De-interlacing processing method and device, and computer-readable storage medium

1. A de-interlacing processing method, comprising:

acquiring image content characteristic information of a pixel point to be interpolated; and

determining according to the image content characteristic information whether a de-interlacing algorithm based on motion adaptive or a de-interlacing algorithm based on motion compensation is adopted to perform de-interlacing processing,

wherein the step of performing de-interlacing processing using the de-interlacing algorithm based on motion compensation comprising:

acquiring a first value of the pixel point to be interpolated calculated by the de-interlacing algorithm based on motion adaptive;

acquiring a second value of the pixel point to be interpolated calculated by a de-interlacing algorithm based on forward motion compensation;

acquiring a third value of the pixel point to be interpolated calculated by a de-interlacing algorithm based on backward motion compensation; and

performing median filtering on the first value, the second value and the third value, and taking a result of the median filtering as a value of the pixel point to be interpolated calculated by the de-interlacing algorithm based on motion compensation.

2. The method according to claim 1 , wherein the image content characteristic information comprises at least one of motion vector information, inter-field matching information, inter-frame matching information, edge direction information, or drawing-feathering intensity information.

3. The method according to claim 2 , wherein the step of determining according to the image content characteristic information whether the de-interlacing algorithm based on motion adaptive or the de-interlacing algorithm based on motion compensation is adopted to perform de-interlacing processing comprises:

when one of the following conditions is satisfied, adopting the de-interlacing algorithm based on motion adaptive to perform de-interlacing processing; and

when none of the following conditions is satisfied, adopting the de-interlacing algorithm based on motion compensation to perform de-interlacing processing;

wherein, the conditions comprise at least one of:

a motion vector of the pixel point to be interpolated is 0;

an absolute value of any component of the motion vector of the pixel point to be interpolated is greater than a preset number of integer pixels;

the inter-field matching information of the pixel point to be interpolated is greater than a first inter-field matching threshold, and the inter-frame matching information of the pixel point to be interpolated is greater than a first inter-frame matching threshold;

the inter-field matching information of the pixel point to be interpolated is greater than a second inter-field matching threshold, and the inter-frame matching information of the pixel point to be interpolated is greater than a second inter-frame matching threshold;

the inter-field matching information of the pixel point to be interpolated is greater than the inter-frame matching information, and the inter-frame matching information of the pixel point to be interpolated is greater than a third inter-frame matching threshold;

first edge direction information of the pixel point to be interpolated in a current field indicates a non-horizontal direction;

second edge direction information of the pixel point to be interpolated in a last field relative to the current field indicates a non-horizontal direction;

third edge direction information of the pixel point to be interpolated in a next field relative to the current field indicates a non-horizontal direction; or

the drawing-feathering intensity information of the pixel point to be interpolated is greater than a drawing intensity threshold.

4. The method according to claim 3 , wherein

the first inter-field matching threshold ranges from 0 to 300;

the second inter-field matching threshold ranges from 200 to 600;

the first inter-frame matching threshold ranges from 200 to 700;

the second inter-frame matching threshold ranges from 100 to 500; and

the third inter-frame matching threshold ranges from 0 to 300.

5. The method according to claim 2 , wherein the inter-field matching information is determined with the following method: selecting a first pixel block composed of M pixels in a last row relative to a current row where the pixel point to be interpolated is located and a second pixel block composed of M pixels in a next row relative to the current row from a current field, determining matched blocks of the first pixel block and the second pixel block in a last field relative to the current field, obtaining a first sum of absolute differences according to the first pixel block and the matched block thereof, and a second sum of absolute differences according to the second pixel block and the matched block thereof, and taking a maximum value of the first sum of absolute differences and the second sum of absolute differences as the inter-field matching information, wherein a central position of the M pixels in the last row and a central position of the M pixels in the next row are in a same column as the pixel point to be interpolated, and M is a positive integer.

6. The method according to claim 2 , wherein the inter-frame matching information is determined with the following method: forming a current frame from a current field where the pixel point to be interpolated is located and a next field relative to the current field, selecting a pixel block in a preset size with the pixel point to be interpolated taken as a center, and matching the pixel block with a pixel block in a last frame relative to the current frame to obtain the inter-frame matching information.

7. The method according to claim 3 , wherein the drawing-feathering intensity information is determined with at least one of the following methods:

an M×N pixel block with the pixel point to be interpolated as a center is taken from a synthesized frame composed of a current field and a last field relative to the current field; for the M×N pixel block, an accumulated value of differences between pixel values corresponding to adjacent pixel points of heterogeneous field in a column and an accumulated value of differences between pixel values corresponding to adjacent pixel points of homogeneous field in the column are calculated column by column; and a total number of the columns satisfying same_parity_value[i]*coeff<dif_parity_value[i] is obtained, wherein same_parity_value[i] is an accumulated value of differences between pixel values corresponding to adjacent pixel points of homogeneous field in Column i in the M×N pixel block, dif_parity_value[i] is an accumulated value of differences between pixel values corresponding to adjacent pixel points of heterogeneous field in Column i in the M×N pixel block, i=0˜N−1, and coeff is greater than 0; and, when the total number of the columns satisfying same_parity_value[i]*coeff<dif_parity_value[i] is greater than a first threshold, the drawing-feathering intensity information is greater than the drawing intensity threshold;

for the M×N pixel block, an accumulated value of pixels in each row is acquired to obtain M accumulated values sum_value_ver[l], where l=0˜M−1; (M−2) thresholds thr[j]=|(sum_value_ver[j+1]*2−sum_value_ver[j]−sum_value_ver[j+2]| are obtained according to the M accumulated values, where j=0˜M−3, M is greater than or equal to 3, and N is greater than or equal to 1;

differences between pixel values corresponding to adjacent pixel points of homogeneous field and differences between pixel values corresponding to adjacent pixel points of heterogeneous field are calculated point by point:

frame_diff[ j ][ i ]=| Y [ j ][ i ]− Y [ j+ 2][ i ]|

field_diff[ j ][ i ]=| Y [ j ][ i ]− Y [ j+ 1][ i ]|

where i=0˜N−1; when a total number of columns satisfying

t

⁢

h

⁢

r

[

j

]

>

frame_diff

[

j

]

[

i

]

*

factor

⁢

1

⁢

thr

[

j

]

<

field_diff

[

j

]

[

i

]

*

factor

⁢

2

is greater than a second threshold, the drawing-feathering intensity information is greater than the drawing intensity threshold, where factor1 and factor2 are regulatory factors for drawing detection and are greater than 0, Y[j][i] is a value of a pixel point in Row j and Column i in the M×N pixel block, frame_diff[j][i] is a difference between pixel values corresponding to adjacent pixel points of homogeneous field for the pixel point in Row j and Column i in the M×N pixel block, and field_diff[j][i] is a difference between pixel values corresponding to adjacent pixel points of heterogeneous field for the pixel point in Row j and Column i in the M×N pixel block.

8. A de-interlacing processing device, comprising a memory storing a program, and a processor, wherein the method according to claim 1 is implemented when the program is read and executed by the processor.

9. A non-transitory computer-readable storage medium storing one or more programs, wherein the one or more programs are executable by one or more processors to perform the method according to claim 1 .

10. A de-interlacing processing device, comprising a memory storing a program, and a processor, wherein the method according to claim 2 is implemented when the program is read and executed by the processor.

11. A de-interlacing processing device, comprising a memory storing a program, and a processor, wherein the method according to claim 3 is implemented when the program is read and executed by the processor.

12. A de-interlacing processing device, comprising a memory storing a program, and a processor, wherein the method according to claim 4 is implemented when the program is read and executed by the processor.

13. A de-interlacing processing device, comprising a memory storing a program, and a processor, wherein the method according to claim 5 is implemented when the program is read and executed by the processor.