CN103763461B - Video jitter detection method and device - Google Patents

Abstract

Disclosed are a video jitter detection method and device. According to the video jitter detection method and device, G_giggle_f, G_giggle_s and G_giggle can be worked out by calculating the two-dimensional global motion vector of a video; judgment results of jitter conditions of the video are obtained by comparing the G_giggle_f, the G_giggle_s and the G_giggle with different threshold values; the jitter conditions are transmitted to subsequent video coding pre-processing and play links to serve as reference in subsequent relevant video quality improvement steps, and for example it is suggested that an anti-jitter processing algorithm should be adopted for improving the quality of videos classified as strong-jitter or medium-jitter videos.

Description

A kind of video jitter detection method and device

Technical field

The present invention relates to video algorithm field, concrete, relate to method and device that video jitter intensity is detected.

Background technology

Along with digital vedio recording product is widely used in individual consumer, increasing people takes the various videos in life by the mobile phone carried with, flat board, digital camera, Digital Video, photographer may be on walking, the automobile that travels, in the subway that starts, thus the float making the video council of shooting occur in various degree; And the shake of picture can be felt under the weather to beholder.

For this problem, there is research and the algorithm realization of a lot of video stabilization, and also had good subjective effect; But current ant-shaking algorithm all can introduce other negative problems such as the reduction at some picture edges substantially.Therefore, if all apply ant-shaking algorithm to all videos, those will inevitably be made not have the video pictures of float to occur, and user does not wish situation about seeing, reduces the effect of these videos.

If can detect whether video is shaken, further, detect shake intensity and the chattering frequency of video, the relevant improvement for the later stage is provided with reference to decision-making, and be conducive to the raising of video quality, avoid the video quality shone due to unnecessary video jitter process to decline.

Summary of the invention

The object of the invention is to propose a kind of video jitter intensity detection algorithm, shake intensity, the chattering frequency of video can be calculated, and the overall jitter value of video, thus be conducive to judging whether to carry out stabilization process to this video.

For reaching this object, the present invention by the following technical solutions:

A kind of video jitter detection method, comprises the steps:

Two dimension global motion vector calculation procedure S110: for video to be analyzed, the two-dimentional global motion vector that the i-th+1 frame utilizes global motion algorithm detecting step to calculate is V _i(x _i, y _i), wherein, the frame number that i=1 ~ N-1, N have for this video;

Parameters Calculation step S120: utilize formula (1) computing parameter,

${\mathrm{fgX}}_i=\left\{\begin{array}{cc}1& x_i\&times;x_{i+1}<0\\ 0& \mathrm{else}\end{array}\right.,{\mathrm{sgX}}_i=\left\{\begin{array}{cc}\left(\right|x_i|+|x_{i+1}\left|\right)/\mathrm{width}& x_i\&times;x_{i+1}<0\\ 0& \mathrm{else}\end{array}\right.$

${\mathrm{fgY}}_i=\left\{\begin{array}{cc}1& y_i\&times;y_{i+1}<0\\ 0& \mathrm{else}\end{array}\right.,{\mathrm{sgY}}_i=\left\{\begin{array}{cc}\left(\right|y_i|+|y_{i+1}\left|\right)/\mathrm{height}& y_i\&times;y_{i+1}<0\\ 0& \mathrm{else}\end{array}\right.,$ Formula (1)

Wherein, width is the wide of video, and height is the height of video.

Preferably, video jitter frequency computation part step S130: utilize formula (2) to calculate video jitter frequency G_giggle_f,

$G\_\mathrm{giggle}\_f=\underset{i=2}{\overset{N-2}{\mathrm{\&Sigma;}}}\mathrm{Max}(\mathrm{fg}{X}_i,\mathrm{fg}{Y}_i)/(N-2)$ Formula (2).

Preferably, video jitter Strength co-mputation step S140: utilize formula (3) to calculate video jitter intensity G_giggle_s,

$G\_\mathrm{giggle}\_s=\left(\sqrt{\underset{i=1}{\overset{N-2}{\mathrm{\&Sigma;}}}\mathrm{sg}{X_i}^2+\underset{i=1}{\overset{N-2}{\mathrm{\&Sigma;}}}\mathrm{sg}{Y_i}^2}\right)/\underset{i=1}{\overset{N-2}{\mathrm{\&Sigma;}}}\mathrm{Max}(\mathrm{fg}{X}_i,\mathrm{fg}{Y}_i)$ Formula (3).

Preferably, video overall jitter value calculation procedure S150: utilize formula (4) to calculate video jitter frequency G_giggle,

$G\_\mathrm{giggle}=G\_\mathrm{giggle}\_f\&times;G\_\mathrm{giggle}\_s=$ $\left(\sqrt{\underset{i=1}{\overset{N-2}{\mathrm{\&Sigma;}}}\mathrm{sg}{X}^2+\underset{i=1}{\overset{N-2}{\mathrm{\&Sigma;}}}\mathrm{sg}{Y_i}^2}\right)/(N-2)$ Formula (4).

Preferably, video overall jitter judges classifying step S160:

The jitter conditions of video is judged according to the numerical value of G_giggle_f and G_giggle,

Satisfy condition G_giggle_f>0.1, and simultaneously during G_giggle>0.0006, video is strong shake;

Satisfy condition G_giggle_f>0.1, and simultaneously during G_giggle≤0.0006, video is medium shake;

Satisfy condition 0.1 >=G_giggle_f>0.06 time, video is weak shake;

Satisfy condition 0.06 >=G_giggle_f>0.01 time, video is micro-shake;

Satisfy condition G_giggle_f≤0.01 time, video is non-jitter.

The invention also discloses a kind of checkout gear of video jitter, it comprises as lower unit:

Two dimension global motion vector computing unit 210: for video to be analyzed, the two-dimentional global motion vector that the i-th+1 frame utilizes global motion algorithm detecting unit to calculate is V _i(x _i, y _i), wherein, the frame number that i=1 ~ N-1, N have for this video;

Parameters Calculation unit 220: utilize formula (1) computing parameter,

${\mathrm{fgX}}_i=\left\{\begin{array}{cc}1& x_i\&times;x_{i+1}<0\\ 0& \mathrm{else}\end{array}\right.,{\mathrm{sgX}}_i=\left\{\begin{array}{cc}\left(\right|x_i|+|x_{i+1}\left|\right)/\mathrm{width}& x_i\&times;x_{i+1}<0\\ 0& \mathrm{else}\end{array}\right.$

${\mathrm{fgY}}_i=\left\{\begin{array}{cc}1& y_i\&times;y_{i+1}<0\\ 0& \mathrm{else}\end{array}\right.,{\mathrm{sgY}}_i=\left\{\begin{array}{cc}\left(\right|y_i|+|y_{i+1}\left|\right)/\mathrm{height}& y_i\&times;y_{i+1}<0\\ 0& \mathrm{else}\end{array}\right.,$ Formula (1)

Wherein, width is the wide of video, and height is the height of video.

Preferably, video jitter frequency computation part unit 230: utilize formula (2) to calculate video jitter frequency G_giggle_f,

$G\_\mathrm{giggle}\_f=\underset{i=2}{\overset{N-2}{\mathrm{\&Sigma;}}}\mathrm{Max}(\mathrm{fg}{X}_i,\mathrm{fg}{Y}_i)/(N-2)$ Formula (2).

Preferably, video jitter Strength co-mputation unit 240: utilize formula (3) to calculate video jitter frequency G_giggle_s,

$G\_\mathrm{giggle}\_s=\left(\sqrt{\underset{i=1}{\overset{N-2}{\mathrm{\&Sigma;}}}\mathrm{sg}{X_i}^2+\underset{i=1}{\overset{N-2}{\mathrm{\&Sigma;}}}\mathrm{sg}{Y_i}^2}\right)/\underset{i=1}{\overset{N-2}{\mathrm{\&Sigma;}}}\mathrm{Max}(\mathrm{fg}{X}_i,\mathrm{fg}{Y}_i)$ Formula (3).

Preferably, video overall jitter value computing unit 250: utilize formula (4) to calculate video jitter frequency G_giggle,

$G\_\mathrm{giggle}=G\_\mathrm{giggle}\_f\&times;G\_\mathrm{giggle}\_s=$ $\left(\sqrt{\underset{i=1}{\overset{N-2}{\mathrm{\&Sigma;}}}\mathrm{sg}{X}^2+\underset{i=1}{\overset{N-2}{\mathrm{\&Sigma;}}}\mathrm{sg}{Y_i}^2}\right)/(N-2)$ Formula (4).

Preferably, video overall jitter judges taxon 260:

The jitter conditions of video is judged according to the numerical value of G_giggle_f and G_giggle,

Satisfy condition G_giggle_f>0.1, and simultaneously during G_giggle>0.0006, video is strong shake;

Satisfy condition G_giggle_f>0.1, and simultaneously during G_giggle≤0.0006, video is medium shake;

Satisfy condition 0.1 >=G_giggle_f>0.06 time, video is weak shake;

Satisfy condition 0.06 >=G_giggle_f>0.01 time, video is micro-shake;

Satisfy condition G_giggle_f≤0.01 time, video is non-jitter.

The present invention can calculate G_giggle_f, G_giggle_s and G_giggle, by with the comparison of different threshold value obtain this video jitter conditions judge, jitter conditions passed to follow-up Video coding pre-treatment and play link, the reference of video quality correlation step is improved for follow-up, such as the video being categorized into strong shake and medium shake, suggestion adds stabilization Processing Algorithm to improve video quality etc.

Accompanying drawing explanation

Fig. 1 is the video jitter detection method according to the specific embodiment of the invention;

Fig. 2 is the video jitter checkout gear according to the specific embodiment of the invention.

Embodiment

Below in conjunction with drawings and Examples, the present invention is described in further detail.Be understandable that, specific embodiment described herein is only for explaining the present invention, but not limitation of the invention.It also should be noted that, for convenience of description, illustrate only part related to the present invention in accompanying drawing but not entire infrastructure.

See Fig. 1, disclose according to video jitter detection method of the present invention.Comprise the steps:

Two dimension global motion vector calculation procedure S110: for video to be analyzed, the two-dimentional global motion vector that the i-th+1 frame utilizes global motion algorithm detecting step to calculate is V _i(x _i, y _i), wherein, the frame number that i=1 ~ N-1, N have for this video.

Parameters Calculation step S120: utilize formula (1) computing parameter,

${\mathrm{fgX}}_i=\left\{\begin{array}{cc}1& x_i\&times;x_{i+1}<0\\ 0& \mathrm{else}\end{array}\right.,{\mathrm{sgX}}_i=\left\{\begin{array}{cc}\left(\right|x_i|+|x_{i+1}\left|\right)/\mathrm{width}& x_i\&times;x_{i+1}<0\\ 0& \mathrm{else}\end{array}\right.$

${\mathrm{fgY}}_i=\left\{\begin{array}{cc}1& y_i\&times;y_{i+1}<0\\ 0& \mathrm{else}\end{array}\right.,{\mathrm{sgY}}_i=\left\{\begin{array}{cc}\left(\right|y_i|+|y_{i+1}\left|\right)/\mathrm{height}& y_i\&times;y_{i+1}<0\\ 0& \mathrm{else}\end{array}\right.,$ Formula (1)

Wherein, width is the wide of video, and height is the height of video;

Video jitter frequency computation part step S130: utilize formula (2) to calculate video jitter frequency G_giggle_f,

$G\_\mathrm{giggle}\_f=\underset{i=2}{\overset{N-2}{\mathrm{\&Sigma;}}}\mathrm{Max}(\mathrm{fg}{X}_i,\mathrm{fg}{Y}_i)/(N-2)$ Formula (2)

Video jitter Strength co-mputation step S140: utilize formula (3) to calculate video jitter frequency G_giggle_s,

$G\_\mathrm{giggle}\_s=\left(\sqrt{\underset{i=1}{\overset{N-2}{\mathrm{\&Sigma;}}}\mathrm{sg}{X_i}^2+\underset{i=1}{\overset{N-2}{\mathrm{\&Sigma;}}}\mathrm{sg}{Y_i}^2}\right)/\underset{i=1}{\overset{N-2}{\mathrm{\&Sigma;}}}\mathrm{Max}(\mathrm{fg}{X}_i,\mathrm{fg}{Y}_i)$ Formula (3)

Video overall jitter value calculation procedure S150: utilize formula (4) to calculate video jitter frequency G_giggle,

$G\_\mathrm{giggle}=G\_\mathrm{giggle}\_f\&times;G\_\mathrm{giggle}\_s=$ $\left(\sqrt{\underset{i=1}{\overset{N-2}{\mathrm{\&Sigma;}}}\mathrm{sg}{X}^2+\underset{i=1}{\overset{N-2}{\mathrm{\&Sigma;}}}\mathrm{sg}{Y_i}^2}\right)/(N-2)$ Formula (4)

Video overall jitter judges classifying step S160:

The jitter conditions of video is judged according to the numerical value of G_giggle_f and G_giggle,

Satisfy condition G_giggle_f>0.1, and simultaneously during G_giggle>0.0006, video is strong shake;

Satisfy condition G_giggle_f>0.1, and simultaneously during G_giggle≤0.0006, video is medium shake;

Satisfy condition 0.1 >=G_giggle_f>0.06 time, video is weak shake;

Satisfy condition 0.06 >=G_giggle_f>0.01 time, video is micro-shake;

Satisfy condition G_giggle_f≤0.01 time, video is non-jitter.

Embodiment 1:

Certain video has 10 frames (N=10), and resolution is 1280x720 (i.e. width=1280, height=720), and the 2nd frame is respectively (12 to the global motion vector of the 10th frame, 13), (21,15), (-10,12), (-12 ,-20), (-10,0), (-12 ,-21), (0,20), (13,5), (2,14).

Calculate

(i=1 ~ 8) are respectively 0, and 1,0,0,0,0,0,0;

(i=1 ~ 8) are respectively 0, and 0,1,0,0,1,0,0;

(i=1 ~ 8) are respectively 0, and 31/1280,0,0,0,0,0,0;

(i=1 ~ 8) are respectively 0, and 0,32/720,0,0,41/720,0,0;

Video jitter frequency G_giggle_f=3/8=0.375

Video jitter intensity G_giggle_s=0.065/3=0.025

The overall jitter value G_giggle=0.025/8=0.003 of video

Due to G_giggle_f=0.375>0.1, G_giggle=0.003>0.0006, so judge that this video is as " shaking by force ".

Embodiment 2:

Certain video has 20 frames (N=20), resolution is 1920x1080 (i.e. width=1920, height=1080), 2nd frame is respectively (2 to the global motion vector of the 20th frame, 3), (1, 5), (-1, 2), (-1,-2), (-7, 0), (-2,-1), (0, 1), (3, 5), (2, 4), (1, 3), (2, 5), (-1, 2), (-1,-2), (7, 0), (-1,-1), (-1, 5), (3, 5), (-1, 4), (0, 5).

Calculate

(i=1 ~ 18) are respectively 0, and 1,0,0,0,0,0,0,0,0,1,0,1,1,0,1,1,0;

(i=1 ~ 18) are respectively 0, and 0,1,0,0,1,0,0,0,0,0,0,1,0,0,1,0,0;

(i=1 ~ 18) are respectively 0, and 2/1920,0,0,0,0,0,0,0,0,3/1920,0,8/1920,8/1920,0,4/1920,4/1920,0;

(i=1 ~ 18) are respectively 0, and 0,4/1080,0,0,2/1080,0,0,0,0,0,0,4/1080,0,0,6/1080,0,0;

Video jitter frequency G_giggle_f=6/18=0.333

Video jitter intensity G_giggle_s=0.01/6=0.0017

The overall jitter value G_giggle=0.01/18=0.0006 of video

Due to G_giggle_f=0.333>0.1, G_giggle=0.0006<0.001, so judge that this video is as " medium shake ".

Further, the invention also discloses a kind of checkout gear of video jitter, it comprises as lower unit:

Two dimension global motion vector computing unit 210: for video to be analyzed, the two-dimentional global motion vector that the i-th+1 frame utilizes global motion algorithm detecting unit to calculate is V _i(x _i, y _i), wherein, the frame number that i=1 ~ N-1, N have for this video,

Parameters Calculation unit 220: utilize formula (5) computing parameter,

${\mathrm{fgX}}_i=\left\{\begin{array}{cc}1& x_i\&times;x_{i+1}<0\\ 0& \mathrm{else}\end{array}\right.,{\mathrm{sgX}}_i=\left\{\begin{array}{cc}\left(\right|x_i|+|x_{i+1}\left|\right)/\mathrm{width}& x_i\&times;x_{i+1}<0\\ 0& \mathrm{else}\end{array}\right.$

${\mathrm{fgY}}_i=\left\{\begin{array}{cc}1& y_i\&times;y_{i+1}<0\\ 0& \mathrm{else}\end{array}\right.,{\mathrm{sgY}}_i=\left\{\begin{array}{cc}\left(\right|y_i|+|y_{i+1}\left|\right)/\mathrm{height}& y_i\&times;y_{i+1}<0\\ 0& \mathrm{else}\end{array}\right.,$ Formula (5)

Wherein, width is the wide of video, and height is the height of video;

Video jitter frequency computation part unit 230: utilize formula (6) to calculate video jitter frequency G_giggle_f,

$G\_\mathrm{giggle}\_f=\underset{i=2}{\overset{N-2}{\mathrm{\&Sigma;}}}\mathrm{Max}(\mathrm{fg}{X}_i,\mathrm{fg}{Y}_i)/(N-2)$ Formula (6)

Video jitter Strength co-mputation unit 240: utilize formula (7) to calculate video jitter frequency G_giggle_s,

$G\_\mathrm{giggle}\_s=\left(\sqrt{\underset{i=1}{\overset{N-2}{\mathrm{\&Sigma;}}}\mathrm{sg}{X_i}^2+\underset{i=1}{\overset{N-2}{\mathrm{\&Sigma;}}}\mathrm{sg}{Y_i}^2}\right)/\underset{i=1}{\overset{N-2}{\mathrm{\&Sigma;}}}\mathrm{Max}(\mathrm{fg}{X}_i,\mathrm{fg}{Y}_i)$ Formula (7)

Video overall jitter value computing unit 250: utilize formula (8) to calculate video jitter frequency G_giggle,

$G\_\mathrm{giggle}=G\_\mathrm{giggle}\_f\&times;G\_\mathrm{giggle}\_s=$ $\left(\sqrt{\underset{i=1}{\overset{N-2}{\mathrm{\&Sigma;}}}\mathrm{sg}{X}^2+\underset{i=1}{\overset{N-2}{\mathrm{\&Sigma;}}}\mathrm{sg}{Y_i}^2}\right)/(N-2)$ Formula (8)

Video overall jitter judges taxon 260:.

The jitter conditions of video is judged according to the numerical value of G_giggle_f and G_giggle,

Satisfy condition G_giggle_f>0.1, and simultaneously during G_giggle>0.0006, video is strong shake

Satisfy condition G_giggle_f>0.1, and simultaneously during G_giggle≤0.0006, video is medium shake;

Satisfy condition 0.1 >=G_giggle_f>0.06 time, video is weak shake;

Satisfy condition 0.06 >=G_giggle_f>0.01 time, video is micro-shake;

Satisfy condition G_giggle_f≤0.01 time, video is non-jitter.

Therefore, the present invention can calculate G_giggle_f, G_giggle_s and G_giggle, by with the comparison of different threshold value obtain this video jitter conditions judge, jitter conditions passed to follow-up Video coding pre-treatment and play link, improve the reference of video quality correlation step for follow-up, such as the video being categorized into strong shake and medium shake, suggestion adds stabilization Processing Algorithm to improve video quality etc.

Above content is in conjunction with concrete preferred implementation further description made for the present invention; can not assert that the specific embodiment of the present invention is only limitted to this; for general technical staff of the technical field of the invention; without departing from the inventive concept of the premise; some simple deduction or replace can also be made, all should be considered as belonging to the present invention by submitted to claims determination protection range.

Claims (8)

1. a video jitter detection method, comprises the steps:

Two dimension global motion vector calculation procedure S110: for video to be analyzed, the two-dimentional global motion vector that the i-th+1 frame utilizes global motion algorithm detecting step to calculate is V _i(x _i, y _i), wherein, the frame number that i=1 ~ N-1, N have for this video;

Parameters Calculation step S120: utilize formula (1) computing parameter,

$\mathrm{fg}{X}_i=\left\{\begin{array}{cc}1& x_i\&times;x_{i+1}<0\\ 0& \mathrm{else}\end{array}\right.,\mathrm{sg}{X}_i=\left\{\begin{array}{cc}\left(\right|x_i|+|x_{i+1}\left|\right)/\mathrm{width}& x_i\&times;x_{i+1}<0\\ 0& \mathrm{else}\end{array}\right.$

$\mathrm{fg}{Y}_i=\left\{\begin{array}{cc}1& y_i\&times;y_{i+1}<0\\ 0& \mathrm{else}\end{array}\right.,\mathrm{sg}{Y}_i=\left\{\begin{array}{cc}\left(\right|y_i|+|y_{i+1}\left|\right)/\mathrm{height}& y_i\&times;y_{i+1}<0\\ 0& \mathrm{else}\end{array}\right.,$ Formula (1)

Wherein, width is the wide of video, and height is the height of video;

Also comprise video jitter frequency computation part step S130: utilize formula (2) to calculate video jitter frequency G_giggle_f,

$G\_\mathrm{giggle}\_f=\underset{i=1}{\overset{N-2}{\mathrm{\&Sigma;}}}\mathrm{Max}(\mathrm{fg}{X}_i,\mathrm{fg}{Y}_i)/(N-2)$ Formula (2).

2. video jitter detection method according to claim 1, also comprises:

Video jitter Strength co-mputation step S140: utilize formula (3) to calculate video jitter intensity G_giggle_s,

$G\_\mathrm{giggle}\_s=\left(\sqrt{\underset{i=1}{\overset{N-2}{\mathrm{\&Sigma;}}}\mathrm{sg}{X_i}^2+\underset{i=1}{\overset{N-2}{\mathrm{\&Sigma;}}}\mathrm{sg}{Y_i}^2}\right)/\underset{i=1}{\overset{N-2}{\mathrm{\&Sigma;}}}\mathrm{Max}(\mathrm{fg}{X}_i,\mathrm{fg}{Y}_i)$ Formula (3).

3. video jitter detection method according to claim 2, also comprises:

Video overall jitter value calculation procedure S150: utilize formula (4) to calculate video overall jitter value G_giggle,

$G\_\mathrm{giggle}=G\_\mathrm{giggle}\_f\&times;G\_\mathrm{giggle}\_s=$ $\left(\sqrt{\underset{i=1}{\overset{N-2}{\mathrm{\&Sigma;}}}\mathrm{sg}{X}^2+\underset{i=1}{\overset{N-2}{\mathrm{\&Sigma;}}}\mathrm{sg}{Y_i}^2}\right)/(N-2)$ Formula (4).

4. video jitter detection method according to claim 3, also comprises:

Video overall jitter judges classifying step S160:

The jitter conditions of video is judged according to the numerical value of G_giggle_f and G_giggle,

Satisfy condition G_giggle_f>0.1, and simultaneously during G_giggle>0.0006, video is strong shake;

Satisfy condition G_giggle_f>0.1, and simultaneously during G_giggle≤0.0006, video is medium shake;

Satisfy condition 0.1 >=G_giggle_f>0.06 time, video is weak shake;

Satisfy condition 0.06 >=G_giggle_f>0.01 time, video is micro-shake;

Satisfy condition G_giggle_f≤0.01 time, video is non-jitter.

5. a checkout gear for video jitter, it comprises as lower unit:

Two dimension global motion vector computing unit 210: for video to be analyzed, the two-dimentional global motion vector that the i-th+1 frame utilizes global motion algorithm detecting unit to calculate is V _i(x _i, y _i), wherein, the frame number that i=1 ~ N-1, N have for this video;

Parameters Calculation unit 220: utilize formula (1) computing parameter,

$\mathrm{fg}{X}_i=\left\{\begin{array}{cc}1& x_i\&times;x_{i+1}<0\\ 0& \mathrm{else}\end{array}\right.,\mathrm{sg}{X}_i=\left\{\begin{array}{cc}\left(\right|x_i|+|x_{i+1}\left|\right)/\mathrm{width}& x_i\&times;x_{i+1}<0\\ 0& \mathrm{else}\end{array}\right.$

$\mathrm{fg}{Y}_i=\left\{\begin{array}{cc}1& y_i\&times;y_{i+1}<0\\ 0& \mathrm{else}\end{array}\right.,\mathrm{sg}{Y}_i=\left\{\begin{array}{cc}\left(\right|y_i|+|y_{i+1}\left|\right)/\mathrm{height}& y_i\&times;y_{i+1}<0\\ 0& \mathrm{else}\end{array}\right.,$ Formula (1)

Wherein, width is the wide of video, and height is the height of video;

Also comprise video jitter frequency computation part unit 230: utilize formula (2) to calculate video jitter frequency G_giggle_f,

$G\_\mathrm{giggle}\_f=\underset{i=1}{\overset{N-2}{\mathrm{\&Sigma;}}}\mathrm{Max}(\mathrm{fg}{X}_i,\mathrm{fg}{Y}_i)/(N-2)$ Formula (2).

6. the detection method of video jitter according to claim 5, also comprises:

Video jitter Strength co-mputation unit 240: utilize formula (3) to calculate video jitter frequency G_giggle_s,

$G\_\mathrm{giggle}\_s=\left(\sqrt{\underset{i=1}{\overset{N-2}{\mathrm{\&Sigma;}}}\mathrm{sg}{X_i}^2+\underset{i=1}{\overset{N-2}{\mathrm{\&Sigma;}}}\mathrm{sg}{Y_i}^2}\right)/\underset{i=1}{\overset{N-2}{\mathrm{\&Sigma;}}}\mathrm{Max}(\mathrm{fg}{X}_i,\mathrm{fg}{Y}_i)$ Formula (3).

7. the detection method of video jitter according to claim 6, also comprises:

Video overall jitter value computing unit 250: utilize formula (4) to calculate video overall jitter value G_giggle,

$G\_\mathrm{giggle}=G\_\mathrm{giggle}\_f\&times;G\_\mathrm{giggle}\_s=$ $\left(\sqrt{\underset{i=1}{\overset{N-2}{\mathrm{\&Sigma;}}}\mathrm{sg}{X}^2+\underset{i=1}{\overset{N-2}{\mathrm{\&Sigma;}}}\mathrm{sg}{Y_i}^2}\right)/(N-2)$ Formula (4).

8. the detection method of video jitter according to claim 7, also comprises:

Video overall jitter judges taxon 260:

The jitter conditions of video is judged according to the numerical value of G_giggle_f and G_giggle,

Satisfy condition G_giggle_f>0.1, and simultaneously during G_giggle>0.0006, video is strong shake;

Satisfy condition G_giggle_f>0.1, and simultaneously during G_giggle≤0.0006, video is medium shake;

Satisfy condition 0.1 >=G_giggle_f>0.06 time, video is weak shake;

Satisfy condition 0.06 >=G_giggle_f>0.01 time, video is micro-shake;

Satisfy condition G_giggle_f≤0.01 time, video is non-jitter.