CN108650511B - Monitoring video frequency distortion optimization coding method based on background distortion propagation - Google Patents

Monitoring video frequency distortion optimization coding method based on background distortion propagation Download PDF

Info

Publication number
CN108650511B
CN108650511B CN201810458872.0A CN201810458872A CN108650511B CN 108650511 B CN108650511 B CN 108650511B CN 201810458872 A CN201810458872 A CN 201810458872A CN 108650511 B CN108650511 B CN 108650511B
Authority
CN
China
Prior art keywords
background
distortion
background block
block
coding
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201810458872.0A
Other languages
Chinese (zh)
Other versions
CN108650511A (en
Inventor
熊健
路丽果
桂冠
杨洁
范山岗
潘金秋
华文韬
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.)
Nanjing University of Posts and Telecommunications
Original Assignee
Nanjing University of Posts and Telecommunications
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 Nanjing University of Posts and Telecommunications filed Critical Nanjing University of Posts and Telecommunications
Priority to CN201810458872.0A priority Critical patent/CN108650511B/en
Publication of CN108650511A publication Critical patent/CN108650511A/en
Application granted granted Critical
Publication of CN108650511B publication Critical patent/CN108650511B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

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
    • H04N19/169—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
    • H04N19/10—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
    • H04N19/147—Data rate or code amount at the encoder output according to rate distortion criteria
    • 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
    • H04N19/149—Data rate or code amount at the encoder output by estimating the code amount by means of a model, e.g. mathematical model or statistical model
    • 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
    • H04N19/189—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the adaptation method, adaptation tool or adaptation type used for the adaptive coding
    • H04N19/19—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the adaptation method, adaptation tool or adaptation type used for the adaptive coding using optimisation based on Lagrange multipliers

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • Algebra (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Pure & Applied Mathematics (AREA)
  • Compression Or Coding Systems Of Tv Signals (AREA)

Abstract

The invention discloses a monitoring video frequency distortion optimization coding method based on background distortion propagation, which comprises the steps of firstly dividing a coding block into a background block and a foreground block, constructing a background block distortion propagation model, carrying out global rate distortion optimization based on the background block distortion propagation model to obtain parameters of the global rate distortion optimization, marking the background block by using a mask, carrying out lossless coding on the mask, coding the background block based on asymmetric quantization parameters, decoding the background block mask and the asymmetric quantization parameter offset value at a decoding end, and realizing the reconstruction of the background block. The invention realizes the improvement of the compression performance and the coding efficiency of the monitoring video and solves the technical problem of low coding efficiency caused by the mutual influence of the background and the foreground in the coding.

Description

Monitoring video frequency distortion optimization coding method based on background distortion propagation
Technical Field
The invention belongs to the field of resolution distortion optimization, and particularly relates to a monitoring video frequency distortion optimization coding method based on background distortion propagation.
Background
With the rapid development of multimedia technology and the increasing popularization of video monitoring equipment, video monitoring is widely applied to occasions such as safety production, intelligent traffic, safe cities and the like due to intuition, accuracy and timeliness and rich information content, and the monitoring video data volume shows explosive growth. At present, the number of video monitoring cameras in a large city of China is usually more than hundreds of thousands, and the number of installed cameras in cities across the country is more than twenty million. Video surveillance has become yet another important video application following digital television, video conferencing, and is increasingly becoming the largest video application system of "massiveness". In the face of the huge amount of monitoring videos collected by a large number of cameras and high storage cost, how to effectively encode and compress the videos has become a significant challenge in the current video monitoring field. The monitoring video compression coding technology is a precondition for links such as storage, transmission, analysis and identification, and is a core technology of video monitoring application.
The current video compression technologies are HEVC and h.264/AVC, and all of the technologies adopt a large number of coding modes and coding parameters. The technique of how to select the optimal coding mode and parameters to achieve the best compression performance is called rate-distortion optimization. The technology is based on the Shannon rate-distortion coding theory, so that the purposes of spending bits as few as possible, obtaining video quality as high as possible and achieving optimal rate-distortion performance are expected. In practical applications, the rate-distortion optimization technique is the most basic and most core technique in video coding, and almost all video coding optimizations are based on the rate-distortion optimization. In view of the importance of surveillance video compression technology and the criticality of rate-distortion optimization technology in video compression, the research on rate-distortion optimization technology oriented to surveillance video coding becomes the key of video multimedia application.
The core problem of video coding rate distortion optimization is how to minimize the distortion of video coding under the limit of bit consumption, which is expressed as min D, s.t.R < RcWhere D represents distortion, R represents bit consumption, R represents bit ratecDenotes the maximum bit consumption, s.t. denotes "under the conditions" which means that the distortion is minimized without exceeding a certain maximum bit consumption.
Sullivan et al propose to use the Lagrangian multiplier method, which, fundamentally, maps bit consumption to distortion using the Lagrangian multiplier, and further converts the constrained optimization problem to the unconstrained optimization problem. This method is widely used in the prior art of coding. On one hand, however, the lagrangian multiplier method is based on the approximation of shannon's theorem under the condition of high bit rate, and shows the performance deficiency in the low bit rate coding application, and a large number of background areas in the monitoring video are coded by using an extremely low number of bits, so that the method is not suitable for the background areas of the monitoring video; the lagrange multiplier, on the other hand, is a function set on the quantization step size according to experimental statistics, whose value is independent of the input video signal. Different parameters will result in different encoding results, and the compression performance obtained by the method is insufficient for diversified video contents in practical application.
The problems existing at present are as follows:
problem 1: the method is lack of research aiming at the rate-distortion characteristic differentiation of a background region and a foreground region of a monitored video, and the existing method is dedicated to constructing an adaptive model based on a Lagrange multiplier method, neglects that the model is not suitable for low-bit-rate coding of the background region of the monitored video, and lacks of introduction of a classification optimization idea.
Problem 2: the method lacks of interference factor research on the background block in the adaptive rate distortion model estimation, and the existing method generally uses the coding information of the background block for the estimation of the adaptive rate distortion model of the foreground block, neglects the interference influence of the coding information of the background block and cannot ensure the accuracy of the adaptive rate distortion model of the foreground block.
Problem 3: a global distortion optimization model established for the distortion propagation characteristics of a background block is lacked, the existing method ignores the extremely strong propagation characteristics of the background block of the monitoring video, and the thinking of global rate distortion optimization is lacked.
Disclosure of Invention
The invention aims to solve the problems and provides a monitoring video rate distortion optimization coding method based on background distortion propagation, which improves the compression performance and coding efficiency of the monitoring video and solves the technical problem of low coding efficiency caused by mutual influence of the background and the foreground in coding.
The invention adopts the following technical scheme that a monitoring video frequency distortion optimization coding method based on background distortion propagation comprises the following specific steps:
1) dividing the coding block into a background block and a foreground block;
2) constructing a background block distortion propagation model;
3) performing global rate distortion optimization based on the background block distortion propagation model to obtain parameters of the global rate distortion optimization;
4) marking a background block by using a mask, and carrying out lossless coding on the mask;
5) encoding the background block based on the asymmetric quantization parameter;
6) the background block mask and a Quantization Parameter (QP) offset value are decoded at a decoding end to reconstruct the background block.
Preferably, the coding block is divided into a background block and a foreground block in step 1), specifically, the background block is classified and detected to obtain a background block with continuous K frames in a time domain;
preferably, the building of the background block distortion propagation model in the step 2) specifically includes:
constructing a background block distortion propagation model based on a static background: for the background blocks of K continuous frames in the time domain, the background blocks are squares, and the distortion and bit consumption of the background block of the ith frame are respectively DiAnd RiWherein i is 1, 2i，jA source pixel value representing the ith frame image at coordinate position j,
Figure GDA0003070444840000031
representing the decoded pixel value of the pixel with the coordinate position j of the ith frame image, wherein the source pixel values of the same coordinate position of each frame background block in the encoding process are the same, namely pi，j＝pi-1，j＝...＝p1，j，pi-1，jAnd p1，jRespectively representing a source pixel value with the coordinate position of j of the i-1 frame image and a source pixel value with the coordinate position of j of the 1 st frame image, wherein the coordinate position j of the pixel is 1, 22N is the side length of the background block; and coding is carried out by adopting SKIP mode, namely the reconstructed pixels at the same position have propaganda,
Figure GDA0003070444840000032
Figure GDA0003070444840000035
and
Figure GDA0003070444840000033
respectively representing the pixel decoded pixel value with j as the coordinate position of the i-1 frame image, the pixel decoded pixel value with j as the coordinate position of the 1 st frame image and the distortion D of the i-th frame background blockiThe following relationships exist:
Figure GDA0003070444840000034
wherein D is1Representing the distortion of the first frame background block.
Preferably, the global rate-distortion optimization in step 3) is performed, and the specific steps for obtaining parameters of the global rate-distortion optimization are as follows:
31) the distortion and bit consumption of the background block of the ith frame are respectively DiAnd RiWherein i is 1, 2.. K, the global rate distortion optimization formula of the background block is based on the goal of the background block distortion propagation model to minimize the rate distortion cost of the background block of the continuous K frames as a whole
Figure GDA0003070444840000041
Wherein λcLagrange multiplier optimized for rate distortion, D1And R1Respectively representing the distortion and bit consumption of the background block of the first frame;
32) based on the background block distortion propagation model, the global rate distortion optimization formula of the background block is rewritten into
Figure GDA0003070444840000042
Are respectively to R1And RiAnd solving the partial derivative to obtain a Lagrange multiplier which has the following relation:
Figure GDA0003070444840000043
wherein λ1Lagrange multiplier, λ, representing the background block of the first frameiThe Lagrange multiplier of the background block of the ith frame is expressed, and the Lagrange multiplier of the subsequent frame is K times of the Lagrange multiplier of the first frame, namely lambdai＝K·λ1。
Preferably, in step 4), the background block is marked by using a mask, and the mask is subjected to lossless coding, specifically, K frames are used as one period at a coding end, each period is marked by using one mask, and the mask is coded by using arithmetic coding.
Preferably, the encoding of the background block based on the asymmetric quantization parameter in step 5) specifically includes:
based on an R-lambda model, the power function relationship between the bit consumption R and the Lagrange multiplier lambda is as follows:
R＝α·λβ
wherein alpha and beta are R-lambda model parameters, and keeping the total bit consumption of the current background block unchanged to obtain
α·λ1 β+(K-1)α·(K·λ1)β＝Kα·λc β
Wherein λcLagrange multiplier optimized for rate distortion, K being a background block of consecutive K frames in the time domain, the above formula being K>When 1, the equation can not be taken, namely the code rates before and after optimization can not be completely equal, but the code rate fluctuation can be effectively controlled, namely the Lagrange multiplier of the first frame background block under the condition of encoding the background block based on the asymmetric quantization parameter is obtained as
Figure GDA0003070444840000051
Wherein,
Figure GDA0003070444840000052
representing the case of encoding a background block based on asymmetric quantization parametersThe lagrangian multiplier of the first frame background block of (1);
the quantization parameter is
QP＝4.2005·ln(λ)+13.7122
Will be provided with
Figure GDA0003070444840000053
And λiAnd substituting the quantization parameters into the formula respectively to obtain the quantization parameter of each frame, and coding all background blocks marked by the mask by adopting the corresponding quantization parameters.
The invention has the following beneficial effects: the invention provides a monitoring video frequency distortion optimization coding method based on background distortion propagation, which realizes the improvement of monitoring video compression performance and coding efficiency and solves the technical problem of low coding efficiency caused by the mutual influence of a background and a foreground in coding; the invention combines the classification detection of the coding blocks with various rate distortion optimization models, optimizes the rate distortion characteristic difference of the coding blocks, has important significance for improving the compression performance and the coding efficiency of the monitoring video, and can be applied to a plurality of fields of video coding, including monitoring videos, video conferences and the like.
Drawings
FIG. 1 is a flow chart of a background block global rate-distortion optimization method based on a background distortion propagation model according to the present invention;
fig. 2 is a schematic diagram of a monitoring video background block distortion propagation model.
Detailed Description
The technical solution of the present invention is further explained with reference to the embodiments according to the drawings.
The invention adopts the following technical scheme, a monitoring video frequency distortion optimization coding method based on background distortion propagation, is realized on an HM16.0(HEVC official test software) experiment platform, and fig. 1 is a flow chart of the invention and comprises the following specific steps:
1) dividing the coding block into a background block and a foreground block;
dividing the coding block into a background block and a foreground block in the step 1), specifically, carrying out classification detection on the background block to obtain K continuous frame background blocks in a time domain; from the background consistency of the surveillance video, it is reasonable to predict that the position coding block is also a background block in the following frames.
2) Constructing a distortion propagation model;
the construction of the background block distortion propagation model in the step 2) specifically comprises the following steps:
constructing a background block distortion propagation model based on a static background: for the background blocks of K continuous frames in the time domain, the background blocks are squares, and the distortion and bit consumption of the background block of the ith frame are respectively DiAnd RiWherein i is 1, 2i，jA source pixel value representing the ith frame image at coordinate position j,
Figure GDA0003070444840000061
representing the decoded pixel value of the pixel with the coordinate position j of the ith frame image, wherein the source pixel values of the same coordinate position of each frame background block in the encoding process are the same, namely pi，j＝pi-1，j＝...＝p1，j，pi-1，jAnd p1，jRespectively representing a source pixel value with the coordinate position of j of the i-1 frame image and a source pixel value with the coordinate position of j of the 1 st frame image, wherein the coordinate position j of the pixel is 1, 22N is the side length of the background block; and coding is carried out by adopting SKIP mode, namely the reconstructed pixels at the same position have propaganda,
Figure GDA0003070444840000062
Figure GDA0003070444840000065
and
Figure GDA0003070444840000063
respectively representing the pixel decoded pixel value with j as the coordinate position of the i-1 frame image, the pixel decoded pixel value with j as the coordinate position of the 1 st frame image and the distortion D of the i-th frame background blockiThe following relationships exist:
Figure GDA0003070444840000064
wherein D is1Representing the distortion of the first frame background block.
3) As shown in fig. 2, the horizontal axis represents the coding bit consumption, the vertical axis represents the corresponding coding distortion, and each point in the figure represents the coding bit consumption and the coding distortion under different coding parameters. Carrying out global rate distortion optimization based on the background block distortion propagation model to obtain parameters of the global rate distortion optimization, and specifically comprising the following steps:
31) the distortion and bit consumption of the background block of the ith frame are respectively DiAnd RiWherein i is 1, 2.. K, the global rate distortion optimization target is to minimize the rate distortion cost of the background block of the continuous K frames, and the global rate distortion optimization formula of the background block is
Figure GDA0003070444840000071
Wherein λcLagrange multiplier optimized for rate distortion, D1And R1Respectively representing the distortion and bit consumption of the background block of the first frame;
32) based on the background block distortion propagation model, the global rate distortion optimization formula of the background block is rewritten into
Figure GDA0003070444840000072
Are respectively to R1And RiAnd solving the partial derivative to obtain a Lagrange multiplier which has the following relation:
Figure GDA0003070444840000073
wherein λ1Lagrange multiplier, λ, representing the background block of the first frameiThe Lagrange multiplier of the background block of the ith frame is expressed, and the Lagrange multiplier of the subsequent frame is K times of the Lagrange multiplier of the first frame, namely lambdai＝K·λ1。
4) The method comprises the steps of marking a background block by using a mask, and carrying out lossless coding on the mask, specifically, taking a K frame as a period at a coding end, marking the background block by using one mask in each period, and coding the mask by adopting arithmetic coding.
5) Encoding the background block based on the asymmetric quantization parameter;
in order to keep the total coding bit changing in a small range, in the actual coding, a corresponding quantization parameter needs to be selected according to the relationship, the adjusted lagrangian multiplier is calculated by using the relationship between the coding bit and the lagrangian multiplier, and the step 5) of coding the background block based on the asymmetric quantization parameter specifically comprises the following steps:
the background block global rate-distortion optimization scheme based on the distortion propagation model is an optimization result based on global rate-distortion cost minimization, the code rate of a first frame is greatly increased after optimization, the overall code rate has large fluctuation, and the change of the code rate before and after optimization is not considered. Therefore, the code rate before and after optimization needs to be constrained, specifically according to an R- λ model:
R＝α·λβ
wherein alpha and beta are R-lambda model parameters, and keeping the total bit consumption of the current background block unchanged to obtain
α·λ1 β+(K-1)α·(K·λ1)β＝Kα·λc β
Wherein λcLagrange multiplier optimized for rate distortion, K being a background block of consecutive K frames in the time domain, the above formula being K>When 1, the equation can not be taken, namely the code rates before and after optimization can not be completely equal, but the code rate fluctuation can be effectively controlled, namely the Lagrange multiplier of the first frame background block under the condition of encoding the background block based on the asymmetric quantization parameter is obtained as
Figure GDA0003070444840000081
Wherein,
Figure GDA0003070444840000082
a Lagrange multiplier representing a first frame background block in the case of encoding the background block based on the asymmetric quantization parameter;
the quantization parameter is
QP＝4.2005·ln(λ)+13.7122
Will be provided with
Figure GDA0003070444840000083
And λiAnd substituting the quantization parameters into the formula respectively to obtain the quantization parameter of each frame, and coding all background blocks marked by the mask by adopting the corresponding quantization parameters. When the value of K is larger, the ratio of K,
Figure GDA0003070444840000084
and λiThe difference value is larger, and the value difference of the quantization parameter of the first frame and the subsequent frame is larger, which is called as asymmetric quantization parameter coding.
6) And decoding the background block mask and the quantization parameter QP offset value at a decoding end to realize the reconstruction of the background block.

Claims (4)

1. The monitoring video frequency distortion optimization coding method based on background distortion propagation is characterized by comprising the following steps:
1) dividing the coding block into a background block and a foreground block;
2) constructing a background block distortion propagation model;
3) performing global rate distortion optimization based on the background block distortion propagation model to obtain parameters of the global rate distortion optimization;
4) marking a background block by using a mask, and carrying out lossless coding on the mask;
5) encoding the background block based on the asymmetric quantization parameter;
6) decoding the background block mask and the asymmetric quantization parameter offset value at a decoding end to realize the reconstruction of the background block;
performing global rate distortion optimization in the step 3), and obtaining parameters of the global rate distortion optimization specifically comprises the following steps:
31) the distortion and bit consumption of the background block of the ith frame are respectively DiAnd RiWherein i is 1, 2.. K, the global rate distortion optimization target is to minimize the rate distortion cost of the background block of the continuous K frames, and the global rate distortion optimization formula of the background block is
Figure FDA0003070444830000011
Wherein λcLagrange multiplier optimized for rate distortion, D1And R1Respectively representing the distortion and bit consumption of the background block of the first frame;
32) based on the background block distortion propagation model, the global rate distortion optimization formula of the background block is rewritten into
Figure FDA0003070444830000012
Are respectively to R1And RiAnd solving the partial derivative to obtain a Lagrange multiplier which has the following relation:
Figure FDA0003070444830000013
wherein λ1Lagrange multiplier, λ, representing the background block of the first frameiThe Lagrange multiplier of the background block of the ith frame is expressed, and the Lagrange multiplier of the subsequent frame is K times of the Lagrange multiplier of the first frame, namely lambdai＝K·λ1；
The encoding of the background block based on the asymmetric quantization parameter in the step 5) specifically comprises:
based on an R-lambda model, the power function relationship between the bit consumption R and the Lagrange multiplier lambda is as follows:
R＝α·λβ
wherein alpha and beta are R-lambda model parameters, and keeping the total bit consumption of the current background block unchanged to obtain
α·λ1 β+(K-1)α·(K·λ1)β＝Kα·λc β
Wherein λcThe Lagrange multiplier for rate distortion optimization is obtained by K being a continuous K frame background block on a time domain
Figure FDA0003070444830000021
Wherein,
Figure FDA0003070444830000023
a Lagrange multiplier representing a first frame background block in the case of encoding the background block based on the asymmetric quantization parameter;
the quantization parameter is
QP＝4.2005·ln(λ)+13.7122
Will be provided with
Figure FDA0003070444830000024
And λiAnd substituting the quantization parameters into the formula respectively to obtain the quantization parameter of each frame, and coding all background blocks marked by the mask by adopting the corresponding quantization parameters.
2. The monitoring video rate-distortion optimized encoding method based on background distortion propagation as claimed in claim 1, wherein in step 1), the encoding block is divided into a background block and a foreground block, specifically, the background block is classified and detected to obtain a background block with continuous K frames in time domain.
3. The monitoring video rate-distortion optimization coding method based on background distortion propagation according to claim 1, wherein the building of the background block distortion propagation model in the step 2) specifically comprises:
constructing a background block distortion propagation model based on a static background: for the background blocks of K continuous frames in the time domain, the background blocks are squares, and the distortion and bit consumption of the background block of the ith frame are respectively DiAnd RiWherein i is 1, 2i，jA source pixel value representing the ith frame image at coordinate position j,
Figure FDA0003070444830000022
representing the decoded pixel value of the pixel with the coordinate position j of the ith frame image, wherein the source pixel values of the same coordinate position of each frame background block in the encoding process are the same, namely pi，j＝pi-1，j＝...＝p1，j，pi-1，jAnd p1，jRespectively representing a source pixel value with j as the coordinate position of the i-1 frame image and a source pixel value with j as the coordinate position of the 1 st frame image, wherein the coordinate position j of the pixel is 1, 22N is the side length of the background block; coding using SKIP mode, i.e.
Figure FDA0003070444830000031
Figure FDA0003070444830000032
Figure FDA0003070444830000033
And
Figure FDA0003070444830000034
respectively representing the pixel decoded pixel value with the coordinate position of j of the i-1 frame image, the pixel decoded pixel value with the coordinate position of j of the 1 st frame image and the distortion D of the i-th frame background blockiThe following relationships exist:
Figure FDA0003070444830000035
wherein D is1Representing the distortion of the first frame background block.
4. The monitored video rate-distortion optimized encoding method based on background distortion propagation as claimed in claim 1, wherein in step 4), the mask is used to mark the background block and perform lossless encoding, specifically, K frames are used as a period at the encoding end, each period is used to mark the background block by using a mask, and the mask is encoded by using arithmetic encoding.
CN201810458872.0A 2018-05-15 2018-05-15 Monitoring video frequency distortion optimization coding method based on background distortion propagation Active CN108650511B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810458872.0A CN108650511B (en) 2018-05-15 2018-05-15 Monitoring video frequency distortion optimization coding method based on background distortion propagation

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810458872.0A CN108650511B (en) 2018-05-15 2018-05-15 Monitoring video frequency distortion optimization coding method based on background distortion propagation

Publications (2)

Publication Number Publication Date
CN108650511A CN108650511A (en) 2018-10-12
CN108650511B true CN108650511B (en) 2021-09-28

Family

ID=63755392

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810458872.0A Active CN108650511B (en) 2018-05-15 2018-05-15 Monitoring video frequency distortion optimization coding method based on background distortion propagation

Country Status (1)

Country Link
CN (1) CN108650511B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109783475B (en) * 2019-01-23 2022-06-14 福州大学 Method for constructing large-scale database of video distortion effect markers

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102547293A (en) * 2012-02-16 2012-07-04 西南交通大学 Method for coding session video by combining time domain dependence of face region and global rate distortion optimization
CN103002289B (en) * 2013-01-08 2015-06-17 中国电子科技集团公司第三十八研究所 Video constant quality coding device for monitoring application and coding method thereof
CN106331711A (en) * 2016-08-26 2017-01-11 北京工业大学 Dynamic bit rate control method based on network feature and video feature
CN107547902A (en) * 2017-07-19 2018-01-05 南京邮电大学 Towards the adaptive rate distortion optimizing method of monitor video coding

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9781449B2 (en) * 2011-10-06 2017-10-03 Synopsys, Inc. Rate distortion optimization in image and video encoding

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102547293A (en) * 2012-02-16 2012-07-04 西南交通大学 Method for coding session video by combining time domain dependence of face region and global rate distortion optimization
CN103002289B (en) * 2013-01-08 2015-06-17 中国电子科技集团公司第三十八研究所 Video constant quality coding device for monitoring application and coding method thereof
CN106331711A (en) * 2016-08-26 2017-01-11 北京工业大学 Dynamic bit rate control method based on network feature and video feature
CN107547902A (en) * 2017-07-19 2018-01-05 南京邮电大学 Towards the adaptive rate distortion optimizing method of monitor video coding

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
"Rate region analysis in a full-duplex-aided cooperative nonorthogonal multiple-access system";Hongji Huang;《IEEE Access( Volume: 5) 2017》;20170830;全文 *

Also Published As

Publication number Publication date
CN108650511A (en) 2018-10-12

Similar Documents

Publication Publication Date Title
Lu et al. Dvc: An end-to-end deep video compression framework
CN111147867B (en) Multifunctional video coding CU partition rapid decision-making method and storage medium
Gao et al. Recent standard development activities on video coding for machines
CN103546749B (en) Method for optimizing HEVC (high efficiency video coding) residual coding by using residual coefficient distribution features and bayes theorem
CN108989802B (en) HEVC video stream quality estimation method and system by utilizing inter-frame relation
CN102065298B (en) High-performance macroblock coding implementation method
CN106385584B (en) The adaptively sampled coding method of distributed video compressed sensing based on spatial correlation
CN101098473A (en) Picture coding method and apparatus
CN107846589B (en) Image compression method based on local dynamic quantization
CN103581647A (en) Depth map sequence fractal coding method based on motion vectors of color video
CN103501438B (en) A kind of content-adaptive method for compressing image based on principal component analysis
CN102281446B (en) Visual-perception-characteristic-based quantification method in distributed video coding
CN107547902B (en) Adaptive rate distortion optimization method for surveillance video coding
CN100574447C (en) Fast intraframe predicting mode selecting method based on the AVS video coding
CN101835042A (en) Wyner-Ziv video coding system controlled on the basis of non feedback speed rate and method
CN103442228B (en) Code-transferring method and transcoder thereof in from standard H.264/AVC to the fast frame of HEVC standard
CN105554502A (en) Distributed compressed sensing video encoding and decoding method based on foreground-background separation
CN110677654B (en) Quantization parameter cascade method of high-efficiency video coding standard low-delay coding structure
Pan et al. Frame-level Bit Allocation Optimization Based on<? brk?> Video Content Characteristics for HEVC
CN102572428A (en) Side information estimating method oriented to distributed coding and decoding of multimedia sensor network
CN108650511B (en) Monitoring video frequency distortion optimization coding method based on background distortion propagation
CN1194544C (en) Video encoding method based on prediction time and space domain conerent movement vectors
CN110493597A (en) A kind of efficiently perception video encoding optimization method
Zhang et al. Macro-block-level selective background difference coding for surveillance video
CN103167282B (en) A kind of online updating method of image compression dynamic regulation

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
GR01 Patent grant
GR01 Patent grant