Method for adaptively selecting filters to interpolate video data

1. A method for adaptively selecting filters to interpolate video data, comprising:

receiving the video data encoded by a 4:2:2 sampling format;

adaptively selecting filters to convert the color components in the video data to a 4:4:4 sampling format based on differences of luminance and color components among a pixel P, a pixel P_ 1 on the left of the pixel P, a pixel P_ 2 on the left of the pixel P_ 1 , a pixel P_ 3 on the left of the pixel P_ 2 , a pixel P 1 on the right of the pixel P, a pixel P 2 on the right of the pixel P 1 , and a pixel P 3 on the right of the pixel P 2 in the video data, comprising:

determining whether a difference between the color components of the pixels P 1 and P_ 1 is larger than a first threshold;

if larger than the first threshold, determining whether a difference between the luminances of the pixels P and P 1 is smaller than a second threshold;

if smaller than the second threshold, using a full left enhancing filter or a full right enhancing filter to convert the video data;

if not smaller than the second threshold, using a center-to-left enhancing filter or a center-to-right enhancing filter to convert the video data;

if not larger than the first threshold, determining whether a difference between the luminances of the pixels P and P_ 1 or the pixels P and P 1 is larger than a third threshold;

if larger than the third threshold, determining whether a difference between a variation of the luminances among pixels P_ 2 to P and a variation of the luminances among pixels P to P 2 is larger than a forth threshold;

if larger than the forth threshold, using the center-to-left enhancing filter or the center-to-right enhancing filter to convert the video data;

if not larger than the forth threshold, using a center-to-left loosing filter or a center-to-right loosing filter to convert the video data; and

if not larger than the third threshold, using a central averaging filter to convert the video.

2. The method for adaptively selecting filters to interpolate video data according to claim 1 , the step of using the full left enhancing filter or the full right enhancing filter to convert the video data comprises:

if the differences between each two of the pixels P_ 3 , P_ 2 , P_ 1 , and P are all smaller than a fifth threshold, using the full left enhancing filter to convert the video data; and

if the differences between each two of the pixels P, P 1 , P 2 , and P 3 are not all smaller than the fifth threshold, using the full right enhancing filter to convert the video data.

3. The method for adaptively selecting filters to interpolate video data according to claim 1 , the step of using the center-to-left enhancing filter or the center-to-right enhancing filter to convert the video data comprises:

if a variation of the luminances among the pixels P_ 2 to P is smaller than a variation of the luminances among the pixels P to P 2 , using the center-to-left enhancing filter to convert the video data; and

if a variation of the luminance component among pixels P 2 to P is smaller than a variation of the luminance component among the pixels P_ 2 to P, using the center-to-right enhancing filter to convert the video data.

4. The method for adaptively selecting filters to interpolate video data according to claim 1 , the step of using a center-to-left loosing filter or a center-to-right loosing filter to convert the video data comprises:

if a variation of the luminances among the pixels P_ 1 to P is smaller than a variation of the luminances among the pixels P 1 to P, using the center-to-left loosing filter to convert the video data; and

if a variation of the luminances among the pixels P 1 to P is smaller than a variation of the luminances among the pixels P_ 1 to P, using the center-to-right loosing filter to convert the video data.

5. The method for adaptively selecting filters to interpolate video data according to claim 1 , the step of converting the video data from the 4:2:2 sampling format to the 4:4:4 sampling format comprises:

calculating a summation of the pixels P_ 3 , P_ 1 , P 1 , P 3 respectively multiplied by a corresponding weight of the filter and dividing the same by a summation of the weights so as to obtain the color component of the pixel P.

6. The method for adaptively selecting filters to interpolate video data according to claim 5 , wherein the weights of the fill left enhancing filter corresponding to the pixels P_ 3 , P_ 1 , P 1 , P 3 are 1, 7, 0, 0, respectively.

7. The method for adaptively selecting filters to interpolate video data according to claim 5 , wherein the weights of the fill right enhancing filter corresponding to the pixels P_ 3 , P_ 1 , P 1 , P 3 are 0, 0, 7, 1, respectively.

8. The method for adaptively selecting filters to interpolate video data according to claim 5 , wherein the weights of the center-to-left enhancing filter corresponding to the pixels P_ 3 , P_ 1 , P 1 , P 3 are 0, 7, 1, 0, respectively.

9. The method for adaptively selecting filters to interpolate video data according to claim 5 , wherein the weights of the center-to-right enhancing filter corresponding to the pixels P_ 3 , P_ 1 , P 1 , P 3 are 0, 1, 7, 0, respectively.

10. The method for adaptively selecting filters to interpolate video data according to claim 5 , wherein the weights of the center-to-left loosing filter corresponding to the pixels P_ 3 , P_ 1 , P 1 , P 3 are 0, 3, 1, 0, respectively.

11. The method for adaptively selecting filters to interpolate video data according to claim 5 , wherein the weights of the center-to-right loosing filter corresponding to the pixels P_ 3 , P_ 1 , P 1 , P 3 are 0, 1, 3, 0, respectively.

12. The method for adaptively selecting filters to interpolate video data according to claim 5 , wherein the weights of the central averaging filter corresponding to the pixels P_ 3 , P_ 1 , P 1 , P 3 are 0, 1, 1, 0, respectively.