CN105187832A - Mobile video code rate control method based on 2.5G wireless network - Google Patents

Abstract

The invention discloses a mobile video code rate control method based on a 2.5G wireless network. The method comprises the steps that an image gray binaryzation method is used to carry out preprocessing to acquire the binaryzation image of each image in a video to be compressed; the complexity of the binaryzation image of the current frame is calculated; the current full scale of the buffer zone of a coded frame, the threshold of the default binaryzation image complexity and a default full scale threshold are acquired and stored; and if the threshold of the default binaryzation image complexity is less than the current frame complexity, or the threshold of the default binaryzation image complexity is greater than the current frame complexity, the current full scale is greater than or equal to the default full scale threshold and the number of compiled frames consecutively abandoned is greater than the default threshold of the number of frames, the actual code rate of the current frame is calculated based on a weighted value dynamically produced by the 2.5G network.

Description

Based on 2.5G wireless network mobile video bit rate control method

Technical field

The present invention relates to video image compression technology, be specifically related to based on 2.5G wireless network mobile video bit rate control method.

Background technology

The every field being applied in people's daily life and work that current wireless communications products is more and more general, in order to meet the demand of people in 2.5G wireless network Video Applications, current rate control algorithm mainly contains two kinds, the first is by TM5 version modified hydrothermal process in MPEG-2, be applied in JVT-F086, it adopts TM5 model, is divided into bit to distribute, rate-distortion optimization for the first time, calculates quantization parameter and again carries out rate-distortion optimization four step.The second is by VM8 version modified hydrothermal process in MPEG-4, and be applied in JVT-G012, proposition flow comes and goes model allocation bit number, and adopts secondary rate distortion function to calculate quantization parameter in macroblock layer.Although above two kinds of algorithms have good encoding efficiency under middle high code check environment, quality performance under the environment such as network bandwidth instability, low delay low bit-rate, video sequence strenuous exercise of transmission of video declines significantly.

Summary of the invention

For above-mentioned deficiency of the prior art, provided by the invention can the compression bit rate of dynamic conditioning frame based on 2.5G wireless network mobile video bit rate control method, still can ensure the quality of transmission of video when the network bandwidth is unstable.

In order to reach foregoing invention object, the technical solution used in the present invention is:

There is provided a kind of based on 2.5G wireless network mobile video bit rate control method, it comprises the following steps:

Adopt gradation of image binarization method, preliminary treatment obtains the binary image of each two field picture in video to be compressed;

Calculate the complexity of present frame binary image;

Obtain the current full scale of storage encoded frame place buffer area, preset binary image complexity threshold values and pre-set full scale threshold values;

When the binary image complexity threshold values preset is less than present frame complexity, or default binary image complexity threshold values is greater than present frame complexity, current full scale is more than or equal to pre-set full scale threshold values and abandon compiling frame number continuously and is greater than when pre-setting frame number threshold values, the weighted value dynamically produced based on 2.5G network calculates the actual bit rate of present frame:

$B_i=(\frac{a\×\mathrm{\ω}\×(N_{GOP}-n)}{N_{GOP}}+1)\×E_i,i=2,3...N_{GOP}$

Wherein, B is the actual bit rate of frame; A is weighted value; E _ifor the target bit rate of frame; ω is residue frame code check mean ratio; N is the sequence number of frame; N _gOPfor the length of GOP.

Beneficial effect of the present invention is: compared with the size between front frame complexity with present frame complexity and default binary image complexity threshold values by default binary image complexity threshold values, determine the present frame changed greatly relative to previous frame, and based on the weighted value that 2.5G network dynamically produces, its actual bit rate is calculated, thus make control coding and transmission being carried out to data volume, thus ensure that the quality of transmission of video.

Accompanying drawing explanation

Fig. 1 is the flow chart based on a 2.5G wireless network mobile video bit rate control method embodiment.

Fig. 2 is the flow chart based on another embodiment of 2.5G wireless network mobile video bit rate control method.

Embodiment

Below the specific embodiment of the present invention is described; so that those skilled in the art understand the present invention; but should be clear; the invention is not restricted to the scope of embodiment; to those skilled in the art; as long as various change to limit and in the spirit and scope of the present invention determined, these changes are apparent, and all innovation and creation utilizing the present invention to conceive are all at the row of protection in appended claim.

The flow chart based on a 2.5G wireless network mobile video bit rate control method embodiment is shown with reference to figure 1, Fig. 1.As shown in Figure 1, should comprise the following steps based on 2.5G wireless network mobile video bit rate control method:

Adopt gradation of image binarization method, preliminary treatment obtains the binary image of each two field picture in video to be compressed;

Because the encode bit number of generation and the motion intense degree of image and picture material complexity have the characteristic of direct linear relationship, frame binary image every in video flowing divides by this programme chronologically, is encoded to original image by the binarization of gray value Image Coding characteristic of order space.

Calculate the complexity of present frame binary image; Actual ask for complexity time, following formulae discovery can be adopted and obtain:

$C_i=\frac{(i-1)E_i}{{\mathrm{\Σ}}_{j=1}^{i-1}{B}_j},i=2,3...N_{GOP},j=1,2...i-1$

Wherein, B is the actual bit rate of frame; E _ifor the target bit rate of frame; E _i=ρ × B _i-1; ρ is linear predictor coefficient; N _gOPfor the length of GOP; I and j is the sequence number of frame.

Obtain the current full scale of storage encoded frame place buffer area, preset binary image complexity threshold values and pre-set full scale threshold values;

By relatively presetting binary image complexity threshold values and present frame complexity and the size between default binary image complexity threshold values and front frame complexity, determine the present frame changed greatly relative to previous frame:

When the binary image complexity threshold values preset is less than present frame complexity, or default binary image complexity threshold values is greater than present frame complexity, current full scale is more than or equal to pre-set full scale threshold values and abandon compiling frame number continuously and is greater than when pre-setting frame number threshold values, show that present frame changes greatly relative to previous frame, need to compile it.

When carrying out Rate Control, preferred binary image complexity threshold values is 80%, and pre-setting frame number threshold values is 2.

To the specific practice that the present frame changed greatly compiles be: the actual bit rate based on the weighted value calculating present frame that 2.5G network dynamically produces:

$B_i=(\frac{a\×\mathrm{\ω}\×(N_{GOP}-n)}{N_{GOP}}+1)\×E_i,i=2,3...N_{GOP}$

Wherein, B is the actual bit rate of frame; A is weighted value; E _ifor the target bit rate of frame; ω is residue frame code check mean ratio; N is the sequence number of frame; N _gOPfor the length of GOP.

Above-mentioned weighted value can adopt following formulae discovery and obtain:

$a=\left\{\begin{array}{cc}1& (\frac{26}{-RSSI-RSRP}\&GreaterEqual;0)\\ 1-\frac{26}{-RSSI-RSRP}& (\frac{26}{-RSSI-RSRP}<0)\end{array}\right.$

Wherein, RSSI is the intensity instruction that terminal receives 2.5G base station signal; RSRP is the mean value that terminal receives 2.5G base station signal power.

And residue frame code check mean ratio can adopt following formulae discovery and obtain:

$\mathrm{\&omega;}=\frac{(1-\frac{n}{N_{GOP}})}{5}$

Wherein, n is the sequence number of frame; N _gOPfor the length of GOP.

In one embodiment of the invention, because the actual bit rate of each frame all can carry out dynamic conditioning, if linear predictor coefficient chooses the target bit rate calculating that fixed value carries out frame, can affect to video transmission quality, so in order to the stability of the quality and transmission that ensure transmission of video, need when the target bit rate carrying out present frame is predicted to upgrade its linear predictor coefficient, also need to comprise the following steps in the complexity step calculating present frame binary image:

Calculate the linear predictor coefficient of present frame:

$\mathrm{\&rho;}=\frac{B_{i-1}}{B_{i-2}},i=2,3...N_{GOP}.$

The flow chart based on another embodiment of 2.5G wireless network mobile video bit rate control method is shown with reference to figure 2, Fig. 2.

When present frame is less relative to previous frame change, then abandon the compiling to present frame.Lose in order to avoid the code check because abandoning the frame compiled and occur that shake occurs picture quality or part picture runs off, with reference to figure 2, the target bit rate mean allocation abandoning compiling frame is left the target bit rate of frame by this programme.

Concrete judgment mode when present frame is less relative to previous frame change is: be less than present frame complexity when current full scale is less than to pre-set full scale threshold values and preset binary image complexity threshold values, or current full scale is more than or equal to pre-set full scale threshold values, establish binary image complexity threshold values to be less than present frame complexity and abandon compiling frame number continuously and is less than or equal to and pre-sets frame number threshold values.

Wherein, the practical operation method target bit rate of present frame being averagely allocated to the target bit rate of subsequent frame is adopt following formula to upgrade to the target bit rate of subsequent frame:

$E_i=\left\{\begin{array}{cc}{E}_i+\frac{R_x}{N_{GOP}-i}& (\frac{R_x}{N_{GOP}-i}\&le;0.23E_i)\\ 1.23E_i& (\frac{R_x}{N_{GOP}-i}>0.23E_i)\end{array}\right.$

Wherein, R abandons compiling the code check of frame, and x is its span of sequence number abandoning frame is 0≤x≤i.

After present frame is abandoned, in order to ensure that storing primary data cache district (uncoded frame region) has constant code stream to export, need to upgrade the full scale of encoded frame place buffer area, its concrete operations mode is:

${\mathrm{\&delta;}}_{i+1}=1-(G_i-\frac{V}{f})$

Wherein, δ is the full scale storing encoded frame place buffer area; G represents the quantity in remaining cache district; V is target bit rate; F is frame per second.

Refer again to Fig. 2, in order to improve the transmission quality of video further, in one embodiment of the invention, between the complexity calculating present frame binary image and the actual bit rate of described calculating present frame, comprise improving one's methods based on DCT coefficient further, judge whether occurrence scene converts present frame:

Obtain the DCT coefficient absolute value of present frame and itself and adjacent spaces frame DCT coefficient absolute value and ratio:

$F_i=\left\{\begin{array}{cc}\frac{D_i}{D_{i-2}}& (D_i>D_{i-2})\\ \frac{D_{i-2}}{D_i}& (D_{i-2}>D_i)\end{array}\right.$

Calculate scene handoff factor:

$T_i=\left\{\begin{array}{cc}\frac{F_i}{{\mathrm{vF}}_i}& (F_i>{\mathrm{vF}}_i)\\ \frac{{\mathrm{vF}}_i}{F_i}& ({\mathrm{vF}}_i>F_i)\end{array}\right.$

Wherein, D _irepresent present frame DCT coefficient absolute value and; VF _ifor the mean value of the ratio F of processed all frames; T is scene handoff factor;

Current scene handoff factor is greater than and switches threshold values in advance, enters the actual bit rate step calculating present frame.

In sum, this can the compression bit rate of dynamic conditioning frame based on 2.5G wireless network mobile video bit rate control method, still can ensure the quality of transmission of video when the network bandwidth is unstable.

Claims (9)

1., based on 2.5G wireless network mobile video bit rate control method, it is characterized in that, comprise the following steps:

Adopt gradation of image binarization method, preliminary treatment obtains the binary image of each two field picture in video to be compressed;

Calculate the complexity of present frame binary image;

Obtain the current full scale of storage encoded frame place buffer area, preset binary image complexity threshold values and pre-set full scale threshold values;

When the binary image complexity threshold values preset is less than present frame complexity, or default binary image complexity threshold values is greater than present frame complexity, current full scale is more than or equal to pre-set full scale threshold values and abandon compiling frame number continuously and is greater than when pre-setting frame number threshold values, the weighted value dynamically produced based on 2.5G network calculates the actual bit rate of present frame:

$B_i=(\frac{a\&times;\mathrm{\&omega;}\&times;(N_{GOP}-n)}{N_{GOP}}+1)\&times;E_i,i=2,3...N_{GOP}$

Wherein, B is the actual bit rate of frame; A is weighted value; E _ifor the target bit rate of frame; ω is residue frame code check mean ratio; N is the sequence number of frame; N _gOPfor the length of GOP.

2. according to claim 1ly it is characterized in that based on 2.5G wireless network mobile video bit rate control method, described complexity adopts following formulae discovery and obtains:

$C_i=\frac{(i-1)E_i}{{\mathrm{\&Sigma;}}_{j=1}^{i-1}{B}_j},i=2,3...N_{GOP},j=1,2...i-1$

Wherein, E _i=ρ × B _i-1; ρ is linear predictor coefficient.

3. according to claim 2ly it is characterized in that based on 2.5G wireless network mobile video bit rate control method, the complexity step of described calculating present frame binary image also comprises:

Calculate the linear predictor coefficient of present frame:

$\mathrm{\&rho;}=\frac{B_{i-1}}{B_{i-2}},i=2,3...N_{GOP}.$

4. arbitrary described based on 2.5G wireless network mobile video bit rate control method according to claim 1-3, it is characterized in that, described weighted value adopts following formulae discovery and obtains:

$a=\left\{\begin{array}{cc}1& (\frac{26}{-RSSI-RSRP}\&GreaterEqual;0)\\ 1-\frac{26}{-RSSI-RSRP}& (\frac{26}{-RSSI-RSRP}<0)\end{array}\right.$

Wherein, RSSI is the intensity instruction that terminal receives 2.5G base station signal; RSRP is the mean value that terminal receives 2.5G base station signal power.

5. according to claim 4ly it is characterized in that based on 2.5G wireless network mobile video bit rate control method, described residue frame code check mean ratio adopts following formulae discovery and obtains:

$\mathrm{\&omega;}=\frac{(1-\frac{n}{N_{GOP}})}{5}$

Wherein, n is the sequence number of frame; N _gOPfor the length of GOP.

6. according to claim 1ly to it is characterized in that based on 2.5G wireless network mobile video bit rate control method,

Present frame complexity is less than when current full scale is less than to pre-set full scale threshold values and preset binary image complexity threshold values, or current full scale is more than or equal to pre-set full scale threshold values, establish binary image complexity threshold values to be less than present frame complexity and abandon compiling frame number continuously and is less than or equal to when pre-setting frame number threshold values, the target bit rate of present frame is averagely allocated to the target bit rate of subsequent frame.

7. according to claim 6ly to it is characterized in that based on 2.5G wireless network mobile video bit rate control method, also comprise the full scale upgrading encoded frame place buffer area:

${\mathrm{\&delta;}}_{i+1}=1-(G_i-\frac{V}{f})$

Wherein, δ is the full scale storing encoded frame place buffer area; G represents the quantity in remaining cache district; V is target bit rate; F is frame per second.

8. according to claim 1,2,3,5,6 or 7 based on 2.5G wireless network mobile video bit rate control method, it is characterized in that, comprise improving one's methods based on DCT coefficient further between the complexity of described calculating present frame binary image and the actual bit rate of described calculating present frame, judge whether occurrence scene converts present frame:

Obtain the DCT coefficient absolute value of present frame and itself and adjacent spaces frame DCT coefficient absolute value and ratio:

$F_i=\left\{\begin{array}{cc}\frac{D_i}{D_{i-2}}& (D_i>D_{i-2})\\ \frac{D_{i-2}}{D_i}& (D_{i-2}>D_i)\end{array}\right.$

Calculate scene handoff factor:

$T_i=\left\{\begin{array}{cc}\frac{F_i}{{\mathrm{vF}}_i}& (F_i>{\mathrm{vF}}_i)\\ \frac{{\mathrm{vF}}_i}{F_i}& ({\mathrm{vF}}_i>F_i)\end{array}\right.$

Wherein, D _irepresent present frame DCT coefficient absolute value and; VF _ifor the mean value of the ratio F of processed all frames; T is scene handoff factor;

Current scene handoff factor is greater than and switches threshold values in advance, enters the actual bit rate step calculating present frame.

9. according to claim 8ly it is characterized in that based on 2.5G wireless network mobile video bit rate control method, described binary image complexity threshold values is 80%, and pre-setting frame number threshold values is 2.