Methods and systems for compensating reconstructed image frames

1 . A method, comprising:

obtaining a reconstructed image frame;

determining a classification manner of pixels in the reconstructed image frame, wherein the classification manner includes a first classification mode and a second classification mode;

classifying the pixels in the reconstructed image frame into at least one first category based on the first classification mode, wherein the first classification mode includes a first classification sub-mode, the pixels in the reconstructed image frame are classified into 17 categories based on the first classification sub-mode, and a formula of the first classification sub-mode is shown below:

Initial: Cl=0;

for |k 1 −i|≤1, |k 2 −j|≤1:

if Y (k 1 ,k 2 ) >Y (i,j) :

Cl+=1

else if Y (k 1 ,k 2 ) <Y (i,j) :

 Cl+=−1,

where different values of Cl refer to different categories, Y (i,j) refers to a pixel value of a current pixel (i, j), Y (k 1 ,k 2 ) refers to a pixel value of a surrounding pixel (k 1 , k 2 ) of the current pixel, a value range of Cl in the first classification sub-mode is [−8, 8], which includes 17 values each of which corresponds to a category;

classifying the pixels in the reconstructed image frame into at least one second category based on the second classification mode;

determining compensation values of the pixels in the reconstructed image frame based on the at least one first category and the at least one second category; and

obtaining a compensated image frame by compensating the reconstructed image frame based on the compensation values of the pixels in the reconstructed image frame.

2 . The method of claim 1 , wherein the method further includes:

determining a secondary classification manner of the pixels in the compensated image frame, the secondary classification manner is the first classification mode;

determining a secondary compensation value of each category of pixels in the compensated image frame that are classified based on the secondary classification manner; and

compensating the compensated image frame based on the secondary compensation value of the each category of pixels in the compensated image frame.

3 . The method of claim 2 , wherein a code stream of the reconstructed image frame or the compensated image frame includes a syntactic element, the syntactic element at least includes a plurality of parameter sets, a filter flag, and preset serial numbers of the plurality of parameter sets associated with the reconstructed image frame or the compensated image frame, the plurality of parameter sets are provided to compensate different coding units in the reconstructed image frame or the compensated image frame, and the method further includes:

determining at least one parameter set, each set of the at least one parameter set including one classification manner of the classification manner or the secondary classification manner and a compensation value corresponding to the classification manner; and

transmitting the classification manner and the corresponding compensation value in the each set of the at least one parameter set based on a preset serial number of the each set of the at least one parameter set, wherein parameter sets of coding units at a same position in the reconstructed image frame or the compensated image frame with different color components are indicated by a same serial number.

4 . The method of claim 3 , wherein the at least one parameter set is applied to a chrominance component of the reconstructed image frame or the compensated image frame.

5 . The method of claim 3 , wherein the method further includes:

determining, based on an opening situation of a filtering operation in a syntactic representation associated with the reconstructed image frame or the compensated image frame, whether to apply the at least one parameter set to the reconstructed image frame or the compensated image frame.

6 . The method of claim 1 , wherein the first classification mode further includes a second classification sub-mode or a third classification sub-mode, the pixels in the reconstructed image frame are classified into 9 categories based on the second classification sub-mode, and the pixels in the reconstructed image frame are classified into 1 category based on the third classification sub-mode.

7 . The method of claim 6 , wherein a formula of the second classification sub-mode is shown below

Initial: Cl=0:

for |k 1 −i|≤1, |k 2 −j|≤1:

if Y (k 1 ,k 2 ) >Y (i,j) :

Cl+=1,

where a value range of Cl in the second classification sub-mode is [0, 8], which includes 9 values each of which corresponds to a category.

8 . The method of claim 7 , wherein

if the first classification mode is the first classification sub-mode or the second classification sub-mode, the second classification mode includes one of 16 classification sub-modes,

if the first classification mode is the third classification sub-mode, the second classification mode includes one of 96 classification sub-modes; and

pixel values of the pixels in the reconstructed image frame are divided into multiple ranges based on each classification sub-mode in the second classification mode.

9 . The system of claim 1 , wherein the first classification mode is configured to classify the pixels in the reconstructed image frame based on the current pixel in the reconstructed image frame and surrounding pixels of the current pixel, and the second classification mode is configured to classify the pixels in the reconstructed image frame based on pixel values of the pixels in the reconstructed image frame.

10 . The system of claim 1 , wherein the first classification mode further includes a second classification sub-mode, the pixels in the reconstructed image frame are classified into 9 categories based on the second classification sub-mode, and a formula of the second classification sub-mode is shown below

Initial: Cl=0;

for |k 1 −i|≤1, |k 2 −j|≤1:

if Y (k 1 ,k 2 ) >Y (i,j) :

Cl+=1,

where a value range of Cl in the second classification sub-mode is [0, 8], which includes 9 values each of which corresponds to a category.

11 . A system, comprising:

at least one storage device including a set of instructions; and

at least one processor in communication with the at least one storage device, wherein when executing the set of instructions, the at least one processor is directed to perform operations including:

obtaining a reconstructed image frame;

determining a classification manner of pixels in the reconstructed image frame, wherein the classification manner includes a first classification mode and a second classification mode;

determining a compensation value of each category of pixels in the reconstructed image frame that are classified based on the classification manner, wherein the first classification mode includes a first classification sub-mode, the pixels in the reconstructed image frame are classified into 17 categories based on the first classification sub-mode, and a formula of the first classification sub-mode is shown below:

Initial: Cl=0;

for |k 1 −i|≤1, |k 2 −j|≤1:

if Y (k 1 ,k 2 ) >Y (i,j) :

 Cl+=1

else if Y (k 1 ,k 2 ) <Y (i,j) :

 Cl+=−1,

where different values of Cl refer to different categories, Y (i,j) refers to a pixel value of a current pixel (i, j), Y (k 1 ,k 2 ) refers to a pixel value of a surrounding pixel (k 1 , k 2 ) of the current pixel, a value range of Cl in the first classification sub-mode is [−8, 8], which includes 17 values each of which corresponds to a category; and

obtaining a compensated image frame by compensating the reconstructed image frame based on the compensation value of each category of pixels in the reconstructed image frame.

12 . The system of claim 11 , wherein the operations further include:

determining a secondary classification manner of the pixels in the compensated image frame, the secondary classification manner is the first classification mode;

determining a secondary compensation value of each category of pixels in the compensated image frame that are classified based on the secondary classification manner; and

compensating the compensated image frame based on the secondary compensation value of the each category of pixels in the compensated image frame.

13 . The system of claim 11 , wherein a code stream of the reconstructed image frame or the compensated image frame includes a syntactic element, the syntactic element at least includes a plurality of parameter sets, a filter flag, and preset serial numbers of the plurality of parameter sets associated with the reconstructed image frame or the compensated image frame, the plurality of parameter sets are provided to compensate different coding units in the reconstructed image frame or the compensated image frame, and the operations further includes:

determining at least one parameter set, each set of the at least one parameter set including one classification manner of the classification manner or the secondary classification manner and a compensation value corresponding to the classification manner; and

transmitting the classification manner and the corresponding compensation value in the each set of the at least one parameter set based on a preset serial number of the each set of the at least one parameter set, wherein parameter sets of coding units at a same position in the reconstructed image frame or the compensated image frame with different color components are indicated by a same serial number.

14 . The system of claim 13 , wherein the at least one parameter set is applied to a chrominance component of the reconstructed image frame or the compensated image frame.

15 . The system of claim 13 , wherein the operations further include:

determining, based on an opening situation of a filtering operation in a syntactic representation associated with the reconstructed image frame or the compensated image frame, whether to apply the at least one parameter set to the reconstructed image frame or the compensated image frame.

16 . The system of claim 11 , wherein the classification manner is a combination of the first classification mode and the second classification mode, and the first classification mode and the second classification mode are combined using a Cartesian product algorithm.

17 . The system of claim 11 , wherein the first classification mode further includes a second classification sub-mode or a third classification sub-mode, the pixels in the reconstructed image frame are classified into 9 categories based on the second classification sub-mode, and the pixels in the reconstructed image frame are classified into 1 category based on the third classification sub-mode.

18 . The system of claim 17 , wherein a formula of the second classification sub-mode is shown below

Initial: Cl=0;

for |k 1 −i|≤1, |k 2 −j|≤1:

if Y (k 1 ,k 2 ) >Y (i,j) :

Cl+=1,

where a value range of Cl in the second classification sub-mode is [0, 8], which includes 9 values each of which corresponds to a category.

19 . The system of claim 18 , wherein

if the first classification mode is the first classification sub-mode or the second classification sub-mode, the second classification mode includes one of 16 classification sub-modes,

if the first classification mode is the third classification sub-mode, the second classification mode includes one of 96 classification sub-modes; and

pixel values of the pixels in the reconstructed image frame are divided into multiple ranges based on each classification sub-mode in the second classification mode.

20 . A non-transitory computer readable medium, comprising executable instructions that, when executed by at least one processor, direct the at least one processor to perform a method, the method comprising:

obtaining a reconstructed image frame;

determining a classification manner of pixels in the reconstructed image frame, wherein the classification manner includes a first classification mode and a second classification mode;

determining a compensation value of each category of pixels in the reconstructed image frame that are classified based on the classification manner, wherein the first classification mode includes a first classification sub-mode, the pixels in the reconstructed image frame are classified into 17 categories based on the first classification sub-mode, and a formula of the first classification sub-mode is shown below:

Initial: Cl=0;

for |k 1 −i|≤1, |k 2 −j|≤1:

if Y (k 1 ,k 2 ) >Y (i,j) :

Cl+=1

else if Y (k 1 ,k 2 ) <Y (i,j) :

 Cl+=−1,

where different values of Cl refer to different categories, Y (i,j) refers to a pixel value of a current pixel (i, j), Y (k 1 ,k 2 ) refers to a pixel value of a surrounding pixel (k 1 , k 2 ) of the current pixel, a value range of Cl in the first classification sub-mode is [−8, 8], which includes 17 values each of which corresponds to a category; and

obtaining a compensated image frame by compensating the reconstructed image frame based on the compensation value of each category of pixels in the reconstructed image frame.