensional gradient operation is greater than a predetermined second threshold value, or non-edge information, if otherwise, when it is determined by the area determining unit that the image data passed through the horizontal or vertical one-dimensioal gradient operator belongs to the global area; an OR-processing unit for performing an OR operation to generate binary edge map information with respect to information output via the area determining unit and the boundary area binary edge information generator, and information output via the area determining unit and the global area binary edge information generator, after the vertical and horizontal one-dimensioal gradient operations are performed with respect to a pixel; a filter determiner for storing the binary edge map information output from the OR-processing unit and classifying the input image data into one of an edge area including at least one piece of edge information and a homogeneous area without having edge information according to the binary edge map information; an average filter for performing a predetermined average filtering with respect to the pixels of a filter window classified into the homogeneous area by the filter determiner; and a weighted filter for performing a predetermined weighted filtering with respect to the pixels of a filter window classified into the edge area by the filter determiner.