[go: up one dir, main page]

CN102006472A - Video bitrate control system and method thereof - Google Patents

Video bitrate control system and method thereof Download PDF

Info

Publication number
CN102006472A
CN102006472A CN 201010549578 CN201010549578A CN102006472A CN 102006472 A CN102006472 A CN 102006472A CN 201010549578 CN201010549578 CN 201010549578 CN 201010549578 A CN201010549578 A CN 201010549578A CN 102006472 A CN102006472 A CN 102006472A
Authority
CN
China
Prior art keywords
image
quantization step
rate
encoder complexity
code check
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
CN 201010549578
Other languages
Chinese (zh)
Inventor
高飞
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.)
Wuxi Vimicro Corp
Original Assignee
Wuxi Vimicro Corp
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 Wuxi Vimicro Corp filed Critical Wuxi Vimicro Corp
Priority to CN 201010549578 priority Critical patent/CN102006472A/en
Publication of CN102006472A publication Critical patent/CN102006472A/en
Pending legal-status Critical Current

Links

Images

Landscapes

  • Compression Or Coding Systems Of Tv Signals (AREA)

Abstract

The invention provides a video bitrate control system and method thereof. The method comprises the following steps of: first estimating coding complexity by combining image complexity with position information and motion information detected and tracked by a motion target, then establishing a model of bitrate to the coding complexity and quantification step size, fast and accurately estimating the bitrate of image coding by utilizing the information, and therefore realizing accurate bitrate control.

Description

Video code rate control system and method thereof
[technical field]
The present invention relates to field of video monitoring, Rate Control system and method thereof during video coding compresses during particularly about video transmission.
[background technology]
In the video transmission process, it during video transmission the picture that video image is resolved into a frame one frame, for the transmission quantity that reduces data need carry out compressed encoding to video data usually, video coding is relevant with the content of each frame of video image, because the content of each frame picture has difference, so the data volume of the frame that compressed can change, the data volume (being the code check of video transmission) in the unit interval just has fluctuation like this.In actual applications, when especially video flowing transmits, need constant code check, so need control to the code check of video transmission.
Code rate control method mainly is according to the code check of the present frame prediction code check of frame down in the existing video compression, i.e. target bit rate, and adjust the target bit rate of next frame as required.The principle of Rate Control is to set up the Mathematical Modeling of code check and quantization step, after determining target bit rate, goes out quantization step by this calculated with mathematical model, thereby quantization step is adjusted the control that realizes target bit rate.
In the supervisory control system of fixing camera head because so background is fixed the foreground area that code check mainly contains motion produces.And how prospect is not rigid objects, and (estimation) prediction result is inaccurate more in its momental big more feasible coding, need the residual error of coding many more, thereby code check can be big more.The big more then representation unit time data quantity transmitted of code check is also big more, is unfavorable for saving bandwidth.
Therefore, how the code check of image encoding is estimated fast and accurately, thereby realized that Rate Control is the problem that needs solve accurately.
[summary of the invention]
The object of the present invention is to provide and a kind ofly can estimate the code check of image encoding fast and accurately, thereby realize accurate code rate control method.
Another object of the present invention is to provide a kind of can estimate the code check of image encoding fast and accurately, thereby realizes the control system of accurate Rate Control.
For reaching aforementioned purpose, the method for a kind of video code rate control of the present invention: it comprises:
Moving target in the image is carried out detection and tracking, obtain the position and the motion vector of moving target;
Image is carried out edge statistics obtain image edge information;
According to motion vector and image edge information estimation encoder complexity;
Calculate quantization step according to encoder complexity and target bit rate;
Control code check by the adjustment quantization step and reach target bit rate.
Further, described image is added up obtains image edge information and comprises:
Utilize edge detection operator to carry out rim detection;
Edge image is carried out binary conversion treatment;
Marginal point in the binary image added up obtain image edge information E.
Further, the aforementioned estimation encoder complexity be utilize Mathematical Modeling C=|MV (X, Y) | * E calculates, and wherein (X Y) is the motion vector of the aforementioned moving target that obtains to MV, and E is aforementioned image edge information.
Further, calculate quantization step according to encoder complexity and target bit rate, comprise set up code check to the model Rate=f of encoder complexity and quantization step (C, QP), perhaps QP=g (Rate, C), wherein QP is a quantization step, Rate is a code check.
For reaching aforementioned another purpose, a kind of Rate Control of the present invention system, it comprises:
Moving object detection and tracking cell are used for the moving target of image is detected, and determine the position and the motion vector of moving target;
The image border statistic unit is used for the marginal point output image marginal information of statistical picture;
Encoder complexity estimation unit, the image edge information that obtains according to the motion vector of moving object detection and the detected moving target of tracking cell and image border statistic unit utilize the publish picture encoder complexity of picture of calculated with mathematical model;
The quantization step computing unit, the numerical value that image encoding complexity that provides according to the encoder complexity estimation unit and target bit rate calculate quantization step;
The code check adjustment unit is used for according to the quantization step that the quantization step computing unit calculates code check being adjusted control.
Further, described image border statistic unit utilizes edge detection operator to carry out that rim detection is carried out binary conversion treatment to edge image then and marginal point in the binary image is added up and obtains image edge information E.
Further, described encoder complexity estimation unit be utilize Mathematical Modeling C=|MV (X, Y) | * E calculates encoder complexity C, and wherein (X Y) is the motion vector of the aforementioned moving target that obtains to MV, and E is aforementioned image edge information.
Further, described quantization step computing unit comprises the Mathematical Modeling Rate=f (C of code check to encoder complexity and quantization step, QP), perhaps QP=g (Rate, C), wherein QP is a quantization step, and Rate is a code check, and described quantization step computing unit utilizes this calculated with mathematical model quantization step according to encoder complexity and target bit rate.
Rate Control system of the present invention and method thereof are carried out detection and tracking to the moving target in the image, obtain the position and the motion vector of moving target; Then image is carried out edge statistics and obtain image edge information; Afterwards according to motion vector and image edge information estimated coding complexity; Calculate quantization step according to encoder complexity and target bit rate again; Control code check by the adjustment quantization step then and reach target bit rate.Method of the present invention utilizes intelligent monitoring information to estimate the code check of image encoding fast and accurately, thereby realizes Rate Control more accurately.
[description of drawings]
Fig. 1 is the schematic flow sheet of bit rate control method of the present invention.
Fig. 2 is the schematic flow sheet of statistical picture marginal information in the bit rate control method of the present invention.
Fig. 3 is the structured flowchart of Rate Control of the present invention system.
[embodiment]
Alleged herein " embodiment " or " embodiment " are meant special characteristic, structure or the characteristic that can be contained at least one implementation of the present invention.Different in this manual local " in one embodiment " that occur not are all to refer to same embodiment, neither be independent or optionally mutually exclusive with other embodiment embodiment.In addition, represent the sequence of unit in method, flow chart or the functional block diagram of one or more embodiment and revocablely refer to any particular order, also be not construed as limiting the invention.
Because the background in the image is fixed, when carrying out encoding compression, code check is mainly produced by the foreground area of motion in the supervisory control system of fixing camera head.And how the object in the image of foreground area is not rigid objects, and prediction result is inaccurate more in its momental big more feasible coding, need the residual error of coding many more, thereby code check can be big more.And the edge of image statistical information has characterized the complexity of this area image, many more its high fdrequency component that comprises of zone in corresponding edge is many more, the efficient of the laggard every trade journey coding of discrete cosine transform in cataloged procedure (DCT) change quantization is low more, also can produce more code stream.
For overcoming of the influence of this two big factor to Rate Control, positional information that combining image complexity of the present invention and moving object detection are followed the tracks of and movable information be the estimation encoder complexity earlier, moving target in the image is carried out detection and tracking, obtain the position and the motion vector of moving target; Then image is carried out edge statistics and obtain image edge information; Afterwards according to motion vector and image edge information estimated coding complexity; Calculate quantization step according to encoder complexity and target bit rate again; Control code check by the adjustment quantization step then and reach target bit rate.Utilize this information fast and accurately the code check of image encoding to be estimated, thereby realize Rate Control accurately.
See also shown in Figure 1ly, it shows the flow chart that carries out Rate Control of one embodiment of the invention.Be to being that the code check of unit is controlled with the frame in the present embodiment.As shown in Figure 1, bit rate control method of the present invention comprises the steps:
Step S11: at first the picture in the present frame of video image is detected, obtain the position of the also definite moving target of moving target in the picture, and moving target is carried out motion tracking, obtain the motion vector of moving target.The motion vector of moving target is the displacement of moving target in former frame and back one two field picture, can use MV (x, y) expression.Can adopt the method for existing various motion detection about the technology of motion detection in the video image, for example frame-to-frame differences point-score, background subtracting method, statistical average method, median filter method or the like are enumerated detailed description no longer one by one about concrete principle and this specification of content of the method for motion detection.
Step S12: the marginal information of statistical picture, see also shown in Figure 2ly, it specifically comprises:
Step S121: utilize edge detection operator to carry out rim detection.Classical rim detection is based on original image, and gray scale step in each its certain field of pixel investigation of image is changed, and utilizes contiguous single order in edge or second order directional derivative Changing Pattern to detect the edge.Edge detection method commonly used has: difference rim detection, gradient rim detection, Roberts edge detection operator, Sobel edge detection operator, Prewitt edge detection operator, Laplace edge detection operator etc.About this specification of concrete grammar of rim detection also detailed explanation no longer one by one.
Step S122: edge image is carried out binary conversion treatment.Promptly set a threshold value T, with the data separated into two parts of T: greater than the pixel group of T with less than the pixel group of T with image.Can set in one embodiment of the invention such as edge image to greater than the pixel group of T, non-edge image then is the pixel group less than T.
Step S123: marginal point in the binary image is added up, formed matrix, thereby obtain image edge information E.
Continue to consult shown in Figure 1, step S13: in conjunction with aforementioned movement vector MV (x, y) and the marginal information E of entire frame set up the Mathematical Modeling estimation encoder complexity of encoder complexity.Concrete Mathematical Modeling can use C=|MV (X, Y) | * E represents that wherein C is an encoder complexity, and (x y) is the motion vector of moving target to MV, and E is an image edge information.
Continue to consult shown in Figure 1, step S14: statistics is set up the model of code check to encoder complexity C and quantization step QP.Concrete Mathematical Modeling can be expressed as Rate=f (C, QP), perhaps QP=g (Rate, C).
Continue to consult shown in Figure 1, step S15: the relationship model according to code check and the encoder complexity and the quantization step of front, calculate quantization step in conjunction with the target bit rate of desiring to reach, make it reach the target bit rate of estimating by adjusting quantization step may command code check.
Be to be the Rate Control of unit in aforementioned embodiments with whole frame, in with macro block or other Rate Control less than frame unit, the marginal information and the motion vector that can adopt preceding method to obtain this unit calculate its encoder complexity, determine code check according to different encoder complexities, and then the quantization step of definite this unit, control code check by adjustment to quantization step.
See also shown in Figure 3, the structured flowchart of its real Rate Control of the present invention system 1.As shown in the figure, Rate Control of the present invention system 1 comprises moving object detection and tracking cell 11, image border statistic unit 12, encoder complexity estimation unit 13, quantization step computing unit 14 and code check adjustment unit 15.
Wherein moving object detection and tracking cell 11 are used for the moving target of image is detected, determine the position of moving target and motion vector MV (x, y).
Image border statistic unit 12 is used for the marginal point of statistical picture and sets up the matrix of image border point, output image marginal information E.Wherein image border statistic unit 12 at first utilizes edge detection operator to carry out rim detection when carrying out the image edge information statistics, then edge image is carried out binary conversion treatment, marginal point in the binary image is added up obtaining image edge information E again.
The image edge information that encoder complexity estimation unit 13 obtains according to the motion vector of moving object detection and tracking cell 11 detected moving targets and image border statistic unit 12 utilize Mathematical Modeling C=|MV (X, Y) | * E calculates the encoder complexity C of image.
Quantization step computing unit 14, its inside comprise the Mathematical Modeling of a code check to encoder complexity and quantization step, for example be Rate=f (C, QP), perhaps QP=g (Rate, C).The numerical value that image encoding complexity C that quantization step computing unit 14 provides according to encoder complexity estimation unit 13 and target bit rate calculate quantization step.
Code check adjustment unit 15 is used for according to the quantization step that quantization step computing unit 14 calculates code check being adjusted control.
Above-mentioned explanation has fully disclosed the specific embodiment of the present invention.It is pointed out that and be familiar with the scope that any change that the person skilled in art does the specific embodiment of the present invention does not all break away from claims of the present invention.Correspondingly, the scope of claim of the present invention also is not limited only to previous embodiment.

Claims (8)

1. the method for video code rate control: it comprises:
Moving target in the image is carried out detection and tracking, obtain the position and the motion vector of moving target;
Image is carried out edge statistics obtain image edge information;
According to motion vector and image edge information estimated coding complexity;
Calculate quantization step according to encoder complexity and target bit rate;
Control code check by the adjustment quantization step and reach target bit rate.
2. the method for video code rate as claimed in claim 1 control is characterized in that: describedly image is carried out edge statistics obtain image edge information and comprise:
Utilize edge detection operator to carry out rim detection;
Edge image is carried out binary conversion treatment;
Marginal point in the binary image added up obtain image edge information E.
3. video code rate control method as claimed in claim 1 or 2, it is characterized in that: described estimated coding complexity be utilize Mathematical Modeling C=|MV (X, Y) | * E calculates, wherein MV (X, Y) be the motion vector of the aforementioned moving target that obtains, E is aforementioned image edge information.
4. video code rate control method as claimed in claim 3, it is characterized in that: calculate quantization step according to encoder complexity and target bit rate, comprise and set up the model Rate=f (C of code check encoder complexity and quantization step, QP), perhaps QP=g (Rate, C), wherein QP is a quantization step, and Rate is a code check.
5. video code rate control system, it comprises:
Moving object detection and tracking cell are used for the moving target of image is detected, and determine the position and the motion vector of moving target;
The image border statistic unit is used for the marginal point output image marginal information of statistical picture;
Encoder complexity estimation unit, the image edge information that obtains according to the motion vector of moving object detection and the detected moving target of tracking cell and image border statistic unit utilize the publish picture encoder complexity of picture of calculated with mathematical model;
The quantization step computing unit, the numerical value that image encoding complexity that provides according to the encoder complexity estimation unit and target bit rate calculate quantization step;
The code check adjustment unit is used for according to the quantization step that the quantization step computing unit calculates code check being adjusted control.
6. video code rate control system as claimed in claim 5 is characterized in that: described image border statistic unit utilizes edge detection operator to carry out that rim detection is carried out binary conversion treatment to edge image then and marginal point in the binary image is added up and obtains image edge information E.
7. as claim 5 or 6 described video code rate control system, it is characterized in that: described encoder complexity estimation unit is to utilize Mathematical Modeling C=|MV (X, Y) | * E calculates encoder complexity C, MV (X wherein, Y) be the motion vector of the aforementioned moving target that obtains, E is aforementioned image edge information.
8. video code rate control system as claimed in claim 7, it is characterized in that: described quantization step computing unit comprises the Mathematical Modeling Rate=f (C of code check to encoder complexity and quantization step, QP), perhaps QP=g (Rate, C), wherein QP is a quantization step, and Rate is a code check, and described quantization step computing unit utilizes this calculated with mathematical model quantization step according to encoder complexity and target bit rate.
CN 201010549578 2010-11-18 2010-11-18 Video bitrate control system and method thereof Pending CN102006472A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN 201010549578 CN102006472A (en) 2010-11-18 2010-11-18 Video bitrate control system and method thereof

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN 201010549578 CN102006472A (en) 2010-11-18 2010-11-18 Video bitrate control system and method thereof

Publications (1)

Publication Number Publication Date
CN102006472A true CN102006472A (en) 2011-04-06

Family

ID=43813489

Family Applications (1)

Application Number Title Priority Date Filing Date
CN 201010549578 Pending CN102006472A (en) 2010-11-18 2010-11-18 Video bitrate control system and method thereof

Country Status (1)

Country Link
CN (1) CN102006472A (en)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104038769A (en) * 2014-06-11 2014-09-10 同济大学 Rate control method for intra-frame coding
CN107113429A (en) * 2014-10-03 2017-08-29 微软技术许可有限责任公司 The adaptive quantizing in interest region
CN110022463A (en) * 2019-04-11 2019-07-16 重庆紫光华山智安科技有限公司 Video interested region intelligent coding method and system are realized under dynamic scene
WO2020243906A1 (en) * 2019-06-04 2020-12-10 SZ DJI Technology Co., Ltd. Method, device, and storage medium for encoding video data base on regions of interests
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

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1480463A2 (en) * 2003-05-23 2004-11-24 LG Electronics Inc. Adaptive moving picture coding method based on block classification
US20060274833A1 (en) * 2005-06-02 2006-12-07 Broadcom Corporation Text recognition during video compression
CN101039432A (en) * 2006-03-16 2007-09-19 华为技术有限公司 Method and apparatus for realizing self-adaptive quantization in coding process
CN101184221A (en) * 2007-12-06 2008-05-21 上海大学 Video Coding Method Based on Visual Attention
CN101466035A (en) * 2007-12-21 2009-06-24 深圳市融合视讯科技有限公司 Method for distributing video image set bit based on H.264
CN101719979A (en) * 2009-11-27 2010-06-02 北京航空航天大学 Video object segmentation method based on time domain fixed-interval memory compensation

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1480463A2 (en) * 2003-05-23 2004-11-24 LG Electronics Inc. Adaptive moving picture coding method based on block classification
US20060274833A1 (en) * 2005-06-02 2006-12-07 Broadcom Corporation Text recognition during video compression
CN101039432A (en) * 2006-03-16 2007-09-19 华为技术有限公司 Method and apparatus for realizing self-adaptive quantization in coding process
CN101184221A (en) * 2007-12-06 2008-05-21 上海大学 Video Coding Method Based on Visual Attention
CN101466035A (en) * 2007-12-21 2009-06-24 深圳市融合视讯科技有限公司 Method for distributing video image set bit based on H.264
CN101719979A (en) * 2009-11-27 2010-06-02 北京航空航天大学 Video object segmentation method based on time domain fixed-interval memory compensation

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
《Circuits and Systems,2006》 20061207 Shen Li Complexity Based Fast Coding Mode Decision for MPEG-2/H.264 Video Transcoding 第575页左栏 1-8 , 2 *

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104038769A (en) * 2014-06-11 2014-09-10 同济大学 Rate control method for intra-frame coding
CN104038769B (en) * 2014-06-11 2017-05-03 同济大学 Rate control method for intra-frame coding
CN107113429A (en) * 2014-10-03 2017-08-29 微软技术许可有限责任公司 The adaptive quantizing in interest region
CN110022463A (en) * 2019-04-11 2019-07-16 重庆紫光华山智安科技有限公司 Video interested region intelligent coding method and system are realized under dynamic scene
WO2020243906A1 (en) * 2019-06-04 2020-12-10 SZ DJI Technology Co., Ltd. Method, device, and storage medium for encoding video data base on regions of interests
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
CN113438488B (en) * 2021-06-18 2023-01-31 北京锐马视讯科技有限公司 Low-bit-rate video optimization coding method, device, equipment and storage medium

Similar Documents

Publication Publication Date Title
CN102263955B (en) Method for detecting video occlusion based on motion vectors
CN101321287B (en) Video encoding method based on movement object detection
JP6016332B2 (en) Image processing apparatus and image processing method
CN102006472A (en) Video bitrate control system and method thereof
US10863185B2 (en) Systems and methods for hybrid video encoding
CN101252687A (en) Method for Realizing Multi-channel Combined Region of Interest Video Coding and Transmission
Cai et al. A novel video coding strategy in HEVC for object detection
CN101855911A (en) Processing video information
Ko et al. An energy-efficient wireless video sensor node for moving object surveillance
EP2735146A1 (en) Method and apparatus for reframing and encoding a video signal
CN106384359B (en) Motion target tracking method and TV
EP1835763A3 (en) Video coding apparatus
CN101729883B (en) Method for adaptively adjusting video decoding complexity
JP5839848B2 (en) Image processing apparatus and image processing method
US20140369618A1 (en) Encoding device and monitoring system
MX2014007041A (en) Method and apparatus for video quality measurement.
CN104244004A (en) Low-power coding method and low-power coding device
KR101868270B1 (en) Content-aware video encoding method, controller and system based on single-pass consistent quality control
CN114286097A (en) Coding block quantization increment parameter optimization method in secondary video coding rate control
CN104125470A (en) Video data transmission method
EP3827588B1 (en) Spatial layer rate allocation
CN101854547B (en) Motion frame in video collection and transmission system, method and system for detecting prospects
CN103079065B (en) A kind of video data handling procedure and device
CN101854466A (en) Moving area detection method and device
CN103942759A (en) Three-dimensional noise reduction method and device based on Gaussian background model in fixed scene

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C12 Rejection of a patent application after its publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20110406