CN101202912A - Method for controlling balanced code rate and picture quality code rate - Google Patents
Method for controlling balanced code rate and picture quality code rate Download PDFInfo
- Publication number
- CN101202912A CN101202912A CN 200710171620 CN200710171620A CN101202912A CN 101202912 A CN101202912 A CN 101202912A CN 200710171620 CN200710171620 CN 200710171620 CN 200710171620 A CN200710171620 A CN 200710171620A CN 101202912 A CN101202912 A CN 101202912A
- Authority
- CN
- China
- Prior art keywords
- frame
- psnr
- bits
- bit
- picture quality
- 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.)
- Pending
Links
Images
Abstract
The invention provides a code rate control method for balancing a code rate and image quality; wherein, the method firstly initializes coding parameters and then makes comparison of the peak signal noise ratios and the deviation values of bit uses of I frame, P frame and B frame with the respective value ranges thereof, so as to judge whether the rations and the deviation values meet relative balance conditions in a balancing way; the method then adjusts the coding parameters (such as quantization parameters, etc.) according to the judging result for a plurality of times and returns the adjusted parameters to recode; the method can use as much code rate bandwidths as possible in the condition of ensuring little changes of image quality, and that is to say, the method better improves the compression quality of videos with the same coding rate. The code rate control method can be successfully applied in an AVS coding algorithm.
Description
Technical field
The present invention relates to the audio/video encoding/decoding technical field, relate in particular to a kind of bit rate control method of video coding.
Background technology
In recent years a collection of digital audio/video encoding and decoding standard of Chu Xianing, representative have international standard H.264/MPEG-4 AVC and an autonomous standard A VS that formulates of China etc., and these standards all relate to Rate Control.Because the code check of code stream during actual transmissions is used (Constant BitRate is called for short CBR) is constant, can guarantee the stable of transport stream.Because the bit rate that most video compression technologies or standard itself produce all is variable, in order can in the communication channel of constant bit rate and network, to transmit, traditional rate control algorithm need be introduced a buffering area mechanism in Rate Control, promptly deposit buffering area in after the code stream compression with variable bit rate earlier, from buffering area, export then with constant code rate.In this method, the task of Rate Control is exactly to guarantee that buffer does not produce under the prerequisite of overflow and underflow, makes coding bit rate reach desired value by adjusting some coding parameters (as quantization parameter).But in fact, when altering a great deal, picture quality is easy to generate the code check peak value, the uncontrollable code check of buffering area, thereby influence the stable of transport stream; In addition, image also may reach good quality under the very low situation of code check, therefore also is easy to generate the code check valley.Therefore, the balance of how to accomplish code check and picture quality is the major issue that Rate Control will be considered.
Summary of the invention
Technical problem to be solved by this invention provides the bit rate control method of a kind of balanced code rate and picture quality, this method can repeatedly be adjusted some coding parameters (as quantization parameter etc.) according to the multi-frame mean quality of coded image and the picture quality and the employed allocation bit ratio of reference frame before, to a certain extent balance code check and picture quality.
In order to solve the problems of the technologies described above, the present invention has adopted following technological means: the bit rate control method of a kind of balanced code rate and picture quality, be used for the code check of code stream is controlled, and described code stream comprises I frame, P frame and B frame, described method comprises the steps:
The coded quantization parameter and the scale factor of step 1, initialization I, P and each frame of B, loop initialization coding frequency n=0;
Step 2, according to initial quantization parameter each frame is encoded, the Y-PSNR and the bit that obtain each frame use deviate;
Step 3, the Y-PSNR of I frame, P frame and B frame and bit are used deviate and separately span compare respectively, equilibrium judges whether to satisfy the relational balance condition;
Step 4, if picture quality and bit number all satisfy equilibrium condition in the step 3, then finish Rate Control, determine the coding parameter of present frame; If picture quality and bit number have at least one not satisfy equilibrium condition in the step 3, determine to increase or the lower quantization parameter, obtain the changing value of new quantization parameter, upgrade coding parameter, continue step 5;
The size of step 5, judgement n and actual maximum cycle N when n equals N, selects one figure of merit as this code check output according to the comparative result of step 3; As n during less than N, jump to step 2, to each frame coding, n adds 1 again.
Further, the described initialization of step 1 comprises:
A) initialization quantization parameter QP, the interval of QP are [0,63];
B) the scale factor IP_factor of initialization I frame and P frame, its interval can be (1,6];
C) the scale factor PB_factor of initialization P frame and B frame, its interval can be (1,6];
D) at video sequence, adopt half second image sets, determine frame per second fps;
E) set constant code rate Rate_CBR according to actual needs, the bit rate Rate_buf that every frame is average, its computing formula is as follows: Rate_buf=Rate_CBR/fps;
F) each frame of pattern-coding of employing IPBBPBB promptly has 1 I frame, 4 P frames and 10 B frames in 15 frames; The frame number of setting I frame among the image sets GOP is I_frames; The frame number of P frame is P_frames in image sets; The frame number of B frame is B_frames in image sets;
G) the bit number Bits_GOP that calculates each image sets needs according to the big or small GOP_size and the average bit rate Rate_buf of every frame of an image sets, its computing formula is as follows:
Bits_GOP＝GOP_size
*Ratef_buf；
H) according to the frame number of each frame and the bit number of each frame of scale factor calculation, i.e. Bits_B, Bits_P and Bits_I, h.1) for first GOP, its computing formula is as follows:
Bits_B＝Bits_GOP/(I_frames
*IP_factor
*PB_factor+P_frames
*PB_factor+B_frames)；
Bits_P＝Bits_B
*PB_factor；
Bits_I＝Bits_B
*IP_factor
*PB_factor；
H.2) for non-first GOP, the code check that actual coding obtains is that previous GOP bit number is removed the I frame bit number of adding next GOP behind the bit number of previous I frame and obtained, and the computing formula of Bits_B, Bits_P and Bits_I is as follows:
Bits_B＝(Bits_GOP-Used_B_Bits-Used_P_Bits)/(I_frames
*IP_factor
*PB_factor+P_frames
*PB_factor+B_frames)；
Bits_P＝Bits_B
*PB_factor；
Bits_I＝Bits_GOP-Used_B_Bits-Used_P_Bits；
Wherein, Used_B_Bits represents B frame bit number encoded among the current GOP, and Used_P_Bits represents P frame bit number encoded among the current GOP.
Further, the bit described in the step 2 uses deviate Err_bits by following calculating gained: Err_bits=Used_bits
*10/Bits_X, wherein X can be I, P or B.
Step 3 is further, to described I frame code check, the bit number that its bit number should an above I frame as a reference owing to when initialization codes, considered to avoid the I frame to use too much bit number to influence the quality of this GOP subsequent image; Therefore only need in the Rate Control to judge bit uses deviate Err_bits whether to satisfy the Rate Control requirement.Described Err_bits when interval, reaches the Rate Control requirement in [9,10], when Err_bits less than 9 the time, then increase the adjustment amount Delta_QP of quantization parameter, when Err_bits greater than 10 the time, then reduce Delta_QP.
Step 3 is further, and to described P frame code check, if present frame is first P frame in the image sets, this P frame needs the average peak signal to noise ratio of reference to be taken as the average peak signal to noise ratio of a GOP; Otherwise, this P frame needs the average peak signal to noise ratio value of reference to be (Average_PSNR+PSNR_P)/(Pn+1), wherein Average_PSNR is the average peak signal to noise ratio of this image sets, and PSNR_P is the Y-PSNR value of P frame accumulative total, and Pn is a P frame accumulative total number.
To described P frame code check, when the Y-PSNR PSNR of previous frame is designated as Last_PSNR greater than 50 the time, then Reference PSNR gets the little value among Average_PSNR and the Last_PSNR; When Last_PSNR less than 50, then Reference_PSNR gets the big value among Average_PSNR and the Last_PSNR.
To described P frame code check, select PSNR to allow the value Delta_PSNR of change according to the present scope of Reference_PSNR, along with Reference_PSNR diminishes, corresponding Delta_PSNR changes and also reduces; When Reference_PSNR less than 30 the time, Delta_PSNR is changed to 0.
To described P frame code check, the span of the picture quality Current_PSNR of current encoded image frame is [Reference_PSNR-Delta_PSNR, Reference_PSNR+Delta_PSNR]; When Current_PSNR is higher than the span upper limit, then use deviate Err_bits to adjust Delta_QP according to bit, when Err_bits is higher than the P ceiling frame, increase quantization parameter QP, different P frames, its upper limit changes, and excursion is [12,23]; Prescribe a time limit when Err_bits is lower than under the P frame, reduce QP, P frame lower limit also changes, and excursion is [5,10].
To described P frame code check, can be more and more littler along with adjusting Delta_QP; The bit that coding comes out uses the deviate Err_bits can be near 10, and Current_PSNR also within the range of permission, promptly can select best coding parameter.
Step 3 is further, to described B frame code check, at first determine to compare, select wherein smaller value as reference Y-PSNR Reference_PSNR with the average peak signal to noise ratio Previous_PSNR of previous frame or the average peak signal to noise ratio Last_PSNR of previous frame and the average peak signal to noise ratio Average_PSNR value of B frame according to reference frame image.
To described B frame code check, select PSNR to allow the value Delta_PSNR of change according to the present scope of Reference_PSNR, along with Reference_PSNR diminishes, corresponding Delta_PSNR changes and also reduces, when Reference_PSNR less than 30 the time, Delta_PSNR is changed to 0.
To described B frame code check, the span of the picture quality Current_PSNR of current encoded image frame is [Reference_PSNR-Delta_PSNR, Reference_PSNR+Delta_PSNR]; When Current_PSNR is higher than the span upper limit, then use deviate Err_bits to adjust Delta_QP according to bit, the upper limit that Err_bits is higher than the B frame will increase QP, and the scope of the upper limit of B frame is [12,20]; The lower limit that is lower than the B frame as Err_bits will reduce QP, and the lower range of B frame is [2,3].
To described B frame code check, along with the process of adjusting, Delta_QP also can be more and more littler; The bit that comes out when coding uses deviate Err_bits near 10, and Current_PSNR also within the range of permission, then can select best coding parameter.
The value of N described in the step 6 require to be determined that by the coding real-time described time the figure of merit satisfies: Err_bits is less than the upper limit of P frame or B frame, and the maximum of the PSNR that obtains for loop coding repeatedly of PSNR.
The present invention compared with prior art, can guarantee to change under the little situation in picture quality owing to adopted above-mentioned technical scheme, and use code check bandwidth as much as possible promptly under the prerequisite of same code rate, has better improved the compression quality of video.
Description of drawings
The bit rate control method of a kind of balanced code rate of the present invention and picture quality is provided by following embodiment and accompanying drawing.
Fig. 1 is a Rate Control flow chart of the present invention;
Fig. 2 is the Y-PSNR schematic diagram of the reflection image brightness of embodiments of the invention.
Embodiment
Below will the bit rate control method of a kind of balanced code rate of the present invention and picture quality be described in further detail.
The present invention will be described in detail below in conjunction with the Rate Control flow chart of Fig. 1 and the coding example of image foreman.yuv commonly used, and the target bit rate of this image is 1000kbps.The 31st frame is encoded to the I frame in the image of present embodiment, and the 32nd frame and the 33rd frame are encoded to the B frame, and the 34th frame is encoded to the P frame.Its bit rate control method following steps realize:
The coded quantization parameter and the scale factor of initialization I, P and each frame of B, loop initialization coding frequency n=0:
A) initialization first round quantization parameter QP is 25; The scale factor IP_factor value of initialization I frame and P frame is 1.5; The scale factor PB_factor value of initialization P frame and B frame is 1.3;
D) at video sequence, adopt half second GOP, be generally for 30 frame/seconds, promptly frame per second fps (how many frames of frames persecond per second) is 30;
E) set constant code rate Rate_CBR=1000kbps according to actual needs, the bit rate Rate_buf=Rate_CBR/fps=33333 that then every frame is average;
F) each frame of pattern-coding of employing IPBBPBB promptly has 1 I frame, 4 P frames and 10 B frames in 15 frames; The frame number of setting I frame among the GOP is I_frames=1; The frame number of P frame is P_frames=4 among GOP; The frame number of B frame is B_frames=10 among GOP;
G) determine the big or small GOP_size=15 of image sets according to above image sets (GOP); And calculating the bit number of each GOP needs according to the average bit rate Rate_buf of every frame, its computing formula is as follows: Bits_GOP=GOP_size
*Rate_buf=15
*33333=499995;
H) according to the frame number of each frame and the bit number of each frame of scale factor calculation, i.e. Bits_B, Bits_P and Bits_I, the scale factor of this moment is initialized as IP_factor=2, PB_factor=1.5.Here be the 3rd GOP, so Bits_I=Bits_GOP-Used_B_Bits-Used_P_Bits, wherein, Used_B_Bits represents B frame bit number encoded among the current GOP, Used_P_Bits represents P frame bit number encoded among the current GOP, its numerical value is respectively 176344 and 127640, so Bits_I=499995-176344-127640=196011.If this is first GOP, then the computing formula of Bits_I is Bits_I=Bits_B
*IP_factor
*PB_factor.
Choosing wherein, the 31st frame I frame is example, then Err_bits=Used_bits
*10/Bits_I=121488
*10/196011=6 less than 9, adjusts quantization parameter QP, promptly reduces the QP value, gets the adjusted value Delta_QP=-6 of quantization parameter, and the Y-PSNR of present image (PSNR) is 39.7549; Second to take turns QP be 21, and bit uses deviate Err_bits=Used_bits
*10/Bits_I=189522
*10/196011=10 just in time satisfies condition, this moment Delta_QP=0, Rate Control finishes.
Choosing wherein, the P frame of the 34th frame is an example, initialization first round QP is 29, average peak signal to noise ratio Average_PSNR=40.1684, previous frame Y-PSNR Last_PSNR=42.6644, then select bigger among both one with reference to Y-PSNR Reference_PSNR, be Reference_PSNR=42.6644, because Reference_PSNR-41.0＞0.8, experience adjustments Delta_PSNR=3.5, the picture quality Current_PSNR excursion that is to say the current encoded image frame is [42.6644-3.5,42.6644+3.5].The Current_PSNR=38.9971 that the first round obtains, in excursion, Err_bits=29917
*10/31402=9, less than 10, the lower limit that does not satisfy P frame this moment requires 10, therefore reduces QP, and Delta_QP is-1.Second to take turns QP be 28, the Current_PSNR=39.3499 that obtains, and in excursion, Err_bits=34833
*10/31402=11, greater than 10, the conditional request of code check and the requirement of picture quality at this moment can be satisfied, and do not need to regulate Delta_QP, thereby just selects the coding parameter of the QP value 28 of this moment as the best, and Rate Control finishes.
Selecting wherein, the B frame of the 32nd frame is an example, the average peak signal to noise ratio Previous_PSNR=40.1553 of previous frame (going up previous frame), the average peak signal to noise ratio Last_PSNR=42.6644 of previous frame, Average_PSNR=40.9917, reference frame is 1, so the smaller of selection Previous_PSNR and Average_PSNR is as Reference_PSNR.Because Reference_PSNR-40.0＞0.8, experience adjustments Delta_PSNR=3.0 that is to say that the Current_PSNR excursion is [40.1553-3.0,40.1553+3.0].The first round, QP got 25, Current_PSNR=40.0446, Err_bits=29543
*10/16593=18, greater than 10, though this moment quality to reach the condition of requirement code check undesirable, regulate Delta_PSNR, increase QP, get Delta_PSNR=8; Because this is first B frame of GOP, so second to take turns QP be 33, Previous_PSNR=40.1553, Last_PSNR=42.6644, Average_PSNR=40.9917, reference frame are 1, so Reference_PSNR selects the smaller of Previous_PSNR and Average_PSNR.Because Reference_PSNR-40.0＞0.8, experience adjustments Delta_PSNR=3.0 that is to say that the Current_PSNR excursion is [40.1553-3.0,40.1553+3.0].But the Current_PSNR=38.1524 through adjusting for the first time satisfies quality requirements, but Err_bits=5792
*10/16593=3, less than 10, code check satisfies B frame lower range this moment.The B frame mainly satisfies quality and meets the requirements of continuity, and code check can suitably hang down, but can not exceed predefined target bit rate.
Table 1 is the code check of each GOP of realizing after Rate Control, this code check and target bit rate 1000kbps are very approaching.Fig. 2 is the Y-PSNR schematic diagram of present embodiment reflection image brightness, the present invention not only encodes for the AVS SD and the little image encoding of AVS provides a kind of constant code check, and guaranteed the continuity of picture quality on the basis of equal code check, to have obtained best picture quality.
Table 1
| The GOP sequence | Code check |
| First GOP_rate | 1134960 |
| Second GOP_rate | 997155 |
| The 3rd GOP_rate | 968560 |
| The 4th GOP_rate | 975280 |
| The 5th GOP_rate | 976368 |
| The 6th GOP_rate | 963152 |
Claims (16)
1. the bit rate control method of balanced code rate and picture quality is used for the code check of code stream is controlled, and described code stream comprises I frame, P frame and B frame, it is characterized in that, described method comprises the steps:
The coded quantization parameter and the scale factor of step 1, initialization I, P and each frame of B, loop initialization coding frequency n=0;
Step 2, according to initial quantization parameter each frame is encoded, the Y-PSNR and the bit that obtain each frame use deviate;
Step 3, the Y-PSNR of I frame, P frame and B frame and bit are used deviate and separately span compare respectively, equilibrium judges whether to satisfy the relational balance condition;
Step 4, if picture quality and bit number all satisfy equilibrium condition in the step 3, then finish Rate Control, determine the coding parameter of present frame; If picture quality and bit number have at least one not satisfy equilibrium condition in the step 3, determine to increase or the lower quantization parameter, obtain the changing value of new quantization parameter, upgrade coding parameter, continue step 5;
The size of step 5, judgement n and actual maximum cycle N when n equals N, selects one figure of merit as this code check output according to the comparative result of step 3; As n during less than N, jump to step 2, to each frame coding, n adds 1 again.
2. the bit rate control method of balanced code rate as claimed in claim 1 and picture quality is characterized in that, described initialization comprises:
A) initialization quantization parameter QP, the interval of QP are [0,63];
B) the scale factor IP_factor of initialization I frame and P frame, its interval can be (1,6];
C) the scale factor PB_factor of initialization P frame and B frame, its interval can be (1,6];
D) at video sequence, adopt half second image sets, determine frame per second fps;
E) set constant code rate Rate_CBR according to actual needs, the bit rate Rate_buf that every frame is average, its computing formula is as follows: Rate_buf=Rate_CBR/fps;
F) each frame of pattern-coding of employing IPBBPBB promptly has 1 I frame, 4 P frames and 10 B frames in 15 frames; The frame number of setting I frame in the image sets is I_frames; The frame number of P frame is P_frames in image sets; The frame number of B frame is B_frames in image sets;
G) the bit number Bits_GOP that calculates each image sets needs according to the big or small GOP_size and the average bit rate Rate_buf of every frame of an image sets, its computing formula is as follows:
Bits_GOP＝GOP_size
*Rate_buf；
H) according to the frame number of each frame and the bit number of each frame of scale factor calculation, i.e. Bits_B, Bits_P and Bits_I,
H.1) for first image sets, its computing formula is as follows:
Bits_B＝Bits_GOP/(I_frames
*IP_factor
*PB_factor+P_frames
*PB_factor+B_frames)；
Bits_P＝Bits_B
*PB_factor；
Bits_I＝Bits_B
*IP_factor
*PB_factor；
H.2) for non-first image sets, the code check that actual coding obtains is that previous image sets bit number is removed the I frame bit number of adding next image sets behind the bit number of previous I frame and obtained, and the computing formula of Bits_B, Bits_P and Bits_I is as follows:
Bits_B＝(Bits_GOP-Used_B_Bits-Used_P_Bits)/(I_frames
*IP_factor
*PB_factor+P_frames
*PB_factor+B_frames)；
Bits_P＝Bits_B
*PB_factor；
Bits_I＝Bits_GOP-Used_B_Bits-Used_P_Bits；
Wherein, Used_B_Bits represents B frame bit number encoded in the present image group, and Used_P_Bits represents P frame bit number encoded in the present image group.
3. the bit rate control method of balanced code rate as claimed in claim 1 and picture quality is characterized in that: described bit uses deviate Err_bits to calculate gained by following formula:
Err_bits=Used_bits
*10/Bits_X, wherein, Used_bits represents the bit number that this frame has used, X can be I, P or B.
4. the bit rate control method of balanced code rate as claimed in claim 1 and picture quality, it is characterized in that: to described I frame code check, its bit number should an above I frame bit number as a reference, only need in Rate Control judge whether bit use deviate Err_bits satisfies the Rate Control requirement.
5. the bit rate control method of balanced code rate as claimed in claim 4 and picture quality, it is characterized in that: described Err_bits is [9, when 10] interval, reach the Rate Control requirement, when Err_bits less than 9 the time, then increase the adjustment amount Delta_QP of quantization parameter, when Err_bits greater than 10 the time, then reduce Delta_QP.
6. the bit rate control method of balanced code rate as claimed in claim 2 and picture quality, it is characterized in that: to described P frame code check, if present frame is first P frame in the image sets, this P frame needs the average peak signal to noise ratio of reference to be taken as the average peak signal to noise ratio of an image sets; Otherwise, this P frame needs the average peak signal to noise ratio value of reference to be (Average_PSNR+PSNR_P)/(Pn+1), wherein Average_PSNR is the average peak signal to noise ratio of this image sets, and PSNR_P is the Y-PSNR value of P frame accumulative total, and Pn is a P frame accumulative total number.
7. the bit rate control method of balanced code rate as claimed in claim 6 and picture quality, it is characterized in that: to described P frame code check, when the PSNR of previous frame is designated as Last_PSNR greater than 50 the time, then get little value among Average_PSNR and the Last_PSNR with reference to Y-PSNR Reference_PSNR; When Last_PSNR less than 50, then Reference_PSNR gets the big value among Average_PSNR and the Last_PSNR.
8. the bit rate control method of balanced code rate as claimed in claim 7 and picture quality, it is characterized in that: to described P frame code check, select PSNR to allow the value Delta_PSNR of change according to the present scope of Reference_PSNR, along with Reference_PSNR diminishes, corresponding Delta_PSNR changes and also reduces; When Reference_PSNR less than 30 the time, Delta_PSNR is changed to 0.
9. the bit rate control method of balanced code rate as claimed in claim 8 and picture quality, it is characterized in that: to described P frame code check, the span of the picture quality Current_PSNR of current encoded image frame is [Reference_PSNR-Delta_PSNR, Reference_PSNR+Delta_PSNR]; When Current_PSNR is higher than the span upper limit, then adjust the adjustment amount Delta_QP of quantization parameter, when Err_bits is higher than the P ceiling frame according to bit use deviate Err_bits, increase quantization parameter QP, different P frames, its upper limit changes, excursion is [12,23]; Prescribe a time limit when Err_bits is lower than under the P frame, reduce QP, P frame lower limit also changes, and excursion is [5,10].
10. the bit rate control method of balanced code rate as claimed in claim 9 and picture quality is characterized in that: to described P frame code check, and can be more and more littler along with adjusting Delta_QP; The bit that coding comes out uses the deviate Err_bits can be near 10, and Current_PSNR also within the range of permission, promptly can select best coding parameter.
11. the bit rate control method of balanced code rate as claimed in claim 1 and picture quality, it is characterized in that: to described B frame code check, at first determine to compare, select wherein smaller value as reference Y-PSNR Reference_PSNR with the average peak signal to noise ratio Previous_PSNR of previous frame or the average peak signal to noise ratio Last_PSNR of previous frame and the average peak signal to noise ratio Average_PSNR value of B frame according to reference frame image.
12. the bit rate control method of balanced code rate as claimed in claim 11 and picture quality, it is characterized in that: to described B frame code check, select Y-PSNR to allow the value Delta_PSNR of change according to the present scope of Reference_PSNR, along with Reference_PSNR diminishes, corresponding Delta_PSNR changes and also reduces, when Reference_PSNR less than 30 the time, Delta_PSNR is changed to 0.
13. the bit rate control method of balanced code rate as claimed in claim 12 and picture quality, it is characterized in that: to described B frame code check, the span of the picture quality Current_PSNR of current encoded image frame is [Reference_PSNR-Delta_PSNR, Reference_PSNR+Delta_PSNR]; When Current_PSNR is higher than the span upper limit, then use deviate Err_bits to adjust the adjustment amount Delta_QP of quantization parameter according to bit, the upper limit that Err_bits is higher than the B frame will increase quantization parameter QP, and the scope of the upper limit of B frame is [12,20]; The lower limit that is lower than the B frame as Err_bits will reduce QP, and the lower range of B frame is [2,3].
14. the bit rate control method of balanced code rate as claimed in claim 13 and picture quality is characterized in that: to described B frame code check, along with the process of adjusting, Delta_QP also can be more and more littler; The bit that comes out when coding uses deviate Err_bits near 10, and Current_PSNR also within the range of permission, then can select best coding parameter.
15. the bit rate control method of balanced code rate as claimed in claim 1 and picture quality is characterized in that: described N is required to determine by the coding real-time.
16. the bit rate control method of balanced code rate as claimed in claim 1 and picture quality, it is characterized in that: described figure of merit satisfies: bit uses the upper limit of deviate Err_bits less than P frame or B frame, and the maximum of the Y-PSNR that obtains for loop coding repeatedly of Y-PSNR.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200710171620 CN101202912A (en) | 2007-11-30 | 2007-11-30 | Method for controlling balanced code rate and picture quality code rate |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200710171620 CN101202912A (en) | 2007-11-30 | 2007-11-30 | Method for controlling balanced code rate and picture quality code rate |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101202912A true CN101202912A (en) | 2008-06-18 |
Family
ID=39517840
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200710171620 Pending CN101202912A (en) | 2007-11-30 | 2007-11-30 | Method for controlling balanced code rate and picture quality code rate |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101202912A (en) |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102118618A (en) * | 2011-03-30 | 2011-07-06 | 北京世纪鼎点软件有限公司 | Method for realizing H.264 multi-code-rate video stream based on PAFF |
| CN103167318A (en) * | 2011-12-13 | 2013-06-19 | 中国电信股份有限公司 | Streaming media processing method, device, server and system |
| WO2013086724A1 (en) * | 2011-12-15 | 2013-06-20 | Mediatek Singapore Pte. Ltd. | Method of clippling transformed coefficients before de-quantization |
| CN103327284A (en) * | 2012-12-18 | 2013-09-25 | 中兴通讯股份有限公司 | Method, device and video communication system for transmitting lossless video |
| CN105744271A (en) * | 2016-02-17 | 2016-07-06 | 浙江大华技术股份有限公司 | Video coding method and device |
| CN106331704A (en) * | 2015-07-07 | 2017-01-11 | 杭州海康威视数字技术股份有限公司 | Video code rate control method and video coding device |
| CN106713842A (en) * | 2015-11-16 | 2017-05-24 | 晶睿通讯股份有限公司 | Image processing method and camera thereof |
| CN107205151A (en) * | 2017-06-26 | 2017-09-26 | 中国科学技术大学 | Coding and decoding device and method based on mixing distortion metrics criterion |
| WO2018113191A1 (en) * | 2016-12-21 | 2018-06-28 | 杭州海康威视数字技术股份有限公司 | Bit rate control method and device |
| CN108574841A (en) * | 2017-03-07 | 2018-09-25 | 北京金山云网络技术有限公司 | A kind of coding method and device based on adaptive quantizing parameter |
| CN110121071A (en) * | 2018-02-05 | 2019-08-13 | 广东欧珀移动通信有限公司 | Method for video coding and Related product |
| CN110895794A (en) * | 2018-09-13 | 2020-03-20 | 浙江宇视科技有限公司 | Video effect self-repairing method and video self-repairing device |
| CN112420059A (en) * | 2020-10-15 | 2021-02-26 | 杭州微帧信息科技有限公司 | Audio coding quantization control method combining code rate layering and quality layering |
| CN112492122A (en) * | 2020-11-17 | 2021-03-12 | 杭州微帧信息科技有限公司 | VMAF-based method for adaptively adjusting sharpening parameters |
| WO2021136110A1 (en) * | 2019-12-31 | 2021-07-08 | 华为技术有限公司 | Coding method and coder |
-
2007
- 2007-11-30 CN CN 200710171620 patent/CN101202912A/en active Pending
Cited By (24)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102118618A (en) * | 2011-03-30 | 2011-07-06 | 北京世纪鼎点软件有限公司 | Method for realizing H.264 multi-code-rate video stream based on PAFF |
| CN102118618B (en) * | 2011-03-30 | 2012-10-31 | 北京世纪鼎点软件有限公司 | Method for realizing H.264 multi-code-rate video stream based on PAFF |
| CN103167318A (en) * | 2011-12-13 | 2013-06-19 | 中国电信股份有限公司 | Streaming media processing method, device, server and system |
| WO2013086724A1 (en) * | 2011-12-15 | 2013-06-20 | Mediatek Singapore Pte. Ltd. | Method of clippling transformed coefficients before de-quantization |
| CN103327284A (en) * | 2012-12-18 | 2013-09-25 | 中兴通讯股份有限公司 | Method, device and video communication system for transmitting lossless video |
| CN103327284B (en) * | 2012-12-18 | 2018-11-06 | 南京中兴软件有限责任公司 | Transmission method, device and the video communications system of lossless video |
| CN106331704A (en) * | 2015-07-07 | 2017-01-11 | 杭州海康威视数字技术股份有限公司 | Video code rate control method and video coding device |
| WO2017004863A1 (en) * | 2015-07-07 | 2017-01-12 | 杭州海康威视数字技术股份有限公司 | Video code rate control method and video encoding apparatus |
| US10542254B2 (en) | 2015-07-07 | 2020-01-21 | Hangzhou Hikvision Digital Technology Co., Ltd | Video bit rate control method and video coding device |
| CN106331704B (en) * | 2015-07-07 | 2019-10-22 | 杭州海康威视数字技术股份有限公司 | A kind of video code rate control method and video coding apparatus |
| CN106713842B (en) * | 2015-11-16 | 2019-07-02 | 晶睿通讯股份有限公司 | Image processing method and camera thereof |
| CN106713842A (en) * | 2015-11-16 | 2017-05-24 | 晶睿通讯股份有限公司 | Image processing method and camera thereof |
| CN105744271A (en) * | 2016-02-17 | 2016-07-06 | 浙江大华技术股份有限公司 | Video coding method and device |
| US10735740B2 (en) | 2016-12-21 | 2020-08-04 | Hangzhou Hikvision Digital Technology Co., Ltd. | Bit rate control method and device |
| WO2018113191A1 (en) * | 2016-12-21 | 2018-06-28 | 杭州海康威视数字技术股份有限公司 | Bit rate control method and device |
| CN108574841A (en) * | 2017-03-07 | 2018-09-25 | 北京金山云网络技术有限公司 | A kind of coding method and device based on adaptive quantizing parameter |
| CN107205151B (en) * | 2017-06-26 | 2020-03-27 | 中国科学技术大学 | Coding and decoding device and method based on mixed distortion measurement criterion |
| CN107205151A (en) * | 2017-06-26 | 2017-09-26 | 中国科学技术大学 | Coding and decoding device and method based on mixing distortion metrics criterion |
| CN110121071A (en) * | 2018-02-05 | 2019-08-13 | 广东欧珀移动通信有限公司 | Method for video coding and Related product |
| CN110895794A (en) * | 2018-09-13 | 2020-03-20 | 浙江宇视科技有限公司 | Video effect self-repairing method and video self-repairing device |
| WO2021136110A1 (en) * | 2019-12-31 | 2021-07-08 | 华为技术有限公司 | Coding method and coder |
| CN112420059A (en) * | 2020-10-15 | 2021-02-26 | 杭州微帧信息科技有限公司 | Audio coding quantization control method combining code rate layering and quality layering |
| CN112420059B (en) * | 2020-10-15 | 2022-04-19 | 杭州微帧信息科技有限公司 | Audio coding quantization control method combining code rate layering and quality layering |
| CN112492122A (en) * | 2020-11-17 | 2021-03-12 | 杭州微帧信息科技有限公司 | VMAF-based method for adaptively adjusting sharpening parameters |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101202912A (en) | Method for controlling balanced code rate and picture quality code rate | |
| US10015497B2 (en) | Video quality of service management and constrained fidelity constant bit rate video encoding systems and methods | |
| AU2003296418B2 (en) | Rate control with picture-based lookahead window | |
| JP3756346B2 (en) | Method and system for processing multiple streams of video frames | |
| US5677969A (en) | Method, rate controller, and system for preventing overflow and underflow of a decoder buffer in a video compression system | |
| Lee et al. | Scalable rate control for MPEG-4 video | |
| US7983160B2 (en) | Method and apparatus for transmitting a coded video signal | |
| US7652993B2 (en) | Multi-stream pro-active rate adaptation for robust video transmission | |
| RU2349054C1 (en) | Contextually-adaptive adjustment of throughput capacity at speed control of video signal transmission | |
| CN100593953C (en) | Video transmission control method based on PID control and received frame rate stable model | |
| US20080151998A1 (en) | Method and Apparatus for Providing Rate Control for Panel-Based Real Time Video Encoder | |
| US20050002453A1 (en) | Network-aware adaptive video compression for variable bit rate transmission | |
| CN101743753A (en) | Utilize the rate controlled based on buffering of frame complexity, buffer level and frame interior position in the video coding | |
| CN105681793B (en) | Based on bit rate control method in the adaptive extremely low delay high-performance video coding frame of complexity of video content | |
| CN101690078A (en) | Bandwidth allocation control in multiple video streaming | |
| Wang et al. | Joint rate control for multi-program video coding | |
| CN105306969A (en) | Adaptive streaming media processing system and method | |
| WO2018005083A1 (en) | Method and system for dynamically adjusting a camera setting for a video stream | |
| WO2014182389A1 (en) | Methods and apparatuses including a statistical multiplexer with bitrate smoothing | |
| KR100498332B1 (en) | Apparatus and method for adaptive rate in video transcoder | |
| Changuel et al. | Joint encoder and buffer control for statistical multiplexing of multimedia contents | |
| US20150208069A1 (en) | Methods and apparatuses for content-adaptive quantization parameter modulation to improve video quality in lossy video coding | |
| Sun et al. | A rate control algorithm for wireless video transmission using perceptual tuning | |
| Yu et al. | New rate control algorithm for MPEG-4 video coding | |
| WO2012076904A1 (en) | Video data transmission system and method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Open date: 20080618 |