CN100515069C - Video evaluation device and method, frame rate determination device, and video process device - Google Patents

Video evaluation device and method, frame rate determination device, and video process device Download PDF

Info

Publication number
CN100515069C
CN100515069C CNB2005101039936A CN200510103993A CN100515069C CN 100515069 C CN100515069 C CN 100515069C CN B2005101039936 A CNB2005101039936 A CN B2005101039936A CN 200510103993 A CN200510103993 A CN 200510103993A CN 100515069 C CN100515069 C CN 100515069C
Authority
CN
China
Prior art keywords
mentioned
frame rate
evaluation
estimate
video
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CNB2005101039936A
Other languages
Chinese (zh)
Other versions
CN1750635A (en
Inventor
加藤禎笃
文仲丞
堀越力
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
NTT Docomo Inc
Original Assignee
NTT Docomo Inc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by NTT Docomo Inc filed Critical NTT Docomo Inc
Publication of CN1750635A publication Critical patent/CN1750635A/en
Application granted granted Critical
Publication of CN100515069C publication Critical patent/CN100515069C/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Television Systems (AREA)
  • Compression Or Coding Systems Of Tv Signals (AREA)

Abstract

The invention provides a video evaluating device, a frame rate determining device, a video processing device, a video evaluating method and a video evaluating program, wherein, the frame rate is determined corresponding to motion smoothness of a video; a variation detecting component (1401) extracts a variation on the basis of a plurality of frame images which are included in input video signals (1403) externally inputted as motion image signals and outputs the variation (1405) to a evaluation value calculating element (1402); the evaluation value calculating element (1402) presumes time variation between various frame images according to the variation (1405) and time intervals between the frame images on the basis of frame rate information (105), and calculates an evaluation value which is used for evaluating the motion smoothness of the inputted video according to the time variation; the evaluation value is taken as a factor for determining the frame rate and outputted to the outside.

Description

Video evaluation device and method, frame rate are determined device, are reached video process apparatus
Technical field
The present invention relates to the video evaluation device, frame rate is determined device, video process apparatus, video evaluation method, and video evaluation program.
Background technology
Comprise the obtaining of video, store, transmit, show, the Video processing of coding, decoding etc. generally handles with fixed frame rate.At this, so-called frame rate is meant the frame number that handle per 1 second.In addition.So-called fixed frame rate is meant that the frame number of handling per 1 second is set to certain frame rate.Object lesson as fixed frame rate, for example in the standard of the U.S. and the Japanese national television standards committee (National Television Standards Committee:NTSC (NTSC system)) that adopts, be defined as 29.97fps (frameper second: frame/second).In addition, in phase alternation capable (Phase Alternating Line:PAL (the Phase Alternate Line)) standard of the national television standards committee that adopts in Europe, be defined as 25fps.And then, use the situation of the fixed frame rate of 15fps and 24fps in addition.So-called " video " is that " two field picture " constitutes continuously by making each rest image in addition.
When carrying out under the situation of Video processing with fixed frame rate, if increase frame rate, the time interval of then continuous frame shortens.Can handle the video of more level and smooth motion thus.For example, compare with the video of handling with the frame rate of 15fps with the video that the frame rate of 30fps is handled, so because the many motions that can show video more meticulously of the frame number in the time per unit, as all can showing with more level and smooth motion.
In addition, except the Video processing that adopts the said fixing frame rate, can also adopt the Video processing of variable frame-rate.The Video processing of this variable frame-rate changes frame rate according to the treating capacity and the data volume of video.For example, when encoded video,, reduce the frame number that frame rate is cut down coding in the time per unit being judged as under the many situations of coded data amount.This is because if the data volume increase cause that increases of needed time of Video processing then.At this, in variable frame-rate, be set to when the time interval of continuous frame images under the situation of T, the frame rate between 2 two field pictures is 1/T.
The technology that is used for changing such frame rate for example is disclosed in patent documentation 1 (spy opens flat 11-112940 communique).
, when handling with fixed frame rate under the situation of video,, then follow treating capacity, data volume and the electric power consumption of Video processing to increase if increase frame rate in order to realize level and smooth motion.If specifically, for example, obtaining under the situation of video, owing in time per unit, need the frame number obtained to increase, thereby treating capacity and follow the electric power consumption of processing to increase.In addition, under the situation that stores video, owing to the frame number that needs in the time per unit to store increases, so data volume increases.
On the other hand, frame rate is reduced, the then motion smoothing reduction of video becomes the video of rough motion.
In addition, when handling under the situation of video with variable frame-rate, if only according to the treating capacity and the data volume change frame rate of video, then the motion smoothing reduction of video becomes rough sport video.
Do not change frame rate if do not consider the motion feature of video like this, then have unnecessary many treating capacities, data volume and power consumption, become the problem of rough video.
Summary of the invention
Thereby, the present invention is in order to address the above problem, its purpose is: provide a kind of flatness of the motion according to video to determine that video evaluation device, the frame rate of frame rate determine that device, video process apparatus, video evaluation method, frame rate determine that method, method for processing video frequency, video evaluation program, frame rate determine program, and video processing program.
The invention provides a kind of video evaluation device, it is characterized in that comprising: the variable quantity detection part, based in a plurality of two field pictures that are included in the incoming video signal based on two continuous two field pictures of the frame rate information that receives from the outside, detect the intensity of variation that is illustrated in the brightness value between above-mentioned continuous two two field pictures or the variable quantity of motion vector; The evaluation of estimate calculating unit, based on by the time interval between the detected variable quantity of above-mentioned variable quantity detection part and above-mentioned continuous two two field pictures, calculating be used to estimate above-mentioned incoming video signal motion flatness evaluation of estimate and it is outputed to the outside, above-mentioned evaluation of estimate calculating unit is based on the time interval between above-mentioned variable quantity and above-mentioned continuous two two field pictures, infer with above-mentioned variable quantity and above-mentioned continuous two two field pictures between time interval time corresponding variable quantity, use this time variation amount to calculate above-mentioned evaluation of estimate.
The present invention also provides a kind of frame rate to determine device, it is characterized in that comprising: the frame rate generation part that the 1st frame rate takes place; The variable quantity detection part, based in a plurality of two field pictures that are included in the incoming video signal based on two continuous two field pictures of above-mentioned the 1st frame rate, detect the intensity of variation of the brightness between above-mentioned continuous two two field pictures of expression or the variable quantity of motion vector; The evaluation of estimate calculating unit based on the time interval between above-mentioned variable quantity and above-mentioned continuous two two field pictures, calculates the evaluation of estimate of the flatness of the motion that is used to estimate above-mentioned incoming video signal; Determine parts with frame rate, use above-mentioned evaluation of estimate and above-mentioned the 1st frame rate, determine to carry out above-mentioned incoming video signal processing the 2nd frame rate and it is outputed to the outside, above-mentioned evaluation of estimate calculating unit is based on the time interval between above-mentioned variable quantity and above-mentioned continuous two two field pictures, infer with above-mentioned variable quantity and above-mentioned continuous two two field pictures between time interval time corresponding variable quantity, use this time variation amount to calculate above-mentioned evaluation of estimate.
The present invention also provides a kind of video process apparatus, it is characterized in that comprising: the buffer unit of storage incoming video signal; The frame rate generation part of the 1st frame rate takes place; The variable quantity detection part, based in a plurality of two field pictures that are included in the above-mentioned incoming video signal based on two continuous two field pictures of above-mentioned the 1st frame rate, detect the intensity of variation of the brightness between above-mentioned continuous two two field pictures of expression or the variable quantity of motion vector; The evaluation of estimate calculating unit based on the time interval between above-mentioned variable quantity and above-mentioned continuous two two field pictures, calculates the evaluation of estimate of the motion smoothing that is used to estimate above-mentioned incoming video signal; Frame rate is determined parts, uses above-mentioned evaluation of estimate and above-mentioned the 1st frame rate, determines to carry out the 2nd frame rate of the processing of above-mentioned incoming video signal; With the Video processing parts, use above-mentioned the 2nd frame rate to read in and carry out Video processing by the above-mentioned incoming video signal of above-mentioned buffer unit storage, above-mentioned evaluation of estimate calculating unit is based on the time interval between above-mentioned variable quantity and above-mentioned continuous two two field pictures, infer with above-mentioned variable quantity and above-mentioned continuous two two field pictures between time interval time corresponding variable quantity, use this time variation amount to calculate above-mentioned evaluation of estimate.
The present invention also provides a kind of video evaluation method, it is characterized in that comprising: variable quantity detects step, based in a plurality of two field pictures that are included in the incoming video signal based on two continuous two field pictures of the frame rate information that receives from the outside, detect the intensity of variation that is illustrated in the brightness value between above-mentioned continuous two two field pictures or the variable quantity of motion vector; With the evaluation of estimate calculation procedure, based on the time interval of detecting at above-mentioned variable quantity between detected variable quantity in the step and above-mentioned continuous two two field pictures, calculating be used to estimate above-mentioned incoming video signal motion flatness evaluation of estimate and it is outputed to the outside, above-mentioned evaluation of estimate calculation procedure is based on the time interval between above-mentioned variable quantity and above-mentioned continuous two two field pictures, infer with above-mentioned variable quantity and above-mentioned continuous two two field pictures between time interval time corresponding variable quantity, use this time variation amount to calculate above-mentioned evaluation of estimate.
If adopt these inventions, then based on a plurality of two field picture change detected amounts that are included in the incoming video signal, based on the time interval between each two field picture of this variable quantity and incoming video signal, calculate the evaluation of estimate of the flatness of the motion be used to estimate incoming video signal.Thereby, can according to according to the variable quantity of incoming video signal of fixed frame rate input, estimate the flatness of the motion of the video in the frame rate when Video processing.In addition, because the evaluation of estimate that obtains according to this evaluation is outputed to the outside, so can determine frame rate based on this evaluation of estimate.That is, can determine frame rate according to the flatness of the motion of this video.
If adopt these inventions, then according to the many two field picture change detected amounts that are included in the incoming video signal, according to the time interval between this converted quantity and the two field picture corresponding with the 1st frame rate, calculating is used to estimate the evaluation of estimate of flatness of the motion of incoming video signal, utilizes this evaluation of estimate to determine to carry out the 2nd frame rate of the processing of incoming video signal.That is,, when estimating the flatness of motion of the video in the 1st frame rate, use this evaluation to determine the 2nd frame rate corresponding with variable quantity according to the incoming video signal of the 1st frame rate input.Thereby, can read in incoming video signal with the 2nd frame rate of determining accordingly with the evaluation of the flatness of the motion of video.That is, with the frame rate of the corresponding definite incoming video signal of flatness of the motion of video in, while can keep the motion smoothing of video to read in incoming video signal.
If adopt these inventions, then according to the many two field picture change detected amounts that are included in the incoming video signal, according to the time interval between this variable quantity and the two field picture corresponding with the 1st frame rate, calculating is used to estimate the evaluation of estimate of flatness of the motion of incoming video signal, utilizes this evaluation of estimate to determine to carry out the 2nd frame rate of the processing of incoming video signal.In addition, use the 2nd frame rate to carry out the Video processing of incoming video signal.That is, corresponding with the variable quantity of the incoming video signal of importing according to the 1st frame rate, in the motion smoothing of the video in estimating the 1st frame rate, use this evaluation to determine the 2nd frame rate.In addition, use the 2nd frame rate of determining according to the evaluation of estimate of relative incoming video signal, carry out the Video processing of incoming video signal.Thereby, according to corresponding the 2nd definite frame rate, can carry out the Video processing of incoming video signal with the evaluation of motion smoothing.That is, with the corresponding definite frame rate of flatness of the motion of video in, while can keep the flatness of the motion of video to carry out the Video processing of incoming video signal.
If adopt these inventions, then detect the moving displacement amounts according to the many two field pictures that are included in the incoming video signal, according to the frame rate of this moving displacement amount and incoming video signal, calculate the evaluation of estimate of the flatness of the motion that is used to estimate incoming video signal.Thereby, can with according to the displacement of motion of incoming video signal of fixed frame rate input, the flatness of the motion of the video in the frame rate when estimating Video processing.In addition, because the evaluation of estimate that obtains by this evaluation is outputed to the outside, so can determine frame rate according to this evaluation of estimate.That is, can determine frame rate according to the motion smoothing of video.
If employing the present invention, then detect the moving displacement amount according to the many two field pictures that are included in the incoming video signal, according to this moving displacement amount and the 1st frame rate, calculating is used to estimate the evaluation of estimate of flatness of the motion of incoming video signal, uses this evaluation of estimate to determine to carry out the 2nd frame rate of the processing of incoming video signal.That is, with the flatness of the motion of correspondingly estimating the video in the 1st frame rate according to the displacement of the motion of the incoming video signal of the 1st frame rate input in, use this evaluation to determine the 2nd frame rate.Thereby, can be with reading in incoming video signal with corresponding the 2nd definite frame rate of evaluation of the flatness of the motion of video.That is, when correspondingly determining the frame rate of incoming video signal, can in the motion smoothing that keeps video, read in incoming video signal with the flatness of video motion.
If employing the present invention, then detect the moving displacement amount according to the many two field pictures that are included in the incoming video signal, according to this moving displacement amount and the 1st frame rate, calculating is used to estimate the evaluation of estimate of flatness of the motion of incoming video signal, utilizes this evaluation of estimate to determine to carry out the 2nd frame rate of the processing of incoming video signal.In addition, use the 2nd frame rate to carry out the Video processing of incoming video signal.That is, in flatness, use this evaluation to determine the 2nd frame rate according to the motion of estimating the video in the 1st frame rate according to the displacement of the motion of the incoming video signal of the 1st frame rate input.In addition, use the 2nd frame rate of determining according to the evaluation of estimate of incoming video signal, carry out the Video processing of incoming video signal.Thereby, according to corresponding the 2nd definite frame rate of evaluation of flatness of motion, can carry out the Video processing of incoming video signal.That is, with the corresponding definite frame rate of flatness of the motion of video in, while can keep the flatness of the motion of video to carry out the Video processing of incoming video signal.
Determine that device, video process apparatus, video evaluation method, frame rate determine that method, method for processing video frequency, video evaluation program, frame rate determine program if adopt video evaluation device of the present invention, frame rate, and video processing program, because can determine frame rate according to the flatness of the motion of video, so when reducing treating capacity, data volume and power consumption, can provide the video of the motion of flatness.
Description of drawings
Fig. 1 is the figure that shows that the function of the video evaluation device in the distortion example of execution mode 1 constitutes.
Fig. 2 is the figure that is used to illustrate the method that detects displacement.(a) being the figure that shows two field picture P0, (b) is the figure that shows two field picture P1.
Fig. 3 is the figure that is used to illustrate the method for estimating the flatness of moving.(a) being the figure that is used to illustrate the method for inferring the interframe movement amount, (b) is to be used to illustrate the figure that calculates the method for evaluation of estimate based on the interframe movement amount.
Fig. 4 is the flow chart of the flow process of showing that the video evaluation of the distortion example of execution mode 1 is handled.
Fig. 5 is the figure that is used to illustrate the scope of trying to achieve based on the size and Orientation of each motion vector.
Fig. 6 is illustrated in the figure that the module of the video evaluation program in the distortion example of execution mode 1 constitutes.
Fig. 7 is that the frame rate in the displaying execution mode 2 is determined the figure that the function of device constitutes.
Fig. 8 (a) is the figure of example the 1st frame rate, (b) is the figure of the sampling rate of example incoming video signal.
Fig. 9 is a flow chart of showing the definite flow process of handling of frame rate in the execution mode 2.
Figure 10 is that the frame rate in the illustrated embodiment 2 is determined the figure that the module of program constitutes.
Figure 11 is the figure that the function of the video process apparatus in the illustrated embodiment 3 constitutes.
Figure 12 is a flow chart of showing the flow process of the Video processing in the execution mode 3.
Figure 13 is the figure that the module of the video processing program in the illustrated embodiment 3 constitutes.
Figure 14 is the figure that the function of the video evaluation device in the illustrated embodiment 1 constitutes.
Figure 15 is the figure that is used to illustrate the method for change detected amount.
Figure 16 is the figure that is used to illustrate the method for estimating the flatness of moving.(a) being the figure that is used to illustrate the method for inferring time variation amount, (b) is to be used to illustrate the figure that calculates the method for evaluation of estimate based on time variation amount.
Figure 17 is a flow chart of showing the flow process of the video evaluation processing in the execution mode 1.
Figure 18 is the figure that the module of the video evaluation program in the example embodiment 1 constitutes.
Embodiment
Below, relate to video evaluation device of the present invention, frame rate based on description of drawings and determine that device, video process apparatus, video evaluation method, frame rate determine that method, method for processing video frequency, video evaluation program, frame rate determine program, and each execution mode of video processing program.And then in each figure, also the repetitive description thereof will be omitted for additional prosign on same key element.
[execution mode 1]
At first, embodiments of the present invention 1 are described.Figure 14 is the figure that the function of the video evaluation device 140 in the illustrated embodiment 1 constitutes.
At this, video evaluation device 140 physically is to possess CPU (central processing unit); The storage device of memory etc., and the computer of communicator etc.Thereby video evaluation device 140 promptly can be the fixed communication terminal of PC terminal etc., also can be the mobile communication terminal of mobile phone etc.That is,, can extensively be suitable for the device that can carry out information processing as video evaluation device 140.
The function that video evaluation device 140 is described with reference to Figure 14 constitutes.As shown in figure 14, video evaluation device 140 comprises: change amount detection 1401; Evaluation of estimate computing unit 1402.
Change amount detection 1401 is being decomposed into two field picture from the outside as the incoming video signal 1403 of motion image signal input.Change amount detection 1401 detects the variable quantity of the intensity of variation between each two field picture that is illustrated in incoming video signal based on a plurality of two field pictures that decompose.Change amount detection 1401 outputs to evaluation of estimate computing unit 1402 to detected variable quantity 1405.
At this, specify the method for change detected amount 1405 with reference to Figure 15.Change amount detection 1401 orders are read in 2 the continuous two field picture that decomposes from incoming video signal 1403.At this, for convenience of explanation, 2 continuous two field pictures with the order of reading in as two field picture P0, two field picture P1 explanation.Change amount detection 1401 is asked the poor of brightness value between the pixel that lays respectively on the same coordinate in two field picture P1 that reads in and two field picture P0, at being included in the square value that two field picture each pixel in all calculates this difference.Change amount detection 1401 is by calculating the mean value of the above-mentioned square value that each pixel is calculated, change detected amount 1405.Thereby the mean value that calculates outputs to evaluation of estimate computing unit 1402 as variable quantity 1405.
Evaluation of estimate computing unit 1402 is according to the variable quantity 1405 that receives from change amount detection 1401, based on the time interval between the two field picture of the frame rate information 1404 that receives from the outside, infer and variable quantity 1405 and each two field picture between corresponding time variation amount of the time interval.Evaluation of estimate computing unit 1402 is based on the time variation amount of inferring, calculates the evaluation of estimate of the flatness of the motion that is used to estimate input video.Evaluation of estimate computing unit 1402 outputs to the outside to the evaluation of estimate 1406 that calculates.At this, be equivalent to as the outside, for example be identified for carrying out the device of frame rate of the best of the Video processing of incoming video signal 1403 based on evaluation of estimate 1406.By to such external device (ED) output evaluation of estimate 1406, can determine the corresponding frame rate of flatness with the motion of the video of incoming video signal 1403.
At this, specify the method for the flatness of estimating motion with reference to Figure 16.At first, with reference to the method for Figure 16 (a) illustrative examples as the time variation amount S1 of deduction in moment T1.Evaluation of estimate computing unit 1402 is inferred time variation amount S1 according to time interval Δ t1 and variation delta c1 based on the moment T1 of the moment T0 of the two field picture P0 of frame rate information 1404 and two field picture P1.In addition, constantly T0 and constantly T1 time interval Δ t1 for example the frame rate of the frame rate information 1404 in moment T1 be under the situation of F1fps, be 1/F1 second.
Further specifically describe for the method for inferring time variation amount.Shown in Figure 16 (a), moment variable quantity S1 in moment T1 for example is set to Δ t1 in the time interval of moment T0 and moment T1, variable quantity in moment T1 is set under the situation of Δ c1, is Δ t1 Δ c1 (area of the oblique line part S1 shown in Figure 16 (a)).Equally, the time variation amount S2 in moment T2 for example is set to Δ t2 in the time interval of moment T1 and moment T2, and the variable quantity in moment T2 is set under the situation of Δ c2, is Δ t2 Δ c2 (area of the oblique line part S2 shown in Figure 16 (a)).
Below, with reference to Figure 16 (b), the method for calculating evaluation of estimate based on time variation amount is described.Evaluation of estimate computing unit 1402 calculates at the moment of each two field picture Tn (n: positive integer for the whole two field picture that is included in the incoming video signal.Below identical) in the evaluation of estimate of flatness of motion.If specifically, then evaluation of estimate computing unit 1402 uses the time variation amount Sn in the moment of each two field picture Tn, calculates the evaluation of estimate of the flatness of the motion in moment Tn.If more particularly, then evaluation of estimate computing unit 1402 calculates the evaluation of estimate of the flatness of the motion in moment Tn for example with the following formula, the α/Sn (α is a constant) that have used time variation amount Sn.In addition, evaluation of estimate computing unit 1402 for example also can be with the following formula that has used time variation amount Sn, a * exp -bSn+ c (a, b, c are constants), the evaluation of estimate of the flatness of the motion of calculating in moment Tn.Evaluation of estimate computing unit 1402 is based on each evaluation of estimate that calculates, and calculates the mean value of evaluation of estimate of the flatness of the motion in whole moment of input video.This mean value outputs to the outside as the final evaluation of estimate 1406 of input video.
Below, with reference to the flow process of the video evaluation processing of Figure 17 explanation in the video evaluation device 140 of execution mode 1.
At first, change amount detection 1401 orders are read in 2 the continuous two field picture (two field picture P0, two field picture P1) (step S1701) that decomposes from incoming video signal 1403.
Below, change amount detection 1401 is asked the poor of brightness value between the pixel that is positioned on the same coordinate respectively in two field picture P1 that reads in and two field picture P0, at being included in the square value (step S1702) that two field picture each pixel in all calculates its difference.
Below, change amount detection 1401 is calculated the mean value (step S1703) of the above-mentioned square value that calculates at each pixel.This mean value that calculates outputs to evaluation of estimate computing unit 1402 as variable quantity 1405.
Below, evaluation of estimate computing unit 1402 is according to the variable quantity 1405 that receives from change amount detection 1401 with based on the time interval between the two field picture of the frame rate information 1404 that receives from the outside, calculate and variable quantity 1405 and each two field picture between corresponding time variation amount of the time interval (step 1704).
Below, 1402 pairs of evaluation of estimate computing units are included in the whole two field picture in the incoming video signal, and service time, variable quantity calculated the evaluation of estimate (step S1705) of the flatness of the motion in the moment of each two field picture Tn.
Below, average calculation unit 1402 is based on each evaluation of estimate that calculates, and calculates the mean value (step S1706) of evaluation of estimate of the flatness of the motion in whole moment of input video.
Below, evaluation of estimate computing unit 1402 outputs to outside (step 1707) to the evaluations of estimate that calculate as the evaluation of estimate 1406 for the flatness of all motions of incoming video signal 1403.
As mentioned above, if adopt the video evaluation device 140 of execution mode 1, then the square value based on the difference of the brightness value of each interframe that is included in a plurality of two field pictures in the incoming video signal calculates variable quantity.In addition, according to this variable quantity and based on the time interval between the two field picture of the frame rate of incoming video signal, calculate the evaluation of estimate of the flatness of the motion that is used to estimate incoming video signal.Thereby, based on according to the variable quantity of brightness value of incoming video signal of fixed frame rate input, can estimate the flatness of the motion of the video in the frame rate when Video processing.In addition, because the evaluation of estimate that obtains by this evaluation is outputed to the outside, so can externally determine frame rate based on this evaluation of estimate.That is, can determine frame rate according to the flatness of the motion of video.
In addition, it is all that the unit that adopts the variable quantity of above-mentioned change amount detection 1401 to detect is not limited to above-mentioned two field picture.For example, also can be block unit, 1 pixel unit, target area unit etc.
In addition, the variable quantity Calculation Method of employing variable quantity computing unit 1401 is not limited to use the mean value of the above-mentioned square value that calculates at each pixel that is included in the above-mentioned two field picture.For example, can use maximum, median or the minimum value of the above-mentioned square value that calculates at each pixel, also can use the subduplicate value of maximum, median or the minimum value of above-mentioned square value, can also use the dispersion value in the two field picture of above-mentioned square value is all.
In addition, the method that adopts the variable quantity of change amount detection 1401 to detect is not limited to use the square value of the difference of the brightness value between pixel on the same coordinate that is positioned at each above-mentioned frame.For example, also can use the value of the difference of the brightness value between the pixel on the same coordinate that is positioned at each frame, perhaps should poor absolute value.
In addition, in adopting the variable quantity detection method of change amount detection 1401, except each above-mentioned value, for example can also use motion vector between the two field picture of incoming video signal etc., be illustrated in all values of variation of the incoming video signal of interframe.
In addition, adopt the deduction method of the time variation amount of evaluation of estimate computing unit 1402 to be not limited to (for example use above-mentioned formula, Δ t1 Δ c1) method is as long as can infer according to variable quantity with based on the time interval between the two field picture of frame rate information.
In addition, the time variation amount of being inferred by evaluation of estimate computing unit 1402 not necessarily must be 1 in frame unit.For example also can be 1 in block unit, pixel unit, target unit.
In addition, the final evaluation of estimate 1406 of the input video that calculates with evaluation of estimate computing unit 1402 is not limited to the mean value of evaluation of estimate of the flatness of the motion in whole moment of above-mentioned input video.For example, also can be maximum, median or the minimum value of evaluation of estimate of the flatness of the motion in whole moment of input video.
In addition, the final evaluation of estimate 1406 of the input video that calculates with evaluation of estimate computing unit 1402 must be 1 for the whole two field picture that is included in the incoming video signal not necessarily.For example, can it be 1 also at each of several two field pictures, 1 two field picture, piece, pixel, target.
At last, explanation is used to make the video evaluation program 180 of computer as above-mentioned video evaluation device 140 performance functions with reference to Figure 18.
As shown in figure 18, video evaluation program 140 comprises: the primary module program 1801 of aggregation process; Variable quantity detection module 1802; Evaluation of estimate computing module 1803.The function that variable quantity detection module 1802 and evaluation of estimate computing module 1803 make computer performance is identical with the function that above-mentioned change amount detection 1401 and evaluation of estimate computing unit 1402 have respectively.
And then video evaluation value program 180 is for example provided by storage medium or the semiconductor memory of CD-ROM, DVD or ROM etc.In addition, video evaluation program 180 also can provide via network as the computer data signal that overlaps on the carrier wave.
[the distortion example of execution mode 1]
Below, the distortion example of above-mentioned execution mode 1 is described.Fig. 1 is illustrated in the figure that the function of the video evaluation device 10 in the distortion example of execution mode 1 constitutes.
At this, video evaluation device 10 physically is the storage device that possesses CPU (central processing unit), memory etc., and the computer of communicator etc.Thereby video evaluation device 10 can be the fixed communication terminal of PC terminal etc., also can be the mobile communication terminal of mobile phone etc.That is, can extensively be suitable for the device that can carry out information processing as video evaluation device 10.
The formation function of video evaluation device 10 is described with reference to Fig. 1.As shown in Figure 1, video evaluation device 10 comprises: displacement detecting unit 101; Eigenvalue calculation unit 102; Evaluation of estimate computing unit 103.
Displacement detecting unit 101 is being decomposed into two field picture from the outside as the incoming video signal 104 of motion image signal input.Displacement detecting unit 101 detects the displacement (moving displacement amount) of the degree of displacement of the motion of representing incoming video signal based on a plurality of two field pictures after decomposing.Displacement detecting unit 101 outputs to eigenvalue calculation unit 102 to detected displacement 106.
In addition, displacement is not limited to the degree of displacement of the motion of incoming video signal, gets final product so long as be illustrated in the amount of the intensity of variation between each two field picture of incoming video signal.
At this, specify the method that detects displacement 106 with reference to Fig. 2.Displacement detecting unit 101 orders are read in 2 the continuous two field pictures that decompose from incoming video signal 104.At this, for convenience of explanation, 2 continuous two field pictures according to the order of reading in as two field picture P0 (with reference to Fig. 2 (a)), two field picture P1 (with reference to Fig. 2 (b)) describes.Displacement detecting unit 101 is divided into the two field picture P1 that reads in the piece of prescribed level.Displacement detecting unit 101 is searched the picture signal pattern the most similar to the picture signal pattern of this each piece for each piece of two field picture P1 from two field picture P0.This is for example searched and can realize by using the processing of searching by piece coupling (correlation method) shown in Figure 2.Displacement detecting unit 101 is based on handling the signal pattern be judged as two the most similar images by searching, detect spatial displacement between the signal pattern of two images and be motion vector V (MVx, MVy).This motion vector V outputs to eigenvalue calculation unit 102 as displacement 106.
Eigenvalue calculation unit 102 calculates the motion characteristic value 107 of the motion characteristics of expression incoming video signal based on the displacement 106 that receives from displacement detecting unit 101.Eigenvalue calculation unit 102 outputs to evaluation of estimate computing unit 103 to the motion characteristic value 107 that calculates.
At this, specifically describe the method for calculating motion characteristic value.Eigenvalue calculation unit 102 uses the motion vector of each piece of the two field picture P1 that receives as displacement 106, asks the size of motion vector of each piece of two field picture P1.The size of this motion vector for example is being arranged to the x composition of the motion vector of the piece arbitrarily on the two field picture P1, y composition respectively under the situation of MVx, MVy, by (MVx 2+ MVy 2) 1/2Try to achieve.Eigenvalue calculation unit 102 calculates the value of the feature that becomes two field picture P1 based on the size of each motion vector.This value that calculates outputs to evaluation of estimate computing unit 103 as motion characteristic value 107.As the value that calculates with eigenvalue calculation unit 102 (motion characteristic value 107), for example be equivalent to maximum, mean value, median or the minimum value of the size of the motion vector of trying to achieve at each piece that is included in the two field picture.
Evaluation of estimate computing unit 103 is according to the motion designated value 107 that receives from eigenvalue calculation unit 102, and, infer the interframe movement vector of the degree that is illustrated in the motion between each two field picture based on the time interval between the two field picture of the frame rate information 105 that receives from the outside.Evaluation of estimate computing unit 103 calculates the evaluation of estimate of the flatness of the motion that is used to estimate input video based on the interframe movement amount inferred.Evaluation of estimate computing unit 103 outputs to the outside to the evaluation of estimate 108 that calculates.At this, be equivalent to as the outside, for example be identified for carrying out the device etc. of frame rate of the best of the Video processing of incoming video signal 104 based on evaluation of estimate 108.By to such external device (ED) output evaluation of estimate 108, can determine the corresponding frame rate of motion smoothing with the video that adopts incoming video signal 104.
And then, when calculating evaluation of estimate, not necessarily must calculate based on the interframe movement amount.For example, can based on and displacement 106 and each two field picture between corresponding time variation amount of the time interval, calculate evaluation of estimate.This time variation amount can be inferred according to the motion characteristic value 107 that receives from eigenvalue calculation unit 102, based on the time interval between the two field picture of the frame rate information 105 that receives from the outside.
At this, specify the method for the flatness of estimating motion with reference to Fig. 3.At first, with reference to Fig. 3 (a) method of inferring the interframe movement amount S1 in moment T1 for example is described.Evaluation of estimate computing unit 103 is inferred interframe movement amount S1 according to based on the time interval Δ t1 of the moment T1 of the moment T0 of the two field picture P0 of frame rate information 105 and two field picture P1 and the motion characteristic value Δ d1 of two field picture P1.In addition, constantly T0 and constantly T1 time interval Δ t1 for example the frame rate of the frame rate information 105 in moment T1 be to be 1/F1 second under the situation of F1fps.
The method of inferring the interframe movement amount is described more specifically.Shown in Fig. 3 (a), interframe movement amount S1 in moment T1 for example is set to Δ t1 in the time interval of moment T0 and moment T1, motion characteristic value among the T1 is set under the situation of Δ d1 constantly, is Δ t1 Δ d1/2 (area of the oblique line part S1 shown in Fig. 3 (a)).Equally, the interframe movement amount S2 in moment T2 for example is set to Δ t2 in the time interval of moment T1 and moment T2, and the motion characteristic value among the T2 is set under the situation of Δ d2 constantly, is Δ t2 Δ d2/2 (area of the oblique line part S2 shown in Fig. 3 (a)).
Below, the method for calculating evaluation of estimate based on the interframe movement amount is described with reference to Fig. 3 (b).103 pairs of evaluation of estimate computing units are included in the whole two field picture in the incoming video signal, calculate the moment Tn (n: positive integer of each two field picture.Below the same) in the mean value of motion smoothing.If specifically, then be the interframe movement amount Sn that evaluation of estimate computing unit 103 uses in the moment of each two field picture Tn, calculate the evaluation of estimate of the flatness of the motion in moment Tn.If more particularly, then be that evaluation of estimate computing unit 103 is for example used following formula, the α/Sn (α is a constant) that has used interframe movement amount Sn, calculate the evaluation of estimate of the flatness of the motion in moment Tn.In addition, evaluation of estimate computing unit 103 is for example with the following formula that has used interframe movement amount Sn, a * exp -bSn+ c (a, b, c are constants) can calculate the evaluation of estimate of the flatness of the motion in moment Tn.Evaluation of estimate computing unit 103 is based on each evaluation of estimate that calculates, and calculates the mean value of evaluation of estimate of the flatness of the motion in whole moment of input video.This mean value outputs to the outside as the final evaluation of estimate 108 of input video.
Below, the flow process that the video evaluation in the video evaluation device 10 of distortion example of execution mode 1 is handled is described with reference to Fig. 4.
At first, displacement detecting unit 101 orders are read in 2 the continuous two field pictures (two field picture P0, two field picture P1) (step S401) that decomposed from incoming video signal 104.
Below, displacement detecting unit 101 is divided into the two field picture P1 that reads in the piece (step S402) of institute's sizing.
Below, displacement detecting unit 101 is searched the picture signal pattern (step S403) the most similar to the picture signal pattern of this each piece for each piece of two field picture P1 from two field picture P0.
Below, displacement detecting unit 101 is based on by searching the signal pattern that is judged as two the most similar images, and the spatial displacement that detects between this picture signal pattern is motion vector (MVx, MVy) (step S404).This motion vector that is detected outputs to eigenvalue calculation unit 102 as displacement 106.
Below, eigenvalue calculation unit 102 uses the motion vector of each piece that is included in the two field picture P1 in the displacement 106, asks the size (step S405) of motion vector of each piece of two field picture P1.
Below, eigenvalue calculation unit 102 calculates the value (step S406) of the feature that becomes two field picture P1 based on the size of each motion vector.This value that calculates outputs to evaluation of estimate computing unit 103 as motion characteristic value 107.
Below, evaluation of estimate computing unit 103 is according to the motion characteristic value 107 that receives from eigenvalue calculation unit 102, based on the time interval between the two field picture of the frame rate information 105 that receives from the outside, and the motion vector that calculates between each two field picture is interframe movement vector (step S407).
Below, evaluation of estimate computing unit 103 uses the mean value (step S408) of the flatness of the motion of interframe movement vector calculation in the moment of each two field picture Tn for the whole two field picture that is included in the incoming video signal.
Below, evaluation of estimate computing unit 103 is based on each evaluation of estimate that calculates, and calculates the mean value (step S409) of evaluation of estimate of the flatness of the motion in whole moment of input video.
Below, the mean values conduct that 103 of evaluation of estimate computing units calculate outputs to outside (step 410) to the evaluation of estimate 108 of the flatness of all motions of incoming video signal 104.
As mentioned above,, then detect motion vector, calculate motion characteristic value based on the size of this motion vector based on a plurality of two field pictures that are included in the incoming video signal if adopt the video evaluation device 10 of the distortion example of execution mode 1.In addition, according to this motion characteristic value and based on the time interval between the two field picture of the frame rate of incoming video signal, calculate the evaluation of estimate of the motion smoothing that is used to estimate incoming video signal.Thereby, the flatness of the video motion in the frame rate in the time of can correspondingly estimating Video processing with the size of the motion vector of the incoming video signal of according to the rules frame rate input.In addition, because output to the outside, so can externally determine frame rate according to this evaluation of estimate estimating resulting evaluation of estimate by this.That is, can determine frame rate according to the flatness of video motion.
In addition, adopt the unit of searching of the picture signal pattern of above-mentioned displacement detecting unit 101 to be not limited to above-mentioned block unit.For example, also can be frame unit, 1 pixel unit, target area unit etc.In addition, adopt the processing method of searching of displacement detecting unit 101 to be not limited to above-mentioned piece coupling.For example, also can be E-test etc.
In addition, displacement detecting unit 101 also can read in the incoming video signal 104 of the motion vector that comprises video as above-mentioned incoming video signal 104.In this case, displacement detecting unit 101 detects motion vector from the incoming video signal 104 that is received by the outside.This detected motion vector is outputed to eigenvalue calculation unit 102 as displacement 106.
In addition, the motion characteristic value 107 that calculates with eigenvalue calculation unit 102 is not limited to maximum, mean value, median or the minimum value of the size of the motion vector of trying to achieve at above-mentioned each piece.For example, can be the size of 1 motion vector of trying to achieve at each two field picture, also can be maximum, mean value, median or the minimum value of the size of the motion vector of trying to achieve at per 1 pixel in the two field picture or each target area.
In addition, it must be 1 that the motion characteristic value 107 that is calculated by eigenvalue calculation unit 102 there is no need in frame unit, for example, also can be 1 in block unit, pixel unit or target unit.In addition, the distribution of the motion vector that motion characteristic value 107 is tried to achieve for block unit, pixel unit or target unit, can be in order to initial point shown in Figure 5 be a plurality of circles at center and the scope R that divides from many lines of the radial extension of initial point (for example, R1, R2 R3) in the unit is 1.
At this, for the method for calculating motion characteristic value 107 in above-mentioned scope R unit, the situation so that 1 two field picture is divided into 9 pieces is that example specifically describes with reference to Fig. 5.At first, on divided 9 block units, ask 1 motion vector.This each motion vector of trying to achieve as motion vector V1～V9.Below, each motion vector V1～V9 is projected on the curve shown in Figure 5.For example, suppose that motion vector V1～V4 is included among the scope R1 shown in Figure 5, motion vector V5, V6 are included on the scope R2 shown in Figure 5, and motion vector V7～V9 is included among the scope R3 shown in Figure 5.In this case, for example, the motion vector VR1 that goes out as the mean value calculation of motion vector V1～V4 is tried to achieve as the motion characteristic value 107 of scope R1, motion vector VR2 as the mean value calculation of motion vector V5, V6 is tried to achieve as the motion characteristic value 107 of scope R2, is tried to achieve as the motion characteristic value 107 of scope R3 as the motion vector VR3 of the mean value calculation of motion vector V7～V9.And then, ask the method for motion vector to be not limited to the block unit of example, for example, also can be pixel unit or target unit.
In addition, the deduction method of the interframe movement amount of employing evaluation of estimate computing unit 103 has been not limited to use the method for above-mentioned formula (for example, Δ t1 Δ d1/2).For example, can be that (MVx, MVy) and the following formula of representing based on the time interval between the two field picture of frame rate information 105, β MVx Δ t1/2+ γ MVy Δ t1/2 (beta, gamma is a constant) infers motion vector also by using displacement 106.
In addition, to there is no need in frame unit must be 1 to the interframe movement amount of being inferred with evaluation of estimate computing unit 103.Can in block unit, pixel unit, target unit or above-mentioned scope R (with reference to Fig. 5) unit, it be 1 also for example.
In addition, the evaluation of estimate of the flatness of the motion in the moment Tn of each two field picture that is calculated by evaluation of estimate computing unit 103 is not limited to above-mentioned α/Sn (α is a constant) or a * exp -bSn+ c (a, b, c are constants) calculates.For example, also can use displacement 106 is that (MVx MVy) and based on the function in the time interval between the two field picture of frame rate information 105 calculates motion vector.
In addition, the final evaluation of estimate 108 of the input video picture that calculates with evaluation of estimate computing unit 103 is not limited to the mean value of evaluation of estimate of the flatness of the motion in whole moment of above-described input video.For example, also can be maximum, median or the minimum value of evaluation of estimate of the flatness of the motion in whole moment of input video.
In addition, the final evaluation of estimate 108 of the input video that calculates with evaluation of estimate computing unit 103 must be 1 for being included in that whole two field picture in the incoming video signal there is no need.For example, can it be 1 also at each of several two field pictures, 1 two field picture, piece, pixel, target or above-mentioned scope R (with reference to Fig. 5).
At last, be used to make computer as above-mentioned video evaluation device 10 and the video evaluation program 50 of performance function with reference to Fig. 6 explanation.
As shown in Figure 6, video evaluation program 50 comprises: the primary module program 501 of aggregation process; Displacement detection module 502; Characteristic value calculating module 503; Evaluation of estimate computing module 504.The function that displacement computing module 502, characteristic value calculating module 503 and evaluation of estimate computing module 504 have computer is identical with the function that above-mentioned displacement detecting unit 101, eigenvalue calculation unit 102 and evaluation of estimate computing unit 103 has respectively.
And then video evaluation program 50 is for example provided by storage medium or the semiconductor memory of CD-ROM, DVD or ROM etc.In addition, video evaluation program 50 also can be used as the computer data signal that overlaps on the carrier wave and is provided via network.
In addition, the video evaluation device 10 unified displacement detecting unit 101 and the eigenvalue calculation unit 102 of the distortion example of execution mode 1, by being rearranged into the displacement detecting unit, can be arranged to execution mode 1 in the function that has of video evaluation device 140 constitute identical functions and constitute.
[execution mode 2]
Below, embodiments of the present invention 2 are described.Fig. 7 is that the frame rate that is illustrated in the execution mode 2 is determined the figure that the function of device 70 constitutes.
At this, frame rate determines that device 70 is the storage devices that physically possess CPU (central processing unit), memory etc., and the computer of communicator etc.Thereby frame rate determines that device 70 can be the fixed communication terminal of PC terminal etc., also can be the mobile communication terminal of mobile phone etc.That is, determine that as frame rate device 70 can be widely applicable for the device that can carry out information processing.
Illustrate that with reference to Fig. 7 frame rate determines that the function of device 70 constitutes.As shown in Figure 7, frame rate determines that device 70 comprises: frame rate generating unit 701; Video evaluation unit 702; Frame rate determining unit 703.
The 1st frame rate 705 takes place in frame rate generating unit 701.Frame rate generating unit 701 outputs to video evaluation unit 702 and frame rate determining unit 703 to the 1st frame rate 705 that takes place.
Video evaluation unit 702 has the function that has with above-mentioned execution mode 1 described video evaluation device 140, perhaps the function identical functions that has of the described video evaluation device 10 of the distortion example of execution mode 1.Promptly, video evaluation unit 702 has the function that has with above-mentioned change amount detection 1401 and evaluation of estimate computing unit 1402, perhaps the function identical functions that has of displacement detecting unit 101, eigenvalue calculation unit 102 and evaluation of estimate computing unit 103.
Video evaluation unit 702 is from by the incoming video signal 704 of outside as motion image signal input, to read in two field picture with the 705 corresponding time intervals of the 1st frame rate, calculates the evaluation of estimate 706 of the smoothing of all motions of relative incoming video signal 704.Video comments unit 702 that the evaluation of estimate 706 of the flatness of the motion that calculates is outputed to frame rate determining unit 703.
At this, there is not a special problem even the sampling rate of the 1st frame rate 705 and incoming video signal 704 is different yet.For example, the 1st frame rate 705 shown in Fig. 8 (a) is 1/15 second relatively, and the sampling rate of the incoming video signal 704 shown in Fig. 8 (b) is 1/30 second.
Frame rate determining unit 703 is determined the 2nd frame rate 707 based on the evaluation of estimate 706 that receives from video evaluation unit 702 with from the 1st frame rate 705 that frame rate generating unit 701 receives.Frame rate determining unit 703 outputs to the frame rate of fixed the 2nd frame rate 707 as the processing that is used to carry out incoming video signal 704 frame rate and determines device 70.
If specifically describe, then frame rate determining unit 703 for example evaluation of estimate 706 greater than during fixed setting, make the 2nd frame rate 707 less than the 1st frame rate 705.In addition, frame rate determining unit 703 for example evaluation of estimate 706 less than during fixed setting, make the 2nd frame rate 707 greater than the 1st frame rate 705.In addition frame rate determine device 70 for example evaluation of estimate 706 with fixed setting when consistent, the 2nd frame rate 707 is arranged to the frame rate the same with the 1st frame rate 705.
If like this, then can determine evaluation of estimate as the flatness of the motion that is used to estimate video converge on the 2nd frame rate in the scope of fixed metewand.That is, Yi Bian can be on one side the flatness of the motion of video remain on the fixed interior incoming video signal that reads in of reference range.In addition, above-mentioned fixed setting can be predefined value, also can be the value that gives from the outside.
Below, illustrate that with reference to Fig. 9 the frame rate of execution mode 2 determines the flow process that the frame rate in the device 70 determine to be handled.
At first, the 1st frame rate 705 (step S901) takes place in frame rate generating unit 701.
Below, video evaluation unit 702 from incoming video signal 704 to read in two field picture (step 902) with the 705 corresponding time intervals of the 1st frame rate.
Below, video evaluation unit 702 is based on each two field picture, calculates the evaluation of estimate 706 (step 903) of the flatness of all motions of relative incoming video signal 704.Promptly, based on each two field picture, handle (with reference to Figure 17) by the video evaluation that carries out the step S1701～S1707 of explanation in above-mentioned execution mode 1, perhaps carry out the video evaluation of step S401～S410 of in the distortion example of above-mentioned execution mode 1, illustrating and handle (with reference to Fig. 4), calculate the evaluation of estimate 706 of all motion smoothings of artificial relatively vision signal 704.
Below, frame rate determining unit 703 is determined the 2nd frame rate 707 (step 904) based on the evaluation of estimate 706 that receives from video evaluation unit 702, from the 1st frame rate 705 that frame rate generating unit 701 receives.
Below, frame rate determining unit 703 outputs to outside (step 905) to the 2nd frame rate 707 as the frame rate of the processing that is used to carry out incoming video signal 704.
As mentioned above, determine device 70 if adopt the frame rate of execution mode 2, then according to the variable quantity of incoming video signal and based on the time interval between the two field picture of the 1st frame rate, calculating is used to estimate the evaluation of estimate of flatness of the motion of incoming video signal, determines to utilize this evaluation of estimate to carry out the 2nd frame rate of the processing of incoming video signal.That is,, use this evaluation to determine the 2nd frame rate simultaneously according to according to the variable quantity evaluation of the incoming video signal of the 1st frame rate input motion smoothing at the video of the 1st frame rate.
In addition, detect motion vector, calculate motion characteristic value based on the size of this motion vector based on a plurality of two field pictures that are included in the incoming video signal.Then,, calculate the evaluation of estimate of the flatness of the motion be used to estimate incoming video signal, utilize this evaluation of estimate to determine to carry out the 2nd frame rate of the processing of incoming video signal according to this motion characteristic value and based on the time interval between the two field picture of the 1st frame rate.That is,, in the motion smoothing of the video of the 1st frame rate, use this evaluation to determine the 2nd frame rate in evaluation according to the motion characteristics amount of the incoming video signal of importing according to the 1st frame rate.
Thereby, when can determine the frame rate of incoming video signal, can in the motion smoothing that keeps video, read in incoming video signal according to the motion smoothing of video.
At last, be used to make computer to determine device 70 with reference to Figure 10 explanation and the frame rate of performance function is determined program 100 as above-mentioned frame rate.
As shown in figure 10, frame rate determines that program 100 comprises: the primary module program 1001 of aggregation process; Frame rate generation module 1002; Video evaluation module 1003; Frame rate determination module 1004.The function that frame rate generation module 1002, video evaluation module 1003 and frame rate determination module 1004 have computer is identical with the function that above-mentioned frame rate generating unit 701, video evaluation unit 702 and frame rate determining unit 703 has respectively.
And then frame rate determines that program 100 is for example provided by storage medium or the semiconductor memory of CD-ROM, DVD or ROM etc.In addition, frame rate determines that program 100 also can be used as the computer data signal that overlaps on the carrier wave and is provided via network.
[execution mode 3]
Below, embodiments of the present invention 3 are described.Figure 11 is the figure that the function of the video process apparatus 110 in the illustrated embodiment 3 constitutes.
At this, video process apparatus 110 physically is the storage device that possesses CPU (central processing unit), memory etc., and the computer of communicator etc.Thereby video process apparatus 110 can be the fixed communication terminal of PC terminal etc., also can be the mobile communication terminal of mobile phone etc.That is, can be widely applicable for the device that can carry out information processing as video process apparatus 110.
The function that video process apparatus 110 is described with reference to Figure 11 constitutes.As shown in figure 11, video process apparatus comprises: buffer cell 1101; Frame rate determining unit 1102; Video processing unit 1103.
Buffer cell 1101 temporarily remains on the incoming video signals of importing as motor message from the outside 1104 on the buffering area on the memory.Buffer cell 1101 outputs to frame rate determining unit 1102 to incoming video signal 1104.In addition, the later video processing unit of narrating 1103 of incoming video signal 1104 usefulness that temporarily is kept on the buffering area reads in as handling vision signal 1106.
Frame rate determining unit 1102 has with above-mentioned execution mode 2 described frame rate determines the function identical functions that device 70 has.That is, frame rate determining unit 1102 has the function identical functions that has with above-mentioned frame rate generating unit 701, video evaluation unit 702 and frame rate determining unit 703.Frame rate determining unit 1102 is used the incoming video signal 1104 that receives from buffering unit 1101, calculates and the suitable Video processing frame rate 1105 of above-mentioned the 2nd frame rate 707.Frame rate determining unit 1102 outputs to video processing unit 1103 to the Video processing frame rate 1105 that calculates.
Video processing unit 1103 is handled vision signal 1106 to read in from buffering unit 1101 with the 1105 corresponding time intervals of Video processing frame rate, carries out Video processing based on this processing vision signal 1106.If specifically, then in the two field picture of video processing unit 1103 in being contained in the processing vision signal 1106 that is stored in the buffer cell 1101, only read in the two field picture with Video processing frame rate 1105 moment corresponding that receive from frame rate determining unit 1102, carry out Video processing.Video processing unit 1103 is outputing to the outside by the video that Video processing generated.In addition, the Video processing of being undertaken by video processing unit 1103 for example is equivalent to obtaining, store, transmit, encode, deciphering of video.
Below, the flow process of the Video processing in the video process apparatus 110 of execution mode 3 is described with reference to Figure 12.
At first, buffer cell 1101 temporarily is kept at this incoming video signal 1104 on the buffering area on the memory (step S1201) when the incoming video signal 1104 imported as motor message from the outside is outputed to frame rate determining unit 1102.
Below, frame rate determining unit 1102 is used the incoming video signal 1104 that receives from buffering unit 1101, calculates Video processing frame rate 1105 (steps 1202).That is, determine to handle (with reference to Fig. 9), calculate Video processing frame rate 1105 as the 2nd frame rate by the frame rate of using incoming video signal 1104 to carry out the step S901～S905 of explanation in above-mentioned execution mode 2.
Below, video processing unit 1103 is handled vision signal 1106 to read at interval with Video processing frame rate 1105 time corresponding from buffering unit 1101, carry out Video processing (step S1203) based on this processing vision signal 1106.
Below, video processing unit 1103 outputs to outside (step 1204) to the video that is generated by Video processing.
As mentioned above, if adopt the video process apparatus 110 of execution mode 3, then according to the variable quantity of incoming video signal and based on the time interval between the two field picture of the 1st frame rate, calculating is used to estimate the evaluation of estimate of flatness of the motion of incoming video signal, determines to utilize this evaluation of estimate to carry out the 2nd frame rate of the processing of incoming video signal.Then, use the 2nd frame rate to carry out the Video processing of incoming video signal.That is, in the flatness of basis, use this evaluation to determine the 2nd frame rate according to the motion of the video of variable quantity evaluation in the 1st frame rate of the incoming video signal of the 2nd frame rate input.Then, use, carry out the Video processing of incoming video signal based on respect to determined the 2nd frame rate of the evaluation of estimate of incoming video signal.
In addition, detect motion vector, calculate motion characteristic value based on the size of this motion vector based on a plurality of two field pictures that are included in the incoming video signal.Then,, calculate the evaluation of estimate of the flatness of the motion be used to estimate incoming video signal, utilize this evaluation of estimate to determine to carry out the 2nd frame rate of the processing of incoming video signal according to this motion characteristic value and based on the time interval between the two field picture of the 1st frame rate.And then, use the 2nd frame rate to carry out the Video processing of incoming video signal.That is, when estimating the motion smoothing of the video in the 1st frame rate, use this evaluation to determine the 2nd frame rate according to the motion characteristics value of the incoming video signal of importing according to the 1st frame rate.Then, use, carry out the Video processing of incoming video signal based on respect to determined the 2nd frame rate of the evaluation of estimate of incoming video signal.
Thereby, based on determined the 2nd frame rate of evaluation, can carry out the Video processing of input signal according to the flatness of moving.That is, can be when determining frame rate, while keep the flatness of the motion of video to carry out the Video processing of incoming video signal according to the flatness of the motion of video.
And then above-mentioned video process apparatus 110 is for example applicable to video acquisition device (for example camera), video conveyer, video coding apparatus, perhaps video decoding apparatus.
At first, video process apparatus 110 is being applicable under the situation of video acquisition device, each above-mentioned function has the function of the following stated.Buffer cell 1101 will utilize sampling rate, and (incoming video signal of for example, 30fps) being imported 1104 cushions.Frame rate determining unit 1102 is used the incoming video signal 1104 that receives from buffering unit 1101, and the Video processing frame rate 1105 of the best when calculating video acquisition device is obtained video (for example, 15fps), outputs to video processing unit 1103.Video processing unit 1103 uses the Video processing frame rate 1105 that receives from frame rate determining unit 1102, and (for example, 15fps), sampling is maintained at the incoming video signal 1104 in the buffer cell 1101.Video processing unit 1103 (for example, 15fps) is obtained the processing vision signal 1106 that is obtained by this sampling by Video processing frame rate 1105.
In addition, when video process apparatus 110 being applicable under the situation of video storage device, above-mentioned each function has the function of the following stated.Buffer cell 1101 will utilize sampling rate, and (incoming video signal of for example, 30fps) being imported 1104 cushions.Frame rate determining unit 1102 is used the incoming video signal 1104 that receives from buffering unit 1101, and the Video processing frame rate 1105 of the best when calculating video storage device storage video (for example, 15fps), outputs to video processing unit 1103.Video processing unit 1103 uses the Video processing frame rate 1105 that receives from frame rate determining unit 1102, and (for example, 15fps), sampling is maintained at the incoming video signal 1104 in the buffer cell 1101.Video processing unit 1103 (for example, 15fps) stores the processing vision signal 1106 that is obtained by this sampling with Video processing frame rate 1105.
In addition, when video process apparatus 110 being adapted under the situation of video conveyer, each above-mentioned function has the function of the following stated.Buffer cell 1101 will utilize sampling rate, and (incoming video signal of for example, 30fps) being imported 1104 cushions.Frame rate determining unit 1102 is used the incoming video signal 1104 that receives from buffering unit 1101, and optimum Video processing frame rate 1105 (for example, 15fps), outputed to video processing unit 1103 when calculating transmitted video at the video conveyer.Video processing unit 1103 uses the Video processing frame rate 1105 that receives from frame rate determining unit 1102, and (for example, 15fps), sampling is stored in the incoming video signal 1104 in the buffer cell 1101.Video processing unit 1103 (for example, 15fps) transmits the processing vision signal 1106 that is obtained by this sampling with Video processing frame rate 1105.
In addition, video process apparatus 110 is being adapted under the situation of video coding apparatus, above-mentioned each function has the function of the following stated.Buffer cell 1101 will utilize sampling rate, and (incoming video signal of for example, 30fps) being imported 1104 cushions.Frame rate determining unit 1102 is used the incoming video signal 1104 that receives from buffering unit 1101, the Video processing frame rate 1105 of the best when calculating the video coding apparatus encoded video (for example, 15fps), outputs to video processing unit 1103.Video processing unit 1103 uses the Video processing frame rate 1105 that receives from frame rate determining unit 1102, and (for example, 15fps), sampling is stored in the incoming video signal 1104 on the buffer cell 1101.Video processing unit 1103 is with Video processing frame rate 1105 (for example, the processing vision signal 1106 of 15fps) encoding and being obtained by this sampling.
In addition, video process apparatus 110 is being adapted under the situation of video decoding apparatus, above-mentioned each function has the function of the following stated.Buffer cell 1101 will utilize sampling rate, and (incoming video signal of for example, 30fps) being imported 1104 cushions.Frame rate determining unit 1102 is used the incoming video signal 1104 that receives from buffering unit 1101, and the Video processing frame rate 1105 of calculating the best when video decoding apparatus decoding video (for example, 15fps), outputs to video processing unit 1103.Video processing unit 1103 uses the Video processing speed 1105 that receives from frame rate determining unit 1102, and (for example, 15fps), sampling is stored in the incoming video signal 1104 in the buffer cell 1101.Video processing unit 1103 is with Video processing frame rate 1105 (for example, 15fps) the processing vision signal 1106 that obtained by this sampling of decoding.
At last, explanation is used to make the video processing program 130 of computer as above-mentioned video process apparatus 110 performance functions with reference to Figure 13.
As shown in figure 13, video processing program 130 comprises: the primary module program 1301 of aggregation process; Buffer module 1302; Frame rate determination module 1303; Video processing module 1304.Buffer module 1302, frame rate determination module 1303 and video processing module 1304 make the function of computer performance identical with the function that above-mentioned buffer cell 1101, frame rate determining unit 1102 and video processing unit 1103 has respectively.
And then video processing program 130 for example provides with storage medium or the semiconductor memory of CD-ROM, DVD or ROM etc.In addition, video processing program 130 can be used as the computer data signal that overlaps on the carrier wave and is provided via network.

Claims (6)

1. video evaluation device is characterized in that comprising:
The variable quantity detection part, based in a plurality of two field pictures that are included in the incoming video signal based on two continuous two field pictures of the frame rate information that receives from the outside, detect the intensity of variation that is illustrated in the brightness value between above-mentioned continuous two two field pictures or the variable quantity of motion vector;
The evaluation of estimate calculating unit, based on by the time interval between the detected variable quantity of above-mentioned variable quantity detection part and above-mentioned continuous two two field pictures, calculate the motion that is used to estimate above-mentioned incoming video signal flatness evaluation of estimate and it is outputed to the outside,
Above-mentioned evaluation of estimate calculating unit is based on the time interval between above-mentioned variable quantity and above-mentioned continuous two two field pictures, infer with above-mentioned variable quantity and above-mentioned continuous two two field pictures between time interval time corresponding variable quantity, use this time variation amount to calculate above-mentioned evaluation of estimate.
2. according to the video evaluation device of claim 1, it is characterized in that:
Above-mentioned variable quantity detection part detects the motion vector between above-mentioned two two field pictures continuously based on two continuous in a plurality of two field pictures that are included in incoming video signal two field pictures,
Described video evaluation device also comprises:
The eigenvalue calculation parts based on the size by the detected motion vector of above-mentioned variable quantity detection part, calculate the motion characteristic value of the motion characteristics of the above-mentioned incoming video signal of expression,
Above-mentioned evaluation of estimate calculating unit calculates above-mentioned evaluation of estimate based on the time interval between above-mentioned motion characteristic value and above-mentioned continuous two two field pictures.
3. a frame rate is determined device, it is characterized in that comprising:
The frame rate generation part of the 1st frame rate takes place;
The variable quantity detection part, based in a plurality of two field pictures that are included in the incoming video signal based on two continuous two field pictures of above-mentioned the 1st frame rate, detect the intensity of variation of the brightness between above-mentioned continuous two two field pictures of expression or the variable quantity of motion vector;
The evaluation of estimate calculating unit based on the time interval between above-mentioned variable quantity and above-mentioned continuous two two field pictures, calculates the evaluation of estimate of the flatness of the motion that is used to estimate above-mentioned incoming video signal; With
Frame rate is determined parts, uses above-mentioned evaluation of estimate and above-mentioned the 1st frame rate, determine to carry out above-mentioned incoming video signal processing the 2nd frame rate and it is outputed to the outside,
Above-mentioned evaluation of estimate calculating unit is based on the time interval between above-mentioned variable quantity and above-mentioned continuous two two field pictures, infer with above-mentioned variable quantity and above-mentioned continuous two two field pictures between time interval time corresponding variable quantity, use this time variation amount to calculate above-mentioned evaluation of estimate.
4. frame rate according to claim 3 is determined device, it is characterized in that:
Above-mentioned frame rate determine parts above-mentioned evaluation of estimate and fixed setting compare, when above-mentioned evaluation of estimate during greater than the afore mentioned rules value, make above-mentioned the 2nd frame rate less than above-mentioned the 1st frame rate, when above-mentioned evaluation of estimate during less than the afore mentioned rules value, make above-mentioned the 2nd frame rate greater than above-mentioned the 1st frame rate, when above-mentioned evaluation of estimate equals the afore mentioned rules value, make above-mentioned the 2nd frame rate equal above-mentioned the 1st frame rate.
5. video process apparatus is characterized in that comprising:
The buffer unit of storage incoming video signal;
The frame rate generation part of the 1st frame rate takes place;
The variable quantity detection part, based in a plurality of two field pictures that are included in the above-mentioned incoming video signal based on two continuous two field pictures of above-mentioned the 1st frame rate, detect the intensity of variation of the brightness between above-mentioned continuous two two field pictures of expression or the variable quantity of motion vector;
The evaluation of estimate calculating unit based on the time interval between above-mentioned variable quantity and above-mentioned continuous two two field pictures, calculates the evaluation of estimate of the motion smoothing that is used to estimate above-mentioned incoming video signal;
Frame rate is determined parts, uses above-mentioned evaluation of estimate and above-mentioned the 1st frame rate, determines to carry out the 2nd frame rate of the processing of above-mentioned incoming video signal; With
The Video processing parts use above-mentioned the 2nd frame rate to read in by the above-mentioned incoming video signal of above-mentioned buffer unit storage and carry out Video processing,
Above-mentioned evaluation of estimate calculating unit is based on the time interval between above-mentioned variable quantity and above-mentioned continuous two two field pictures, infer with above-mentioned variable quantity and above-mentioned continuous two two field pictures between time interval time corresponding variable quantity, use this time variation amount to calculate above-mentioned evaluation of estimate.
6. video evaluation method is characterized in that comprising:
Variable quantity detects step, based in a plurality of two field pictures that are included in the incoming video signal based on two continuous two field pictures of the frame rate information that receives from the outside, detect the intensity of variation that is illustrated in the brightness value between above-mentioned continuous two two field pictures or the variable quantity of motion vector; With
The evaluation of estimate calculation procedure, based on the time interval of detecting at above-mentioned variable quantity between detected variable quantity in the step and above-mentioned continuous two two field pictures, calculate the motion that is used to estimate above-mentioned incoming video signal flatness evaluation of estimate and it is outputed to the outside,
Above-mentioned evaluation of estimate calculation procedure is based on the time interval between above-mentioned variable quantity and above-mentioned continuous two two field pictures, infer with above-mentioned variable quantity and above-mentioned continuous two two field pictures between time interval time corresponding variable quantity, use this time variation amount to calculate above-mentioned evaluation of estimate.
CNB2005101039936A 2004-09-16 2005-09-16 Video evaluation device and method, frame rate determination device, and video process device Active CN100515069C (en)

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
JP2004270123 2004-09-16
JP2004270123 2004-09-16
JP2005229618 2005-08-08

Publications (2)

Publication Number Publication Date
CN1750635A CN1750635A (en) 2006-03-22
CN100515069C true CN100515069C (en) 2009-07-15

Family

ID=36605865

Family Applications (1)

Application Number Title Priority Date Filing Date
CNB2005101039936A Active CN100515069C (en) 2004-09-16 2005-09-16 Video evaluation device and method, frame rate determination device, and video process device

Country Status (2)

Country Link
JP (1) JP5101673B2 (en)
CN (1) CN100515069C (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10764578B2 (en) 2018-04-03 2020-09-01 KKStream Limited Bit rate optimization system and method

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20080022276A (en) * 2006-09-06 2008-03-11 엘지전자 주식회사 Method and apparatus for controlling screen of (an) image display device
JP5315162B2 (en) * 2009-08-05 2013-10-16 株式会社日立製作所 Video processing apparatus and video processing method
CN108322737B (en) * 2018-02-08 2020-02-14 Oppo广东移动通信有限公司 Method and device for measuring camera shooting frame rate
JP7273460B2 (en) * 2018-04-11 2023-05-15 シャープ株式会社 Video processing device, video processing method, television receiver, control program, and recording medium
CN113347487B (en) * 2021-06-17 2022-09-27 微微科技有限公司 Low-power browser sequence diagram video method and terminal for automatic compatible equipment

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6618439B1 (en) * 1999-07-06 2003-09-09 Industrial Technology Research Institute Fast motion-compensated video frame interpolator
US20030169933A1 (en) * 2002-03-09 2003-09-11 Samsung Electronics Co., Ltd. Method for adaptively encoding motion image based on temporal and spatial complexity and apparatus therefor
CN1453936A (en) * 2002-04-25 2003-11-05 松下电器产业株式会社 Image coding apparatus and image coding method
CN1520180A (en) * 2003-02-03 2004-08-11 三星电子株式会社 Image processor having frame speed conversion and its method

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP3558384B2 (en) * 1994-10-12 2004-08-25 三菱電機株式会社 Video recording and playback method
JP3542911B2 (en) * 1998-08-24 2004-07-14 日本電気株式会社 Video compression device
JP2001245303A (en) * 2000-02-29 2001-09-07 Toshiba Corp Moving picture coder and moving picture coding method

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6618439B1 (en) * 1999-07-06 2003-09-09 Industrial Technology Research Institute Fast motion-compensated video frame interpolator
US20030169933A1 (en) * 2002-03-09 2003-09-11 Samsung Electronics Co., Ltd. Method for adaptively encoding motion image based on temporal and spatial complexity and apparatus therefor
CN1453936A (en) * 2002-04-25 2003-11-05 松下电器产业株式会社 Image coding apparatus and image coding method
CN1520180A (en) * 2003-02-03 2004-08-11 三星电子株式会社 Image processor having frame speed conversion and its method

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
A Sliding Window Approach to Real-time H.263+ Frame RateAdjustment. Hwangjun Song, Jongwon Kim,C.-C. Jay Kuo.Conference Record of the Thirty-Second Asilomar Conference on Signals,Systems&Computers,1998.,Vol.1 No.1. 1998
A Sliding Window Approach to Real-time H.263+Frame RateAdjustment. Hwangjun Song, Jongwon Kim,C.-C. Jay Kuo.Conference Record of the Thirty-Second Asilomar Conference on Signals,Systems&Computers,1998.,Vol.1 No.1. 1998 *
A Virtual Frame Rate ControlAlgorithmforEfficientMPEG-2Video Encoding. Byung Cheol Song, Kang Wook Chun.IEEE Transactions on Consumer Electronics,Vol.49 No.2. 2003
Content adaptive frame skipping for low bit rate video coding. Feng Pan, Z.P.Lin, X.Lin,S.Rahardja, William uwono, Ferdinand Slamet.ICICS-PCM 2003,Vol.1 . 2003

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10764578B2 (en) 2018-04-03 2020-09-01 KKStream Limited Bit rate optimization system and method

Also Published As

Publication number Publication date
JP2010268525A (en) 2010-11-25
JP5101673B2 (en) 2012-12-19
CN1750635A (en) 2006-03-22

Similar Documents

Publication Publication Date Title
EP1638338B1 (en) Video evaluation device, frame rate determination device, video process device, video evaluation method, and video evaluation program
CN1041043C (en) Sport calculating method and device
CN100515069C (en) Video evaluation device and method, frame rate determination device, and video process device
US5784115A (en) System and method for motion compensated de-interlacing of video frames
CN100421455C (en) Video electronic flutter-proof method
US20120162450A1 (en) Digital image stabilization device and method
US8588513B2 (en) Method and system for motion compensation
CN101163247A (en) Interpolation method for a motion compensated image and device for the implementation of said method
CN101889452A (en) Image generation device and image generation method
US8630338B2 (en) Method and device for image interpolation systems based on motion estimation and compensation
US6847333B2 (en) Method of and system for low-bandwidth transmission of color video
US7961961B2 (en) Image processing apparatus and image processing program using motion information
CN113691758A (en) Frame insertion method and device, equipment and medium
CN113168502A (en) Dynamic adaptive convolutional neural network for variable computing resources
US8160151B2 (en) Motion vector detection apparatus, motion vector processing method and program
JPS5885684A (en) Forecasting encoding device of television signal
US5731851A (en) Method for determining feature points based on hierarchical block searching technique
EP1627533B1 (en) Biased motion vector interpolation for reduced video artifacts
JPS6028392A (en) Movement interpolation system of dynamic picture signal
CN109005414A (en) Video transmission method, device, equipment and system
US20100266043A1 (en) Design of low power video compression chipset for protable application
JPH0730859A (en) Frame interpolation device
US6735374B1 (en) Picture processing apparatus and method
JP2806332B2 (en) Video motion compensation circuit
JP3799842B2 (en) Static video detection method and apparatus

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant