Systems and methods for evaluating video quality

1. A processor-implemented method for evaluating quality of a video sequence, the method comprising:

receiving a video sequence including one or more image frames, the video sequence being associated with a video processing system;

generating, using a data processor, a frame quality factor for each image frame in the video sequence based on a comparison between the image frame and a reference image frame;

generating, using the data processor, a fluctuation value and a weight value for the image frames, the fluctuation value of the image frames indicating a variation of the frame quality factors of the image frames, the weight value of the image frames being determined based on an average of the frame quality factors of the image frames; and

outputting, using the data processor, a video quality factor based on the frame quality factors of the image frames, the fluctuation value of the image frames, and the weight value of the image frames.

2. The method of claim 1 , wherein the video quality factor is used for adjusting parameters of the video processing system if the video quality factor is lower than a threshold associated with the video processing system.

3. The method of claim 1 , wherein a frame quality factor of an image frame includes a peak signal-to-noise ratio (PSNR) value of the image frame that is determined according to the following equation:

PSNR

=

10

×

log

10

⁢

Max

⁢

⁢

Err

2

×

w

×

h

∑

i

=

0

,

j

=

0

w

,

h

⁢

(

x

ij

-

y

ij

)

2

where x ij represents a pixel value of a location (i, j) in the image frame, y ij represents a pixel value of the location (i, j) in a reference frame, MaxErr represents a maximum difference of pixel values between the image frame and the reference frame, w represents a width of the image frame, and h represents a height of the image frame.

4. The method of claim 3 , wherein the average of the frame quality factors of the image frames is represented by an average PSNR value of the image frames.

5. The method of claim 4 , wherein the average PSNR value of the image frames is determined according to the following equation:

PSNR

average

=

1

N

×

∑

n

=

0

N

⁢

PSNR

n

where PSNR n represents a PSNR value of an image frame contained in the video sequence and N represents the number of image frames contained in the video sequence.

6. The method of claim 4 , wherein the weight value of the image frames is determined based on the average PSNR value of the image frames.

7. The method of claim 6 , wherein the weight value of the image frames is determined according to the following equations:

W

=

{

0.5

PSNR

average

≤

16

1

40

×

PSNR

average

+

0.1

16

<

PSNR

average

<

28

0.8

28

≤

PSNR

average

≤

36

-

1

80

×

PSNR

average

+

1.25

36

<

PSNR

average

<

60

0.5

PSNR

average

≥

60

where W represents the weight value of the image frames, and PSNR average represents the average PSNR value of the image frames.

8. The method of claim 4 , wherein the fluctuation value of the image frames is determined according to the following equation:

P

fluctuation

=

∑

i

=

0

N

⁢

(

PSNR

i

-

PSNR

average

)

2

N

where PSNR average represents the average PSNR value of the image frames, PSNR i represents a PSNR value of an image frame, and N represents the number of image frames contained in the video sequence.

9. The method of claim 4 , wherein the video quality factor is determined according to the following equation:

P =PSNR average −W×P fluctuation

where P represents the video quality factor, PSNR average represents the average PSNR value of the image frames, P fluctuation represents the fluctuation value of the image frames, and W represents the weight value of the image frames.

10. The method of claim 1 , wherein the frame quality factor for an image frame includes a structural-similarity-based image quality assessment (SSIM) index of the image frame.

11. The method of claim 10 , wherein the average of the frame quality factors of the image frames is represented by an average SSIM index of the image frames.

12. The method of claim 11 , wherein the video quality factor is determined according to the following equation:

P =SSIM average −W×F fluctuation

where P represents the video quality factor, SSIM average represents the average SSIM index of the image frames, P fluctuation represents the fluctuation value of the image frames, and W represents the weight value of the image frames;

wherein P fluctuation is related to SSIM indices of the image frames; and

wherein W is related to the SSIM indices of the image frames.

13. A system for evaluating quality of a video sequence, said system comprising:

one or more data processors;

a computer-readable storage medium encoded with instructions for commanding the one or more data processors to execute steps including:

receiving a video sequence including one or more image frames, the video sequence being associated with a video processing system;

generating a frame quality factor for each image frame in the video sequence based on a comparison between the image frame and a reference image frame;

generating a fluctuation value and a weight value of the image frames, the fluctuation value of the image frames indicating a variation of the frame quality factors of the image frames, the weight value of the image frames being determined based on an average of the frame quality factors of the image frames; and

outputting a video quality factor based on the frame quality factors of the image frames, the fluctuation value of the image frames, and the weight value of the image frames.

14. The system of claim 13 , wherein a frame quality factor of an image frame includes a peak signal-to-noise ratio (PSNR) value of the image frame.

15. The system of claim 14 , wherein the average of the frame quality factors of the image frames is represented by an average PSNR value of the image frames.

16. The system of claim 15 , wherein the video quality factor is determined according to the following equation:

P =PSNR average −W×P fluctuation

where P represents the video quality factor, PSNR average represents the average PSNR value of the image frames, P fluctuation represents the fluctuation value of the image frames, and W represents the weight value of the image frames.

17. The system of claim 13 , wherein the frame quality factor for an image frame includes a structural-similarity-based image quality assessment (SSIM) index of the image frame.

18. The system of claim 17 , wherein the average of the frame quality factors of the image frames is represented by an average SSIM index of the image frames, and the video quality factor is determined according to the following equation:

P =SSIM average −W×P fluctuation

where P represents the video quality factor, SSIM average represents the average SSIM index of the image frames, P fluctuation represents the fluctuation value of the image frames, and W represents the weight value of the image frames;

wherein P fluctuation is related to SSIM indices of the image frames; and

wherein W is related to the SSIM indices of the image frames.

19. The system of claim 13 , wherein the video quality factor is used for adjusting parameters of the video processing system if the video quality factor is lower than a threshold associated with the video processing system.

20. A non-transitory computer-readable storage medium comprising programming instructions for evaluating quality of a video sequence, the programming instructions being adapted to cause a processing system to execute steps comprising:

receiving a video sequence including one or more image frames, the video sequence being associated with a video processing system;

generating a frame quality factor for each image frame in the video sequence based on a comparison between the image frame and a reference image frame;

generating a fluctuation value and a weight value of the image frames, the fluctuation value of the image frames indicating a variation of the frame quality factors of the image frames, the weight value of the image frames being determined based on an average of the frame quality factors of the image frames; and

outputting a video quality factor based on the frame quality factors of the image frames, the fluctuation value of the image frames, and the weight value of the image frames.