CN109076212A - The method and code device of code rate control - Google Patents

The method and code device of code rate control Download PDF

Info

Publication number
CN109076212A
CN109076212A CN201780018104.7A CN201780018104A CN109076212A CN 109076212 A CN109076212 A CN 109076212A CN 201780018104 A CN201780018104 A CN 201780018104A CN 109076212 A CN109076212 A CN 109076212A
Authority
CN
China
Prior art keywords
image
encoded
frame
parameter
complexity
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
Application number
CN201780018104.7A
Other languages
Chinese (zh)
Inventor
缪泽翔
郑萧桢
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.)
SZ DJI Technology Co Ltd
Shenzhen Dajiang Innovations Technology Co Ltd
Original Assignee
Shenzhen Dajiang Innovations Technology Co Ltd
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 Shenzhen Dajiang Innovations Technology Co Ltd filed Critical Shenzhen Dajiang Innovations Technology Co Ltd
Publication of CN109076212A publication Critical patent/CN109076212A/en
Pending legal-status Critical Current

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
    • H04N19/103—Selection of coding mode or of prediction mode
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
    • H04N19/124—Quantisation
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/134—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or criterion affecting or controlling the adaptive coding
    • H04N19/146—Data rate or code amount at the encoder output
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/134—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or criterion affecting or controlling the adaptive coding
    • H04N19/157—Assigned coding mode, i.e. the coding mode being predefined or preselected to be further used for selection of another element or parameter
    • H04N19/159—Prediction type, e.g. intra-frame, inter-frame or bidirectional frame prediction
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
    • H04N19/17—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object
    • H04N19/176—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object the region being a block, e.g. a macroblock
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/50—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/60—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding
    • H04N19/61—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding in combination with predictive coding
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/90—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using coding techniques not provided for in groups H04N19/10-H04N19/85, e.g. fractals
    • H04N19/91—Entropy coding, e.g. variable length coding [VLC] or arithmetic coding

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Compression Or Coding Systems Of Tv Signals (AREA)

Abstract

A kind of image processing method and device are provided, this method comprises: receiving the image complexity of image to be encoded, which obtained by the image processor being connected with encoder；Determine the pre-arranged code bit of the image to be encoded；According to the image complexity and the pre-arranged code bit, the quantization parameter of the image to be encoded is obtained；According to the quantization parameter of the image to be encoded, which is encoded, the computation complexity of code rate control accuracy and encoder can be taken into account.

Description

The method and code device of code rate control
Technical field
This application involves coding and decoding video fields, and more specifically, are related to the method and coding of a kind of code rate control Device.
Background technique
Visual information is one of the most important source that the mankind obtain external information, but camera acquisition gained is uncompressed The memory space that occupies of original video data it is very huge, be 1920 × 1080, frame per second with resolution ratio be that 30 frames are per second, duration For 10 minutes videos, the memory space that the video needs to occupy 104.28G byte is stored in the case where no compression.In order to It reduces video and stores and transmits occupied bandwidth, need to carry out encoding compression processing to video data.
Video coding process includes prediction, transformation, code rate control, quantization and entropy coding.Wherein, code rate control exists Play very important role in video coding process.Since the content and feature of different video sequence are all different, if The bit rate output and different degrees of video distortion that coding will create a difference very big are carried out using same coding parameter, in order to Overcome this problem, just proposes code rate control.
Rate control algorithm is to be adaptively adjusted coding parameter according to the feature of different video image.Generally, based on mark Quasi- video encoder has very big flexibility in the selecting party face of coding parameter, after the selection of coding parameter may be to coding The code rate of video code flow has a great impact.Rate control algorithm can make, and encoder can be in a series of coding parameter collection Suitable coding parameter is selected between conjunction, to realize that actual coding bit approaches pre-assigned coded-bit.
In the industry cycle in existing rate control algorithm, some algorithms will lead to code rate control inaccuracy, some algorithms are solving It certainly will increase the computation complexity of encoder while code rate control inaccuracy again.
It is, therefore, desirable to provide a kind of coding staff for the computation complexity that can be compatible with code rate control accuracy and encoder Method.
Summary of the invention
The application provides a kind of bit rate control method and device, and code rate precise control can be taken into account in rate control process The computation complexity of degree and encoder.
In a first aspect, providing a kind of method of code rate control, the method is applied to video encoder, the method packet It includes: receiving the image complexity of image to be encoded, described image complexity is obtained by the image processor being connected with encoder 's；Determine the pre-arranged code bit of the image to be encoded；According to described image complexity and the pre-arranged code bit, obtain The quantization parameter of the image to be encoded；According to the quantization parameter of the image to be encoded, the image to be encoded is compiled Code.
Second aspect provides a kind of code device, and the code device includes: Transmit-Receive Unit, for receiving figure to be encoded The image complexity of picture, described image complexity are obtained by the image processor being connected with encoder；Processing unit is used for Determine the pre-arranged code bit of the image to be encoded；According to described image complexity and the pre-arranged code bit, institute is obtained State the quantization parameter of image to be encoded；According to the quantization parameter of the image to be encoded, the image to be encoded is encoded.
The third aspect provides a kind of code device, and the code device includes memory and processor, and the memory is used In store instruction, the processor is used to execute the instruction of the memory storage, and to the finger stored in the memory The execution of order is so that the processor executes the method that first aspect provides.
Fourth aspect provides a kind of chip, and the chip includes processing module and communication interface, and the processing module is used for It controls the communication interface to be communicated with outside, the processing module is also used to realize the method that first aspect provides.
5th aspect, provides a kind of computer readable storage medium, is stored thereon with computer program, the computer journey Realize the computer in any possible implementation of first aspect or first aspect Method.Specifically, the computer can be above-mentioned * * equipment.
6th aspect, provides a kind of computer program product comprising instruction, makes when described instruction is computer-executed The computer realizes the method that first aspect provides.
In conclusion the embodiment of the present invention when carrying out code rate control, is obtained by the image processor outside encoder Image complexity, a large amount of calculate that can will likely occur are completed by the image processor (such as ISP) except encoder, thus The calculation amount and implementation complexity of encoder can be reduced, and then secondary coding bring coding can be carried out to avoid encoder and prolonged Late.Meanwhile the embodiment of the present invention is by obtaining quantization parameter according to image complexity and pre-arranged code bit, so as to one Determine to guarantee rate control accuracy in degree.Therefore, the scheme of the embodiment of the present invention can be before not sacrificing rate control accuracy It puts, the calculation amount and implementation complexity of encoder can be reduced, so as to improve code efficiency.
Detailed description of the invention
Fig. 1 is the schematic flow chart of video coding process.
Fig. 2 is the schematic block diagram of the image processing system of the embodiment of the present invention.
Fig. 3 is the schematic flow chart of the method for the code rate control of the embodiment of the present invention.
Fig. 4 is the schematic diagram that full frame image is divided into several image blocks in the embodiment of the present invention.
Fig. 5 is the schematic diagram of the block of pixels in the embodiment of the present invention.
Fig. 6 is another schematic flow chart of the method for the code rate control of the embodiment of the present invention.
Fig. 7 is the schematic block diagram of the code device of the embodiment of the present invention.
Fig. 8 is another schematic block diagram of the code device of the embodiment of the present invention.
Specific embodiment
In order to facilitate understanding with the scheme of description the application, Fig. 1 is combined to describe the process of Video coding first below.
As shown in Figure 1, video coding process includes prediction, transformation, quantization and entropy coding.The explanation of these steps It is as follows.
Prediction it is intended that removing the redundancy of current image block to be encoded using prediction block message.Prediction packet Include intra prediction and inter-prediction two types.Intra prediction is the information acquisition prediction block data using this frame image.For example, Intra prediction is using the spatial information (si) of current encoded frame for eliminating redundancy.Inter-prediction utilizes the information acquisition of reference frame Prediction block data.For example, the time domain frame information adjacent with before and after current encoded frame can be used for eliminating redundancy in inter-prediction Information.The process of inter-prediction includes: firstly, image block to be encoded is divided into several subimage blocks；Then, for each Subimage block, search is with the most matched image block of current subimage block as prediction block in reference frame；Thereafter, by the subgraph The respective pixel values of block and the prediction block subtract each other to obtain residual error, and by the corresponding residual error combination of obtained each subimage block one It rises, obtains the residual error of image block to be encoded.
In Video coding processing, it is referred to as I frame using only the video frame of intra prediction mode；It can be used preceding to reference The video frame that frame carries out inter-prediction is known as P frame；The video frame that bi-directionally referenced frame can be possessed is known as B frame.
The purpose of transformation is to remove the redundancy of image block.Specifically, transformation matrix removal image block can be used Residual error correlation, that is, remove image block redundancy, to improve code efficiency.The transformation of data block is logical in image block Frequently with two-dimensional transform, i.e., the residual information of data block is multiplied with a transformation matrix and its transposed matrix respectively, is multiplied it What is obtained afterwards is transformation coefficient.
The purpose of quantization is, obtains quantization parameter according to variation coefficient.Specifically, according to quantization parameter to transformation coefficient Quantified, obtains corresponding quantization parameter.
The purpose of entropy coding is, by carrying out entropy coding to quantization parameter, obtains bit stream.
As shown in Figure 1, in an encoding process, inverse quantization and inverse transformation will also include.What inverse quantization referred to be exactly with The opposite process of quantizing process.What inverse transformation referred to is exactly the process opposite with conversion process.
Coding side after completing image coding, can the coding mode by bit stream that entropy coding obtains and after being encoded believe Breath, such as intra prediction mode, motion vector information, are stored or are sent to decoding end.
In decoding end, after obtaining entropy coding bit stream, firstly, carrying out entropy decoding to the bit stream, obtain corresponding residual Difference；Then, according to the coding mode informations such as obtained motion vector or intra prediction are decoded, forecast image block is obtained；Finally, root It is predicted that the residual error of image block and image block obtains the value of each pixel in current subimage block.
In video coding process, in addition to above-mentioned prediction, transformation, quantization and entropy coding and etc. other than, code rate control Play very important role.
Its content of different video sequences and feature are all different.If using same coding parameter to different videos Sequence is encoded, the bit rate output and different degrees of video distortion that can create a difference very big.In order to overcome this problem, It needs to be adaptively adjusted coding parameter according to the feature of different video image, the process of this adjustment coding parameter is known as code Rate control.
The purpose of code rate control is, by dynamically adjusting coding parameter, to make under the premise of guaranteeing certain picture quality Compressed video code flow can be by the channel real-time Transmission of fixed-bandwidth, or meets ASCII stream file ASCII after the interior compression of specific time It must be smaller than the constraint of some predetermined size.
It is realized in general, rate control algorithm can be divided into following two step:
The first step allocates coded-bit in advance before each encoded hierarchy coding.
Encoded hierarchy is from high in the end successively are as follows: picture group (group of pictures, GOP) grade, single video frame level and base This subimage block grade.
In this step, can also consider encoder buffer takes degree.
Second step, in the cataloged procedure of each encoded hierarchy, by adjusting coding parameter, so that actual coding bit It matches with the pre-assigned coded-bit of the first step.
The coding parameter dynamically adjusted in rate control process includes: the direction mode of intra prediction, the fortune of inter-prediction Dynamic vector, the coding parameters such as quantization parameter in quantizing process.
In quantizing process, the scaling of transformation coefficient, Input transformation coefficient, output quantization system are realized by quantization parameter Number, shown in the following formula of the process:
Wherein, quantizedCoeff indicates quantization parameter；TransformedCoeff indicates transformation coefficient；QP expression amount Change parameter；F (QP) indicates quantization step, is the scalar determined by quantization parameter QP.
Binary code stream can be further encoded into entropy coder since quantization parameter enters, and the ratio of coded quantization coefficient Spy's amount accounting in the code stream that entire video is encoded into is very high, and the amplitude range of quantization parameter is controlled by quantization step, amount Change step-length directly determined by quantization parameter, therefore the size of quantization parameter by determine coding video frames after bit rate output size. In the case where other coding parameters are fixed, when larger quantization parameter normally results in encoder coding same video output compared with Small code rate, on the contrary biggish code rate is exported when lesser quantization parameter will lead to encoder coding same video.
Therefore, in rate control process, the coding parameter that frequently uses first is that quantization parameter.
In the industry cycle in existing rate control algorithm, in order to improve the accuracy of code rate control, start to encode in encoder Before, the characteristic information that some images are obtained by pretreatment is controlled for code rate.But before encoding, encoder executes pre- The process of processing can bring coding delay, to influence the speed of service of encoder, be unfavorable for real-time coding realization.
In view of the above-mentioned problems, the application proposes the method and code device of a kind of code rate control, code rate control can be taken into account The computation complexity of accuracy and encoder.
In this application, firstly, image complexity is introduced rate control algorithm.Secondly, by the image outside encoder Processor, such as image-signal processor (image signal processor, ISP) generate the image complexity of image to be encoded Spend information.
Due to executing the calculating of image complexity without encoder, it can thus be avoided introducing additional meter for encoder Calculate complexity and coding time.Simultaneously as vision signal is after camera acquires, then is transferred to before video encoder, The processing of image processor (such as ISP) can be first passed around, therefore, image complexity information is generated by image processor, it will not Additional component is introduced in image processing system.
Fig. 2 is the configuration diagram of the image processing system of the embodiment of the present invention.The system includes 210 He of image processor Code device 220.Code device 220 includes prediction module 221, conversion module 222, rate control module 223, quantization modules 224 with entropy code module 225.
Image processor 210 is used for, and receives image sequence to be encoded, calculates its image complexity, and image is complicated Degree is input in rate control module 223.
Specifically, image processor 210 is ISP.
Prediction module 221 is used for, and is received image sequence to be encoded, is predicted it, and corresponding residual signals are obtained.
Variation module 222 is used for, and is received the residual signals that prediction module 221 exports, is converted to it, obtains corresponding Transformation parameter.
Rate control module is used for, and 223 is complicated according to pre-arranged code bit and the image from the acquisition of image processor 210 Degree, obtains corresponding quantization parameter.
Quantization modules 224 are used for, the amount that the transformation parameter and rate control module 223 for receiving variation module 222 export Change parameter, and running parameter is quantified using quantization parameter, obtains corresponding quantization parameter.
Entropy code module 225 is used for, and receives the quantization parameter that quantization modules 224 export, and carry out entropy volume to quantization parameter Code obtains coded bit stream.
As shown in Fig. 2, further including inverse transform block and inverse quantization module in system.The function and transformation mould of inverse transform block The function of block is opposite.The function of inverse quantization module and the function of quantization modules are opposite.
In embodiments of the present invention, image complexity is calculated by the image processor outside encoder, is encoded in image In the process, code rate control is carried out using the image complexity.On the one hand, guarantee rate control accuracy to a certain extent；It is another Aspect, it would be possible to a large amount of calculate of appearance is completed by the image processor outside encoder, can reduce encoder calculating and Implementation complexity.
Fig. 3 is the schematic flow chart of the method for the code rate control of the embodiment of the present invention.This method can be by Video coding Device executes, such as is executed by code device 220 shown in Fig. 2.As shown in figure 3, this method comprises the following steps.
310, the image complexity of image to be encoded is received, which is the image procossing by being connected with encoder What device obtained.
Specifically, image processor obtains the image complexity of image to be encoded, and then, encoder connects from image processor Receive the image complexity of image to be encoded.The image processor is located at the outside of encoder.For example, the image processor is ISP.
Specifically, the image complexity of the image to be encoded is obtained according to the texture information of the image to be encoded.It answers Understand, the image complexity that the texture information by calculating image obtains can characterize the information content of the image.Alternatively, should be wait compile The image complexity of code image is to determine that is, this is to be encoded according to the fluctuation severe degree of the image inner region pixel to be encoded The image complexity of image indicates the fluctuation severe degree of the image inner region pixel to be encoded.
320, determine the pre-arranged code bit of the image to be encoded.
Specifically, pre-arranged code ratio refers in particular to, before image coding, pre-assigned coded-bit.
330, according to the image complexity and the pre-arranged code bit, obtain the quantization parameter of the image to be encoded.
It should be understood that image complexity can characterize the information content of image before encoding, and pre-arranged code bit indicates encoded The information content of compressed images, therefore it is to close that the method by combining image complexity and pre-arranged code bit, which calculates quantization parameter, Reason is effective.
340, according to the quantization parameter of the image to be encoded, which is encoded.
It should be understood that above-mentioned steps 310 to step 330 belongs to rate control process.Specifically, step 310 to step 330 can To be executed by the rate control module of decoder internal, step 340 can be by the coding module of decoder internal (for example, realizing The module of entropy coding) it executes.
It is complicated that the embodiment of the present invention obtains image when carrying out code rate control, through the image processor outside encoder Degree, a large amount of calculate that can will likely occur are completed by the image processor (such as ISP) except encoder, can reduce coding The calculation amount and implementation complexity of device, so as to avoid the coding delay of encoder from influencing the speed of service of encoder, in turn It is advantageously implemented real-time coding.Meanwhile the embodiment of the present invention is quantified by being obtained according to image complexity and pre-arranged code bit Parameter, so as to guarantee rate control accuracy to a certain extent.Therefore, the scheme of the embodiment of the present invention can not sacrifice Under the premise of rate control accuracy, the calculation amount and implementation complexity of encoder can be reduced, so as to improve code efficiency.
Optionally, in some embodiments, image to be encoded is an image block in picture frame.
Specifically, encoder is when encoding a frame image, first by full frame image be divided into several 64 × 64 Image block indicates that a size is 64 × 64 for each piece in Fig. 4 as shown in figure 4, a frame image is divided into 128 image blocks Image block.Then, encoder in order successively encodes each image block.
Optionally, in some embodiments, image to be encoded can also directly be a frame image.
For ease of description with understanding, hereafter retouched so that image to be encoded is an image block in picture frame as an example It states, but the embodiment of the present invention is not construed as limiting this.
Optionally, in some embodiments, the picture frame where the image to be encoded is I frame.
Specifically, in Video coding processing, it is referred to as I frame using only the video frame of intra prediction mode.
The residual signals of I frame image are generated by I picture signal estimation when due to coding, pass through getting frame The complexity information of interior image is beneficial for rate control algorithm.
Optionally, in some embodiments, the picture frame where the image to be encoded is P frame.
Specifically, in Video coding processing, it is known as P frame using the video frame that forward reference frame carries out inter-prediction.
Optionally, in some embodiments, step 330 specifically includes: the quantization parameter is calculated using first function model, It does not include exponent arithmetic in the first function model, the independent variable of the first function model includes the image complexity and this is default Coded-bit.
Optionally, in some embodiments, the first function model only include logarithm operation, multiplying, addition and/or Subtraction, the independent variable of the first function model include the image complexity and the pre-arranged code bit.
Specifically, logarithm operation as mentioned in the embodiments of the present invention can be various forms of logarithm operations, for example, with e For the logarithm operation at bottom, denary logarithm operation, alternatively, being using other numerical value as the logarithm operation of the truth of a matter, the present invention is real It applies example and this is not construed as limiting.
Optionally, in some embodiments, the coefficient in the first function model includes at least one following parameter: work multiplies First parameter of method operation, the second parameter for making multiplying and the third parameter for making addition and/or subtraction, this first Parameter, second parameter and the third parameter are the parameter for code rate control.
Specifically, first parameter, rate control module that second parameter and the third parameter are decoder internal from Some code rate control parameters.
Specifically, the quantization parameter of the coded image is calculated using first function model, comprising: calculate according to the following formula The quantization parameter of the coded image:
QP=A1log (complexity)+A2log (targetBits)+B (2)
Wherein, QP indicates that the quantization parameter, complexity indicate described image complexity, and targetBits indicates institute Pre-arranged code bit is stated, A1 indicates first parameter, and A2 indicates second parameter, and B indicates the third parameter, log () Indicate logarithm operation.
Optionally, in some embodiments, the absolute value of the first parameter A1 and the second parameter A2 are identical.
Optionally, in some embodiments, the image complexity of image to be encoded is the texture according to the image to be encoded What information obtained.In other words, the image processor outside encoder (such as ISP) calculates image according to the texture information of image Complexity.
Specifically, it is assumed that the image to be encoded includes M block of pixels, and the size of the block of pixels is that N × N, M and N are positive The image complexity of integer, the image to be encoded isSADjFor the pixel of j-th of block of pixels in M block of pixels The sum of poor absolute value, andAvejFor the pixel average of j-th of block of pixels, and Indicate the pixel in j-th of block of pixels.
As an example, a width complete image (being denoted as picture frame) is divided into the image that several sizes are 64 × 64 Block.By taking resolution ratio is 1920 × 1080 picture frame as an example, 510 (30 × 17) a image blocks can be divided into.With size for 2 × 2 Block of pixels as calculate image complexity minimum unit, be made of as shown in Figure 5 four pixels x0, x1, x2, x3 Block of pixels.It should be understood that including 1024 block of pixels in an image block.
Firstly, calculating the pixel average average of a block of pixels according to the following formula2×2:
average2×2=(x0+x1+x2+x3)/4 (3)
Secondly, calculating the sum of the pixel difference absolute value of block of pixels SAD according to the following formula2×2:
Wherein, xi indicates the pixel value of ith pixel in the block of pixels.
Then, the image complexity complexity of an image block is calculated according to the following formula64×64:
Wherein, SAD2x2iIndicate the sum of the pixel difference absolute value of ith pixel block in the image block.
In other words, by by the sum of the pixel difference absolute value of all pixels block in same image block SAD2×2It is added, obtains Obtain the image complexity of the image block.
The image complexity of picture frame can be added by the image complexity of all image blocks in the picture frame and be obtained.
The image complexity of image block can be described as block grade image complexity.The image complexity of picture frame can be described as frame level figure As complexity.
It should be understood that the above-mentioned description in conjunction with Fig. 5 is merely illustrative and non-limiting.It, can be as the case may be in practical application Determine the minimum unit for calculating image complexity.
Optionally, in some embodiments, the image complexity of image to be encoded indicates image inner region pixel to be encoded Fluctuation severe degree.
As a kind of optional implementation, the image complexity of image to be encoded is obtained by frequency-domain calculations method.
For example, discrete (just) cosine transform or the transformation of Hadamard transform of the image processor by statistics image to be encoded Coefficient determines the image complexity of image to be encoded.
As another optional implementation, the image complexity of image to be encoded is obtained by pixel domain calculation method ?.
For example, image processor determines image to be encoded by the variance or standard deviation of pixel in calculating image to be encoded Image complexity.
Optionally, in some embodiments, image to be encoded is an image block in picture frame, and step 320 is specifically wrapped Include: according to the image complexity of frame level pre-arranged code bit, frame level image complexity and the image to be encoded, determining should be wait compile The pre-arranged code bit of code image, wherein frame level pre-arranged code bit indicates the pre-arranged code bit of the picture frame, frame level image Complexity refers to the sum of the image complexity of all image blocks in the picture frame.
Specifically, it is assumed that the image to be encoded is i-th of image block in picture frame, and calculating according to the following formula should be wait compile The pre-arranged code bit targetBits of code imagei:
Wherein, targetBits indicates frame level pre-arranged code bit, complexityiIndicate i-th of figure in picture frame As the image complexity of block (i.e. the image to be encoded), ∑jcomplexityjIndicate frame level image complexity, i.e., institute in picture frame There is the sum of the image complexity of image block.
Specifically, the process for calculating image complexity refers to description in conjunction with Fig. 5, and which is not described herein again.
Specifically, the picture frame in the present embodiment is I frame.
It is alternatively possible to before coding begins, distribute pre-arranged code bit in advance for each image block in picture frame.
Optionally, in some embodiments, image to be encoded is an image block in picture frame, which is P frame, Step 320 specifically includes: determining the default of the image to be encoded according to the product of the frame level predetermined bit of the P frame and the first coefficient Bit, wherein first coefficient is determined according at least one of following: institute in the overview image complexity of the P frame, the P frame There is the frame level of the sum of weight of image block, the image complexity of the image to be encoded, the weight of the image to be encoded and the P frame Predetermined bit, wherein the weight is determined by the code rate control parameter of the P frame.
Specifically, which is the overview image complexity that the P frame is accounted for according to the image complexity of the image to be encoded The product of the ratio of degree and the second coefficient accounts for the sum of weight of all image blocks in the P frame with the weight of the image to be encoded What ratio and the sum of products of third coefficient determined, wherein second coefficient and the third coefficient are all larger than the 0, and second coefficient With third coefficient and be equal to 1.
As an example, it is assumed that the image to be encoded is i-th of image block in picture frame, is calculated according to the following formula The pre-arranged code bit targetBits of the image to be encodedi:
Wherein, targetBits indicates frame level pre-arranged code bit；complexityiIndicate i-th in the picture frame The image complexity of image block (i.e. the image to be encoded)；∑jcomplexityjIndicate frame level image complexity, the i.e. picture frame The sum of the image complexity of interior all image blocks；weightiIndicate i-th image block (i.e. the image to be encoded) in picture frame Weight, ∑jweightjIndicate the sum of the weight of all image blocks in the picture frame；C1 indicates that the second coefficient, C2 indicate third system Number.For example, the second coefficient C1 is 1/4, third coefficient C2 is 3/4.In this example, the weight of image block is the code rate according to P frame The weight that control parameter determines.
Specifically, the process for calculating image complexity refers to description in conjunction with Fig. 5, and which is not described herein again.
It is alternatively possible to before coding begins, distribute pre-arranged code bit in advance for each image block in picture frame.
From formula (6) and formula (7) it is found that in scheme provided by the present application, for I frame and P frame, pre-arranged code is determined The method of bit is different.In other words, the Rate Control Schem used when I frame encodes and the code rate used when P frame encodes Control strategy is different, is conducive to the accuracy for improving code rate control in this way, and reason is as follows.
Prediction residual is generated since I frame does not depend on the reference information of time domain, the residual error letter obtained when encoding I frame What breath indicated is certain coded sub-blocks and the difference between its space adjacent image.And the residual information obtained when encoding P frame is more Ground indicates current encoded frame and the difference of reference frame in the time domain.In other words, it is obtained in I frame coding and P frame coding residual Poor information is different, accordingly, also should be corresponding different to the Rate Control Schem of P frame for I frame in code rate control.In industry In the existing some rate control algorithms in boundary, Rate Control Schem identical with P frame is used in I frame coding, it is this to do Method will lead to code rate control inaccuracy, to influence the quality of encoded video.
Therefore, the application is encoded by using different Rate Control Schem processing I frame codings with P frame, so as to mention High code rate precise control.
Optionally, in certain embodiments, one in picture frame of the image to be encoded where the image to be encoded Image block, this method further include: according to the frame level image complexity of the picture frame and frame level pre-arranged code bit, obtain the image The quantization parameter of frame.Step 340 specifically includes: according to the quantization parameter of the quantization parameter of the image to be encoded and the picture frame Size relation determines the quantization parameter for being encoded to the image to be encoded, then utilizes finally determining quantization parameter The image to be encoded is encoded.
Optionally, in certain embodiments, model of the quantization parameter of the image to be encoded in the quantization parameter of the picture frame In enclosing, step 340 is specifically included: according to the quantization parameter of the image to be encoded, encoding to the image to be encoded.
Optionally, in certain embodiments, the quantization parameter of the image to be encoded is not in the quantization parameter of the picture frame In range, step 340 is specifically included: the quantization parameter of the image to be encoded is cut out to the range of the quantization parameter of the picture frame Interior, the quantization parameter of the image to be encoded after being cut out according to this encodes the image to be encoded.
Optionally, in some embodiments, which indicates the image block in full frame image.In other words, with Image block is the coding that minimum code unit carries out full frame image.
Specifically, full frame image is first divided into several image blocks, then in order successively to each image block according to Step 310 to step 340 is encoded, and when completing the coding of all image blocks in picture frame, completes the coding of full frame image.
Optionally, in some embodiments, after the coding for completing the picture frame where the image to be encoded, this method is also Include: the frame level actual coding bit and frame level pre-arranged code bit according to the picture frame, update the third parameter, the frame level is real Border coded-bit indicates the actual coding bit of the picture frame, which indicates that the frame level of the picture frame is default Coded-bit.
Specifically, according to second function model, updated third parameter, the independent variable packet of the second function model are obtained Include the third parameter before the frame level actual coding bit, the frame level pre-arranged code bit and update, the second function model Comprise at least one of the following operation: logarithm operation, multiplying, addition and/or subtraction.
As an example, the third parameter B of update is obtained according to the following formulanew:
Bnew=Bold+k·A1·(log(actualBits)-log(targetBits)) (8)
Wherein, the third parameter before indicating to update, i.e., upper one is completed third parameter used by the picture frame of coding； Indicate the first parameter；K indicates renewal rate, such as can be taken as empirical value 2；ActualBits indicates frame level actual coding ratio It is special；TargetBits indicates frame level pre-arranged code bit.
It should be understood that code rate control parameter A1 (i.e. the first parameter), code rate control parameter B (i.e. third parameter) are in encoder The own parameter of portion's rate control module, the characteristics of to make rate control module more adapt to different video image, in cataloged procedure In, after every coding for completing a frame image, code rate control parameter B can be according to frame level actual coding bit and frame level pre-arranged code ratio Special gap carries out adaptive iteration update.
The present embodiment, by the way that after the coding for completing a frame image, iteration updates code rate control parameter, so that The accuracy of code rate control can be improved in the characteristics of rate control module in encoder more adapts to different video image in this way.
It should be understood that the separate equations described herein are merely illustrative and non-limiting.For example, the parameter or coefficient in formula can To determine according to actual needs.
Fig. 6 is another schematic flow chart of the image encoding method of the embodiment of the present invention, and this method is executed by encoder. In the description of the present embodiment, by the image complexity of image block, pre-arranged code bit, actual coding bit, quantization parameter point It is not denoted as: block grade image complexity, block grade pre-arranged code bit, block grade actual coding bit, block level quantized parameter.By whole frame figure Image complexity, pre-arranged code bit, actual coding bit, the quantization parameter of picture are denoted as respectively: frame level image complexity, frame level Pre-arranged code bit, frame level actual coding bit, frame level quantization parameter.As shown in fig. 6, this method comprises the following steps.
6001, block grade image complexity is obtained from image processor, which is located at the outside of encoder.
6002, determine block grade pre-arranged code bit.
6003, the block grade pre-arranged code ratio that the block grade image complexity and step 6002 determined according to step 6001 determines Spy, and code rate control parameter (the first parameter, the second parameter and third parameter in such as foregoing embodiments) is combined, calculation block grade amount Change parameter.
6004, frame level image complexity is obtained from image processor.
6005, determine frame level pre-arranged code bit.
6006, the frame level pre-arranged code ratio that the frame level image complexity and step 6005 determined according to step 6004 determines Spy, and code rate control parameter (the first parameter, the second parameter and third parameter in such as foregoing embodiments) is combined, calculate frame level amount Change parameter.
6007, the frame level quantization parameter whether the block level quantized parameter that judgment step 6003 obtains obtains in step 6006 In range, if so, step 6008 is gone to, if it is not, going to step 6009.
6008, the block level quantized parameter that step 6003 obtains is cut out the one of the frame level quantization parameter obtained to step 6006 Determine in range, goes to step 6009.
6009, the coding of image block will be carried out in finally obtained piece of level quantized parameter input quantization modules.
Specifically, if the judging result of step 6007 is block level quantized parameter input quantity yes, that step 6003 is obtained Change in module；If the judging result of step 6007 be it is no, will be defeated by the block level quantized parameter after cutting out in step 6008 Enter in quantization modules.
6010, judge whether the coding for completing to work as former frame complete image, if so, step 6011 is gone to, if it is not, going to step Rapid 6012.
6011, according to the frame level actual coding bit and frame level pre-arranged code bit of the current picture frame for completing coding, more The code rate control parameter used in new step 6006.Go to step 6006.
Specifically, the third parameter as described in foregoing embodiments is updated, associated description as detailed above is no longer superfluous here It states.
6012, according to the block grade actual coding bit and block grade pre-arranged code bit of the current image block for completing coding, be Next image block to be encoded distributes block grade pre-arranged code bit.Go to step 6002.
Specifically, it after the completion of an image block coding, can be compiled according to the actual coding bit of the image block with default The difference of code bit distributes pre-arranged code bit for next image block to be encoded.
It should be understood that can guarantee the same of rate control accuracy by setting frame level quantization parameter and block level quantized parameter When make stable image quality in a frame.
It should also be understood that step 6001, step 6002, step 6004 in the above process, between step 6005 without stringent Ordinal relation.
In conclusion the embodiment of the present invention when carrying out code rate control, is obtained by the image processor outside encoder Image complexity, a large amount of calculate that can will likely occur are completed by the image processor (such as ISP) except encoder, thus The calculation amount and implementation complexity of encoder can be reduced, and then secondary coding bring coding can be carried out to avoid encoder and prolonged Late.Meanwhile the embodiment of the present invention is by obtaining quantization parameter according to image complexity and pre-arranged code bit, so as to one Determine to guarantee rate control accuracy in degree.Therefore, the scheme of the embodiment of the present invention can be before not sacrificing rate control accuracy It puts, the calculation amount and implementation complexity of encoder can be reduced, so as to improve code efficiency.
Described above is the present processes embodiments, and the Installation practice of the application is described below.It should be understood that device The description of embodiment is corresponded to each other with the description of embodiment of the method, and therefore, the content being not described in detail may refer to previous methods Embodiment, for sake of simplicity, which is not described herein again.
Fig. 7 is the schematic block diagram of code device 700 provided in an embodiment of the present invention, which includes:
Transmit-Receive Unit 710, for receiving the image complexity of image to be encoded, the image complexity be by with encoder phase What image processor even obtained；
Processing unit 720, for determining the pre-arranged code bit of the image to be encoded；It is pre- with this according to the image complexity If coded-bit, the quantization parameter of the image to be encoded is obtained；According to the quantization parameter of the image to be encoded, to the figure to be encoded As being encoded.
Optionally, in some embodiments, which is specifically used for, and calculates the amount using first function model Change parameter, which only includes logarithm operation, multiplying, addition and/or subtraction, the first function model Independent variable include the image complexity and the pre-arranged code bit.
Optionally, in some embodiments, the coefficient in the first function model includes at least one following parameter: work multiplies First parameter of method operation, the second parameter for making multiplying and the third parameter for making addition and/or subtraction, this first Parameter, second parameter and the third parameter are the parameter for code rate control.
Optionally, in some embodiments, the first function model are as follows: the logarithm operation value of the image complexity and this The product of one parameter, the product of the logarithm operation value of the pre-arranged code bit and the third parameter, with second parameter and.
Optionally, in some embodiments, first parameter is identical with the absolute value of second parameter.
Optionally, in some embodiments, after the coding for completing the picture frame where the image to be encoded, the processing list Member 720 is also used to: according to the frame level actual coding bit of the picture frame and frame level pre-arranged code bit, the third parameter is updated, The frame level actual coding bit indicates the actual coding bit of the picture frame, which indicates the picture frame Frame level pre-arranged code bit.
Optionally, in some embodiments, which is specifically used for, and according to second function model, is updated Third parameter afterwards, the independent variable of the second function model include the frame level actual coding bit, the frame level pre-arranged code bit, And the third parameter before updating, which comprises at least one of the following operation: logarithm operation, multiplying, Addition and/or subtraction.
Optionally, in some embodiments, second function model are as follows: the logarithm operation value of the frame level actual coding bit And the product of the difference of the logarithm operation value of the frame level pre-arranged code bit and renewal rate value and first parameter, with update The sum of preceding third parameter.
Optionally, in some embodiments, one in picture frame of the image to be encoded where the image to be encoded Image block；The processing unit 720 is specifically used for, according to the frame level pre-arranged code bit of the picture frame, the frame level figure of the picture frame As complexity and the image complexity of the image to be encoded, the pre-arranged code bit of the image to be encoded is determined.
Optionally, in some embodiments, which is specifically used for, multiple according to the image of the image to be encoded Miscellaneous to spend the ratio for accounting for the frame level image complexity of the picture frame, the product with the frame level pre-arranged code bit of the picture frame determines The pre-arranged code bit of the image to be encoded.
Optionally, in some embodiments, one in picture frame of the image to be encoded where the image to be encoded Image block；The processing unit 720 is also used to, and according to the image complexity of the picture frame and frame level pre-arranged code bit, obtaining should The quantization parameter of picture frame；When encoding to the image to be encoded, which is specifically used for, when this is to be encoded When the quantization parameter of image is in the range of the quantization parameter of the picture frame, according to the quantization parameter of the image to be encoded, to this Image to be encoded is encoded.
Optionally, in some embodiments, when encoding to the image to be encoded, which is specifically used In when in the range of quantization parameter of the quantization parameter of the image to be encoded not in the picture frame, by the image to be encoded Quantization parameter is cut out to the quantization parameter of the picture frame, the quantization ginseng of the image to be encoded after being cut out according to this It is several that the image to be encoded is encoded.
Optionally, in some embodiments, the picture frame where the image to be encoded is I frame.
Optionally, in some embodiments, the picture frame where the image to be encoded is P frame, which is the P An image block in frame；The processing unit 720 is specifically used for, according to the frame level predetermined bit of the P frame and multiplying for the first coefficient Product determines the predetermined bit of the image to be encoded, wherein first coefficient is determined according at least one of following: the P frame The sum of weight of all image blocks in overview image complexity, the P frame, the image to be encoded image complexity, this is to be encoded The frame level predetermined bit of the weight of image and the P frame, wherein the weight is determined by the code rate control parameter of the P frame.
Optionally, in some embodiments, which is specifically used for, which is to be encoded according to this The image complexity of image accounts for the ratio of the overview image complexity of the P frame and the product of the second coefficient, with the image to be encoded Weight account for what the sum of products of the ratio of the sum of weight of all image blocks and third coefficient in the P frame determined, wherein this It is that two coefficients and the third coefficient are all larger than the 0, and second coefficient and third coefficient and be equal to 1.
Optionally, in some embodiments, the image complexity of the image to be encoded is the line according to the image to be encoded Reason information obtains.
Optionally, in some embodiments, the image complexity of the image to be encoded indicates the image inner region to be encoded The fluctuation severe degree of pixel.
Optionally, in some embodiments, the image complexity of the image to be encoded is by counting the image to be encoded It is discrete just/transformation coefficient of cosine transform or Hadamard transform obtains.
Optionally, in some embodiments, the image complexity of the image to be encoded is by calculating the image to be encoded What the variance or standard deviation of interior pixel obtained.
Optionally, in some embodiments, which is image-signal processor ISP.
Processing unit 710 in the present embodiment specifically includes: code rate control unit, predicting unit, converter unit, inverse transformation Unit, quantifying unit, inverse quantization unit, entropy code unit.Wherein, code rate control unit is for realizing code rate control, for quantization Unit provides quantization parameter.Specifically, code rate control unit is used to receive image complexity from image processor, then according to figure As complexity and pre-arranged code bit, quantization parameter is calculated.
As shown in figure 8, the embodiment of the present invention also provides a kind of code device 800, processor 810 and memory 820, this is deposited For storing instruction, which is used to execute the instruction of the memory 820 storage to reservoir 820, and to the memory The execution of the instruction stored in 820 so that, the method which is used to execute embodiment of the method above.
Specifically, which further includes transceiver 830, for connecing from external image processor (such as ISP) Receive image complexity information.
The embodiment of the present invention also provides a kind of computer storage medium, is stored thereon with computer program, the computer journey Make when sequence is computer-executed, the method which executes embodiment of the method above.
The embodiment of the present invention also provides a kind of computer program product comprising instruction, which is characterized in that the instruction is counted The method that calculation machine makes computer execute embodiment of the method above when executing.
In the above-described embodiments, can come wholly or partly by software, hardware, firmware or any other combination real It is existing.When implemented in software, it can entirely or partly realize in the form of a computer program product.The computer program Product includes one or more computer instructions.When loading on computers and executing the computer program instructions, all or It partly generates according to process or function described in the embodiment of the present invention.The computer can be general purpose computer, dedicated meter Calculation machine, computer network or other programmable devices.The computer instruction can store in computer readable storage medium In, or from a computer readable storage medium to the transmission of another computer readable storage medium, for example, the computer Instruction can pass through wired (such as coaxial cable, optical fiber, number from a web-site, computer, server or data center User's line (digital subscriber line, DSL)) or wireless (such as infrared, wireless, microwave etc.) mode to another Web-site, computer, server or data center are transmitted.The computer readable storage medium can be computer capacity Any usable medium enough accessed either includes that the data such as one or more usable mediums integrated server, data center are deposited Store up equipment.The usable medium can be magnetic medium (for example, floppy disk, hard disk, tape), optical medium (such as digital video light Disk (digital video disc, DVD)) or semiconductor medium (such as solid state hard disk (solid state disk, SSD)) etc..
Those of ordinary skill in the art may be aware that list described in conjunction with the examples disclosed in the embodiments of the present disclosure Member and algorithm steps can be realized with the combination of electronic hardware or computer software and electronic hardware.These functions are actually It is implemented in hardware or software, the specific application and design constraint depending on technical solution.Professional technician Each specific application can be used different methods to achieve the described function, but this realization is it is not considered that exceed Scope of the present application.
In several embodiments provided herein, it should be understood that disclosed systems, devices and methods, it can be with It realizes by another way.For example, code device embodiment described above is only schematical, for example, the list Member division, only a kind of logical function partition, there may be another division manner in actual implementation, for example, multiple units or Component can be combined or can be integrated into another system, or some features can be ignored or not executed.Another point is shown The mutual coupling, direct-coupling or communication connection shown or discussed can be through some interfaces, between device or unit Coupling or communication connection are connect, can be electrical property, mechanical or other forms.
The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple In network unit.It can select some or all of unit therein according to the actual needs to realize the mesh of this embodiment scheme 's.
It, can also be in addition, each functional unit in each embodiment of the application can integrate in one processing unit It is that each unit physically exists alone, can also be integrated in one unit with two or more units.
The above, the only specific embodiment of the application, but the protection scope of the application is not limited thereto, it is any Those familiar with the art within the technical scope of the present application, can easily think of the change or the replacement, and should all contain Lid is within the scope of protection of this application.Therefore, the protection scope of the application should be based on the protection scope of the described claims.

Claims (43)

1. a kind of method of code rate control, which is characterized in that the method is applied to video encoder, which comprises
The image complexity of image to be encoded is received, described image complexity is obtained by the image processor being connected with encoder 's；
Determine the pre-arranged code bit of the image to be encoded；
According to described image complexity and the pre-arranged code bit, the quantization parameter of the image to be encoded is obtained；
According to the quantization parameter of the image to be encoded, the image to be encoded is encoded.
2. the method according to claim 1, wherein described according to described image complexity and the pre-arranged code Bit obtains the quantization parameter of the image to be encoded, comprising:
Calculate the quantization parameter using first function model, the first function model only include logarithm operation, multiplying, Addition and/or subtraction, the independent variable of the first function model include described image complexity and the pre-arranged code ratio It is special.
3. according to the method described in claim 2, it is characterized in that, the coefficient in the first function model include it is following at least One parameter: make the first parameter of multiplying, make the second parameter of multiplying and make the of addition and/or subtraction Three parameters, first parameter, second parameter and the third parameter are the parameter for code rate control.
4. according to the method described in claim 3, it is characterized in that, the first function model includes: described image complexity Logarithm operation value and the product of first parameter, the logarithm operation value of the pre-arranged code bit and the third parameter Product, with second parameter and.
5. the method according to claim 3 or 4, which is characterized in that first parameter and second parameter it is absolute It is worth identical.
6. method according to any one of claim 3 to 5, which is characterized in where completing the image to be encoded Picture frame coding after, the method also includes:
According to the frame level actual coding bit of described image frame and frame level pre-arranged code bit, the third parameter is updated, it is described Frame level actual coding bit indicates that the actual coding bit of described image frame, the frame level pre-arranged code bit indicate described image The frame level pre-arranged code bit of frame.
7. according to the method described in claim 6, it is characterized in that, the frame level actual coding bit according to described image frame With frame level pre-arranged code bit, the third parameter is updated, comprising:
According to second function model, updated third parameter is obtained, the independent variable of the second function model includes the frame Third parameter before grade actual coding bit, the frame level pre-arranged code bit and update, the second function model packet Include following at least one operation: logarithm operation, multiplying, addition and/or subtraction.
8. being compiled the method according to the description of claim 7 is characterized in that the second function model includes: that the frame level is practical Difference and the renewal rate value of the logarithm operation value of the logarithm operation value and frame level pre-arranged code bit of code bit and described The product of first parameter, with the third parameter before update and.
9. the method according to any one of claims 1 to 5, which is characterized in that the image to be encoded is described wait compile An image block in picture frame where code image；
The pre-arranged code bit of the determination image to be encoded, comprising:
According to the frame level pre-arranged code bit of described image frame, the frame level image complexity of described image frame and described to be encoded The image complexity of image determines the pre-arranged code bit of the image to be encoded.
10. according to the method described in claim 9, it is characterized in that, the pre-arranged code ratio of the determination image to be encoded Spy specifically includes:
The ratio that the frame level image complexity of described image frame is accounted for according to the image complexity of the image to be encoded, with the figure As the product of the frame level pre-arranged code bit of frame, the pre-arranged code bit of the image to be encoded is determined.
11. according to claim 1 to method described in any one of 5,9 or 10, which is characterized in that the image to be encoded is institute State an image block in the picture frame where image to be encoded；
The method also includes:
According to the image complexity of described image frame and frame level pre-arranged code bit, the quantization parameter of described image frame is obtained；
The quantization parameter according to the image to be encoded, encodes the image to be encoded, comprising:
When the quantization parameter of the image to be encoded is in the range of the quantization parameter of described image frame, according to described to be encoded The quantization parameter of image encodes the image to be encoded.
12. according to the method for claim 11, which is characterized in that the quantization parameter according to the image to be encoded, The image to be encoded is encoded, further includes:
It, will be described to be encoded when the quantization parameter of the image to be encoded is not in the range of the quantization parameter of described image frame The quantization parameter of image is cut out to the quantization parameter of described image frame, according to described to be encoded after described cut out The quantization parameter of image encodes the image to be encoded.
13. method according to any one of claim 1 to 12, which is characterized in that the figure where the image to be encoded As frame is I frame.
14. the method according to claim 1, wherein the picture frame where the image to be encoded is P frame, institute Stating image to be encoded is an image block in the P frame；
The pre-arranged code bit of the determination image to be encoded, comprising:
The predetermined bit of the image to be encoded is determined according to the product of the frame level predetermined bit of the P frame and the first coefficient, In, first coefficient is determined according at least one of following: being owned in the overview image complexity of the P frame, the P frame The sum of weight of image block, the image complexity of the image to be encoded, the weight of the image to be encoded and the P frame Frame level predetermined bit, wherein the weight is determined by the code rate control parameter of the P frame.
15. according to the method for claim 14, which is characterized in that first coefficient is according to the image to be encoded Image complexity accounts for the ratio of the overview image complexity of the P frame and the product of the second coefficient, with the image to be encoded Weight accounts for what the ratio of the sum of weight of all image blocks and the sum of products of third coefficient in the P frame determined, wherein described Second coefficient and the third coefficient are all larger than 0, and second coefficient and third coefficient and be equal to 1.
16. according to claim 1 to method described in any one of 15, which is characterized in that the image of the image to be encoded is multiple Miscellaneous degree is obtained according to the texture information of the image to be encoded.
17. according to claim 1 to method described in any one of 15, which is characterized in that the image of the image to be encoded is multiple Miscellaneous degree indicates the fluctuation severe degree of the image inner region pixel to be encoded.
18. according to the method for claim 17, which is characterized in that the image complexity of the image to be encoded is to pass through system Count the image to be encoded it is discrete just/transformation coefficient of cosine transform or Hadamard transform obtains.
19. according to the method for claim 17, which is characterized in that the image complexity of the image to be encoded is to pass through meter What the variance or standard deviation for calculating the pixel in the image to be encoded obtained.
20. according to claim 1 to method described in any one of 19, which is characterized in that described image processor is image letter Number processor ISP.
21. a kind of code device characterized by comprising
Transmit-Receive Unit, for receiving the image complexity of image to be encoded, described image complexity is by being connected with encoder What image processor obtained；
Processing unit, for determining the pre-arranged code bit of the image to be encoded；According to described image complexity with it is described pre- If coded-bit, the quantization parameter of the image to be encoded is obtained；According to the quantization parameter of the image to be encoded, to it is described to Coded image is encoded.
22. code device according to claim 21, which is characterized in that the processing unit is specifically used for, and utilizes first Function model calculates the quantization parameter, and the first function model only includes logarithm operation, multiplying, addition and/or subtracts Method operation, the independent variable of the first function model include described image complexity and the pre-arranged code bit.
23. code device according to claim 22, which is characterized in that the coefficient in the first function model include with At least one lower parameter: make the first parameter of multiplying, make the second parameter of multiplying and make addition and/or subtraction fortune The third parameter of calculation, first parameter, second parameter and the third parameter are the parameter for code rate control.
24. code device according to claim 23, which is characterized in that the first function model includes: described image The product of the logarithm operation value of complexity and first parameter, the logarithm operation value of the pre-arranged code bit and the third The product of parameter, with second parameter and.
25. the code device according to claim 23 or 24, which is characterized in that first parameter and second parameter Absolute value it is identical.
26. the code device according to any one of claim 22 to 25, which is characterized in that complete the figure to be encoded As the picture frame at place coding after, the processing unit is also used to:
According to the frame level actual coding bit of described image frame and frame level pre-arranged code bit, the third parameter is updated, it is described Frame level actual coding bit indicates that the actual coding bit of described image frame, the frame level pre-arranged code bit indicate described image The frame level pre-arranged code bit of frame.
27. code device according to claim 26, which is characterized in that the processing unit is specifically used for, according to second Function model, obtains updated third parameter, and the independent variable of the second function model includes the frame level actual coding ratio Third parameter before special, the described frame level pre-arranged code bit and update, the second function model includes following at least one Kind operation: logarithm operation, multiplying, addition and/or subtraction.
28. code device according to claim 27, which is characterized in that the second function model includes: the frame level The difference of the logarithm operation value of actual coding bit and the logarithm operation value of the frame level pre-arranged code bit and renewal rate value with And the product of first parameter, with the third parameter before update and.
29. the code device according to any one of claim 21 to 25, which is characterized in that the image to be encoded is institute State an image block in the picture frame where image to be encoded；
The processing unit is specifically used for, according to the frame level pre-arranged code bit of described image frame, the frame level figure of described image frame As complexity and the image complexity of the image to be encoded, the pre-arranged code bit of the image to be encoded is determined.
30. code device according to claim 29, which is characterized in that the processing unit is specifically used for, according to described The image complexity of image to be encoded accounts for the ratio of the frame level image complexity of described image frame, pre- with the frame level of described image frame If the product of coded-bit, the pre-arranged code bit of the image to be encoded is determined.
31. the code device according to any one of claim 21 to 25,29 or 30, which is characterized in that described to be encoded An image block of the image in the picture frame where the image to be encoded；
The processing unit is also used to, according to the image complexity of described image frame and frame level pre-arranged code bit, described in acquisition The quantization parameter of picture frame；
When encoding to the image to be encoded, the processing unit is specifically used for, when the quantization of the image to be encoded When parameter is in the range of the quantization parameter of described image frame, according to the quantization parameter of the image to be encoded, to described wait compile Code image is encoded.
32. code device according to claim 31, which is characterized in that when being encoded to the image to be encoded, The processing unit is specifically used for, when the quantization parameter of the image to be encoded is not in the range of the quantization parameter of described image frame When interior, the quantization parameter of the image to be encoded is cut out to the quantization parameter of described image frame, is cut according to described The quantization parameter of the image to be encoded after sanction encodes the image to be encoded.
33. the code device according to any one of claim 21 to 32, which is characterized in where the image to be encoded Picture frame be I frame.
34. code device according to claim 21, which is characterized in that the picture frame where the image to be encoded is P Frame, the image to be encoded are an image block in the P frame；
The processing unit is specifically used for, according to the product of the frame level predetermined bit of the P frame and the first coefficient determine it is described to The predetermined bit of coded image, wherein first coefficient is determined according at least one of following: the overall pattern of the P frame As the sum of weight of all image blocks in complexity, the P frame, the image complexity, described to be encoded of the image to be encoded The frame level predetermined bit of the weight of image and the P frame, wherein the weight is determined by the code rate control parameter of the P frame.
35. code device according to claim 34, which is characterized in that first coefficient is according to the figure to be encoded The image complexity of picture accounts for the ratio of the overview image complexity of the P frame and the product of the second coefficient, with the figure to be encoded The weight of picture accounts for what the ratio of the sum of weight of all image blocks and the sum of products of third coefficient in the P frame determined, wherein Second coefficient and the third coefficient are all larger than 0, and second coefficient and third coefficient and be equal to 1.
36. the code device according to any one of claim 21 to 35, which is characterized in that the figure of the image to be encoded As complexity is obtained according to the texture information of the image to be encoded.
37. the code device according to any one of claim 21 to 36, which is characterized in that the figure of the image to be encoded As complexity indicates the fluctuation severe degree of the image inner region pixel to be encoded.
38. the code device according to claim 37, which is characterized in that the image complexity of the image to be encoded is logical Cross count the image to be encoded it is discrete just/transformation coefficient of cosine transform or Hadamard transform obtains.
39. the code device according to claim 37, which is characterized in that the image complexity of the image to be encoded is logical Cross what the variance for calculating the pixel in the image to be encoded or standard deviation obtained.
40. the code device according to any one of claim 21 to 39, which is characterized in that described image processor is figure As signal processor ISP.
41. a kind of code device characterized by comprising memory and processor, the memory for storing instruction, institute Processor is stated for executing the instruction of memory storage, and the execution to the instruction stored in the memory so that, The processor is used to execute the method as described in any one of claims 1 to 20.
42. a kind of computer storage medium, which is characterized in that be stored thereon with computer program, the computer program is counted Calculation machine makes when executing, and the computer executes the method as described in any one of claims 1 to 20.
43. a kind of computer program product comprising instruction, which is characterized in that make to calculate when described instruction is computer-executed Machine executes the method as described in any one of claims 1 to 20.
CN201780018104.7A 2017-12-27 2017-12-27 The method and code device of code rate control Pending CN109076212A (en)

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2017/119111 WO2019127136A1 (en) 2017-12-27 2017-12-27 Bit rate control method and encoding device

Publications (1)

Publication Number Publication Date
CN109076212A true CN109076212A (en) 2018-12-21

Family

ID=64812361

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201780018104.7A Pending CN109076212A (en) 2017-12-27 2017-12-27 The method and code device of code rate control

Country Status (2)

Country Link
CN (1) CN109076212A (en)
WO (1) WO2019127136A1 (en)

Cited By (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110536168A (en) * 2019-09-11 2019-12-03 北京达佳互联信息技术有限公司 Video method for uploading, device, electronic equipment and storage medium
CN110740324A (en) * 2019-09-11 2020-01-31 浙江大华技术股份有限公司 Coding control method and related device
CN111247800A (en) * 2019-01-25 2020-06-05 深圳市大疆创新科技有限公司 Method and device for determining image information quantity
CN111264062A (en) * 2019-02-21 2020-06-09 深圳市大疆创新科技有限公司 Encoder, encoding system and encoding method
CN111279697A (en) * 2019-01-25 2020-06-12 深圳市大疆创新科技有限公司 Code rate allocation method and device
CN111447446A (en) * 2020-05-15 2020-07-24 西北民族大学 HEVC (high efficiency video coding) rate control method based on human eye visual region importance analysis
CN112218083A (en) * 2020-10-16 2021-01-12 西安邮电大学 Method for estimating intra-frame image code rate of high-efficiency video coding standard
CN112492305A (en) * 2020-11-18 2021-03-12 腾讯科技（深圳）有限公司 Data processing method and device and computer readable storage medium
CN113438488A (en) * 2021-06-18 2021-09-24 北京锐马视讯科技有限公司 Low-bit-rate video optimization coding method, device, equipment and storage medium
CN113453001A (en) * 2020-03-24 2021-09-28 合肥君正科技有限公司 Method for improving H264 coding efficiency by using ISP information to adaptively allocate QP
CN113784126A (en) * 2021-09-17 2021-12-10 Oppo广东移动通信有限公司 Image encoding method, apparatus, device and storage medium
WO2022012183A1 (en) * 2020-07-17 2022-01-20 Oppo广东移动通信有限公司 Encoding method, encoder and computer readable storage medium
TWI774383B (en) * 2021-03-22 2022-08-11 瑞昱半導體股份有限公司 Encoder and associated signal processing method
WO2022252222A1 (en) * 2021-06-04 2022-12-08 深圳市大疆创新科技有限公司 Encoding method and encoding device
WO2023019910A1 (en) * 2021-08-20 2023-02-23 上海商汤智能科技有限公司 Video processing method and apparatus, electronic device, storage medium, computer program and computer program product
WO2023138532A1 (en) * 2022-01-21 2023-07-27 杭州海康威视数字技术股份有限公司 Video decoding method and apparatus, video decoder and storage medium
WO2024051299A1 (en) * 2022-09-08 2024-03-14 华为技术有限公司 Encoding method and apparatus, and decoding method and apparatus

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101094411A (en) * 2007-07-03 2007-12-26 芯瀚电子技术（上海）有限公司 Code rate control method of video code
CN101127909A (en) * 2007-09-29 2008-02-20 华为技术有限公司 A control method and device for image code rate
CN101754003A (en) * 2008-12-11 2010-06-23 北京威速科技有限公司 Code rate control method for stabilizing video quality
CN102067610A (en) * 2008-06-16 2011-05-18 杜比实验室特许公司 Rate control model adaptation based on slice dependencies for video coding
CN102938840A (en) * 2012-11-26 2013-02-20 南京邮电大学 Key frame quantization parameter selecting method applied to multi-viewpoint video coding system
CN104754335A (en) * 2015-03-09 2015-07-01 湖南国科微电子有限公司 Video coding rate control method
US20150189297A1 (en) * 2011-10-04 2015-07-02 Thomson Licensing Device and method for intra-encoding image blocks, storage medium carrying and use of a processing device for decoding intra-encoded image blocks

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040120404A1 (en) * 2002-11-27 2004-06-24 Takayuki Sugahara Variable length data encoding method, variable length data encoding apparatus, variable length encoded data decoding method, and variable length encoded data decoding apparatus
CN101632308B (en) * 2007-03-14 2011-08-03 日本电信电话株式会社 Encoding bit rate control method and device
CN102036062B (en) * 2009-09-29 2012-12-19 华为技术有限公司 Video coding method and device and electronic equipment
JP6094956B2 (en) * 2012-09-26 2017-03-15 パナソニックＩｐマネジメント株式会社 Image encoding apparatus, captured image recording system, imaging apparatus, image encoding method, and image encoding program
CN105208390B (en) * 2014-06-30 2018-07-20 杭州海康威视数字技术股份有限公司 The bit rate control method and its system of Video coding

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101094411A (en) * 2007-07-03 2007-12-26 芯瀚电子技术（上海）有限公司 Code rate control method of video code
CN101127909A (en) * 2007-09-29 2008-02-20 华为技术有限公司 A control method and device for image code rate
CN102067610A (en) * 2008-06-16 2011-05-18 杜比实验室特许公司 Rate control model adaptation based on slice dependencies for video coding
CN101754003A (en) * 2008-12-11 2010-06-23 北京威速科技有限公司 Code rate control method for stabilizing video quality
US20150189297A1 (en) * 2011-10-04 2015-07-02 Thomson Licensing Device and method for intra-encoding image blocks, storage medium carrying and use of a processing device for decoding intra-encoded image blocks
CN102938840A (en) * 2012-11-26 2013-02-20 南京邮电大学 Key frame quantization parameter selecting method applied to multi-viewpoint video coding system
CN104754335A (en) * 2015-03-09 2015-07-01 湖南国科微电子有限公司 Video coding rate control method

Cited By (25)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111247800A (en) * 2019-01-25 2020-06-05 深圳市大疆创新科技有限公司 Method and device for determining image information quantity
CN111279697A (en) * 2019-01-25 2020-06-12 深圳市大疆创新科技有限公司 Code rate allocation method and device
CN111247800B (en) * 2019-01-25 2022-07-19 深圳市大疆创新科技有限公司 Method and device for determining image information quantity
CN111264062A (en) * 2019-02-21 2020-06-09 深圳市大疆创新科技有限公司 Encoder, encoding system and encoding method
WO2020168520A1 (en) * 2019-02-21 2020-08-27 深圳市大疆创新科技有限公司 Encoder, encoding system, and encoding method
CN110536168B (en) * 2019-09-11 2021-09-17 北京达佳互联信息技术有限公司 Video uploading method and device, electronic equipment and storage medium
CN110740324A (en) * 2019-09-11 2020-01-31 浙江大华技术股份有限公司 Coding control method and related device
CN110740324B (en) * 2019-09-11 2023-03-31 浙江大华技术股份有限公司 Coding control method and related device
CN110536168A (en) * 2019-09-11 2019-12-03 北京达佳互联信息技术有限公司 Video method for uploading, device, electronic equipment and storage medium
CN113453001A (en) * 2020-03-24 2021-09-28 合肥君正科技有限公司 Method for improving H264 coding efficiency by using ISP information to adaptively allocate QP
CN111447446B (en) * 2020-05-15 2022-08-23 西北民族大学 HEVC (high efficiency video coding) rate control method based on human eye visual region importance analysis
CN111447446A (en) * 2020-05-15 2020-07-24 西北民族大学 HEVC (high efficiency video coding) rate control method based on human eye visual region importance analysis
WO2022012183A1 (en) * 2020-07-17 2022-01-20 Oppo广东移动通信有限公司 Encoding method, encoder and computer readable storage medium
CN112218083B (en) * 2020-10-16 2021-09-17 西安邮电大学 Method for estimating intra-frame image code rate of high-efficiency video coding standard
CN112218083A (en) * 2020-10-16 2021-01-12 西安邮电大学 Method for estimating intra-frame image code rate of high-efficiency video coding standard
CN112492305A (en) * 2020-11-18 2021-03-12 腾讯科技（深圳）有限公司 Data processing method and device and computer readable storage medium
CN112492305B (en) * 2020-11-18 2022-02-11 腾讯科技（深圳）有限公司 Data processing method and device and computer readable storage medium
US11523117B2 (en) 2021-03-22 2022-12-06 Realtek Semiconductor Corp. Encoder using coding tree unit level adaptive quantization mode to adjust frame level quantization parameter and associated signal processing method
TWI774383B (en) * 2021-03-22 2022-08-11 瑞昱半導體股份有限公司 Encoder and associated signal processing method
WO2022252222A1 (en) * 2021-06-04 2022-12-08 深圳市大疆创新科技有限公司 Encoding method and encoding device
CN113438488A (en) * 2021-06-18 2021-09-24 北京锐马视讯科技有限公司 Low-bit-rate video optimization coding method, device, equipment and storage medium
WO2023019910A1 (en) * 2021-08-20 2023-02-23 上海商汤智能科技有限公司 Video processing method and apparatus, electronic device, storage medium, computer program and computer program product
CN113784126A (en) * 2021-09-17 2021-12-10 Oppo广东移动通信有限公司 Image encoding method, apparatus, device and storage medium
WO2023138532A1 (en) * 2022-01-21 2023-07-27 杭州海康威视数字技术股份有限公司 Video decoding method and apparatus, video decoder and storage medium
WO2024051299A1 (en) * 2022-09-08 2024-03-14 华为技术有限公司 Encoding method and apparatus, and decoding method and apparatus

Also Published As

Publication number Publication date
WO2019127136A1 (en) 2019-07-04

Similar Documents

Publication Publication Date Title
CN109076212A (en) The method and code device of code rate control
CN106716997A (en) Video encoding method using in-loop filter parameter prediction and apparatus therefor, and video decoding method and apparatus therefor
US6160846A (en) Apparatus and method for optimizing the rate control in a coding system
CN104602010B (en) Method and apparatus for encoding and decoding coding unit of picture boundary
CN104796714B (en) Use the enhancing intraframe predictive coding of plane expression
CN110235444A (en) Use the intra prediction of a plurality of reference line
JP2000511366A (en) Apparatus and method for variable block size motion estimation based on quadrant tree
CN110199522A (en) Processing is used for the reference sample of image block intra prediction
CN104641643A (en) Decomposition of residual data during signal encoding, decoding and reconstruction in a tiered hierarchy
CN104219522B (en) Bit rate control method and device in a kind of encoding video pictures
KR20010021879A (en) Apparatus and method for macroblock based rate control in a coding system
CN106688232A (en) Perceptual optimization for model-based video encoding
CN110876060B (en) Code rate adjusting method and device in coding process
WO2004104930A2 (en) Hybrid video compression method
JP4949836B2 (en) System and method for encoding and decoding enhancement layer data using descriptive model parameters
CN107846593B (en) Rate distortion optimization method and device
CN107667528A (en) Method and system for the adaptive reference frame buffer of coding and decoding video
CN107005713A (en) Apply the method for video coding and equipment and video encoding/decoding method and equipment of edge type skew
Zheng et al. A high-efficiency compressed sensing-based terminal-to-cloud video transmission system
CN105850136B (en) Use the method and apparatus of prediction signal and transformation compiling signal estimation vision signal
CN104754335B (en) A kind of code rate controlling method for video coding
CN108353175A (en) The method and apparatus of prediction processing vision signal caused by coefficient of utilization
CN108781298A (en) Encoder, image processing system, unmanned plane and coding method
CN108040256A (en) It is a kind of based on bit rate control method H.265, system and device
CN108322749A (en) The coefficient optimization method of RDOQ, the accelerating method and device of RDOQ

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20181221

WD01 Invention patent application deemed withdrawn after publication