CN112291564B - HEVC intra-frame code rate control method for optimizing and monitoring video perception quality - Google Patents

HEVC intra-frame code rate control method for optimizing and monitoring video perception quality Download PDF

Info

Publication number
CN112291564B
CN112291564B CN202011305568.6A CN202011305568A CN112291564B CN 112291564 B CN112291564 B CN 112291564B CN 202011305568 A CN202011305568 A CN 202011305568A CN 112291564 B CN112291564 B CN 112291564B
Authority
CN
China
Prior art keywords
frame
video
lcu
mth
determining
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
CN202011305568.6A
Other languages
Chinese (zh)
Other versions
CN112291564A (en
Inventor
公衍超
王玲
刘颖
杨楷芳
王富平
林庆帆
高梓铭
韦同胜
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.)
Xian University of Posts and Telecommunications
Original Assignee
Xian 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 Xian University of Posts and Telecommunications filed Critical Xian University of Posts and Telecommunications
Priority to CN202011305568.6A priority Critical patent/CN112291564B/en
Publication of CN112291564A publication Critical patent/CN112291564A/en
Application granted granted Critical
Publication of CN112291564B publication Critical patent/CN112291564B/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
    • H04N19/172—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 picture, frame or field
    • 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
    • H04N19/70—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals characterised by syntax aspects related to video coding, e.g. related to compression standards
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N7/00—Television systems
    • H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast

Abstract

The HEVC intra-frame rate control method for optimizing perceptual quality of a surveillance video comprises the steps of coding a 1 st frame of the surveillance video, determining an average frame difference of an LCU, determining a motion perception sensitive factor of the LCU, determining an average gradient of the LCU, determining a texture perception sensitive factor of the LCU, determining a target bit of the LCU and coding a subsequent I frame of the surveillance video, distributing more rate codes for a visual sensitive area with medium texture complexity and low motion speed, using relatively less rate codes in an area with flat texture, excessive complexity and high motion speed, solving the technical problems that the perceptual characteristic of a human visual system and the complexity of parameter calculation are not fully considered in the existing rate control technology, and particularly improving the perceptual quality of a reconstructed video under medium and low rate. The method has the advantages of simple parameter calculation, high bit control precision, consistency of the reconstructed video quality and the human visual system perception characteristic and the like, and can be used in the technical field of monitoring video code rate control.

Description

HEVC intra-frame code rate control method for optimizing and monitoring video perception quality
Technical Field
The invention belongs to the technical field of code rate control in video coding, and particularly relates to an HEVC intra-frame code rate control method for optimizing and monitoring video perceptual quality.
Background
With the rapid development of the internet and multimedia technologies, multimedia applications have already penetrated various aspects of our lives. Video, an important component of multimedia, has become an important source of our information. Video monitoring is one of important video applications, and plays an extremely important role in the fields of public security, intelligent home and the like. Video coding is a key technology for ensuring the application of surveillance video in a practical environment. The international telecommunication union telecommunication standardization sector and the international organization for standardization/international electrotechnical commission jointly provide a new generation of video coding standard-HEVC in 2013, and compared with H.264/AVC, the video coding standard saves the coding rate by nearly 50% on the premise of ensuring the unchanged coding quality.
Although the compression efficiency of HEVC is greatly improved, in the existing environment of real-time video transmission, the channel bandwidth is limited. In order to effectively transmit video data and ensure the playing quality of video services under the condition of meeting the limitations of channel bandwidth and transmission delay, code rate control needs to be performed on the video coding process. The code rate control method based on lambda field includes the first target bit for coding unitAllocating, namely allocating proper target bits for the coding units according to the video content, the channel bandwidth and the buffer status; the Quantization Parameter (QP) is then independently determined for the coding unit using an R- λ model to achieve the pre-allocated target bits. In a target bit allocation process considering only intra rate control, the target bit allocation is divided into a picture level and a maximum coding unit (LCU) level according to a hierarchical structure of an encoding process. The picture level is a coding unit of a full picture level in a video sequence, and the LCU level is a coding unit of a size of 64 × 64 pixels, which is further divided into coding units of the picture level. The target bits at the picture level are allocated according to the number of remaining uncoded pictures, coding complexity and the condition of the remaining target bits. The target bits at the LCU level are allocated according to the number of remaining unencoded LCUs in the current coded picture, the coding complexity and the condition of the remaining target bits. When LCU-level bit allocation is carried out on an I frame by a code rate control method based on a lambda field in the current HEVC, a target bit B of each LCU is determined by adopting the following formulam,n：
Figure BDA0002788227980000011
Target bit B 'of LCU as coding advances'm,nThe update is performed as follows:
Figure BDA0002788227980000021
relevant research on human visual characteristics has shown that human eyes are selective in watching video, and the visual focus is attracted by certain specific information. Objects with moderate texture complexity and moderate and slow motion speed often catch human eyes, such as pedestrians in video, vehicles driving on urban roads. In contrast, for flat textured areas such as sky and road surfaces and almost stationary objects, the information content of the flat textured areas is small and is often easily ignored by human eyes. For the objects with complicated textures such as grasslands, trees and the like and high movement speed, due to the influence of texture covering and movement covering, detailed information of the objects is difficult to distinguish by naked eyes, so that distortion is difficult to perceive. Aiming at the perception characteristic of human eyes, limited bit resources can be redistributed, more code rates are distributed to the regions concerned by the vision focus, and relatively less code rates are distributed to other unimportant regions, so that the perception quality of the video can be still ensured on the premise of relatively insufficient code rates.
In the current HEVC code rate control method based on the lambda domain, in the process of code rate allocation, the code rate allocation is mainly performed by considering objective factors such as a coding structure, video sequence content characteristics and the like. Since the final receiving end of the video stream is the human visual system, a reasonable code rate allocation strategy should allocate more code rates to the regions concerned by human eyes and allocate less code rates to the regions not concerned by human eyes. Therefore, the performance of HEVC currently adopting a λ -domain based rate control method without considering the perceptual characteristics of the human visual system can be further improved.
For the HEVC standard, some more effective rate control methods considering the perceptual characteristics of the human visual system are proposed. The general idea of the methods is to firstly divide the regions in the video into different levels according to the importance or the significance, and then allocate more code rates and use smaller QP for coding the more important or significant regions in modules such as code rate control bit allocation or QP determination and the like, so as to preferentially ensure the quality of the regions and improve the coding perception quality of the whole video image. However, the existing methods have one or more of the following problems: first, there are some methods that aim at improving the perceptual quality of video images, but use peak signal-to-noise ratio (PSNR), Mean Square Error (MSE), and other measures that do not consider the human visual perception characteristics as quality measures in model optimization or performance testing. A great deal of literature research shows that the quality measure of the PSNR, MSE and the like without considering the human visual perception characteristics has poor consistency with the actual quality of the human visual system perception characteristics. The perception quality improved by the code rate control method is limited; second, there are some methods that use complex operations in the calculation and updating of model parameters, such as precoding, motion estimation, various mathematical optimization model iterations, etc. Video monitoring is a video application with high real-time requirement, so a method with complex model parameter calculation and updating is not suitable for the video monitoring application due to high coding delay; third, some methods are mainly proposed for inter-frame images (P-frames or B-frames) in HEVC, and these methods are not suitable for intra-frame images because there is a significant difference between intra-frame images and inter-frame images in terms of the adopted coding technique and the coding rate distortion performance; a fourth, some code rate control methods based on neural network are also proposed, but these methods have very high requirements on system hardware, and are not suitable for being used at a monitoring camera in a video monitoring system. In addition, the coding time delay of the method is very large, and the method is not suitable for being applied to a video monitoring system.
Disclosure of Invention
The technical problem to be solved by the present invention is to overcome the shortcomings of the prior art, and provide an HEVC intra-frame rate control method for optimizing and monitoring video perceptual quality, which is suitable for a video monitoring system, considers human visual system perceptual characteristics, and has low coding delay.
The technical scheme adopted for solving the technical problems comprises the following steps:
(1) coding surveillance video frame 1
And (4) giving a target code rate, and coding the 1 st frame of the monitoring video according to the HEVC code rate control method based on the lambda field.
(2) Determining average frame difference for LCU
Average frame difference D of nth LCU in mth I frame of videom,nDetermined according to equation (1):
Figure BDA0002788227980000031
wherein m is more than 1, m and n are finite positive integers, w and h are the width and height of the nth LCU in the mth I frame of the video respectively, and xm,nAnd (I, j) is the brightness component value of the nth LCU in the mth I frame of the video at the pixel point (I, j).
(3) Determining motion perception sensitivity factors for LCUs
Mth videoMotion perception sensitivity factor d of nth LCU in I framem,nDetermined according to equation (2):
Figure BDA0002788227980000032
wherein p is1、q1、p2、q2Is a model parameter, p1∈(0,1]，q1∈[0.5,1]，p2∈[-0.02,0)，q2∈[2,5]。
(4) Determining average gradient of LCU
Average gradient T of nth LCU in mth I frame of videom,nDetermined according to equation (3):
Figure BDA0002788227980000033
(5) determining texture perception sensitivity factors for LCUs
Texture perception sensitivity factor t of nth LCU in mth I frame of videom,nDetermined according to equation (4):
Figure BDA0002788227980000034
wherein a is1、a2、a3、a4、a5Is a model parameter, a1∈[-2×10-5,-1×10-5]，a2∈[0.001,0.002]，a3∈[-0.1,-0.05]，a4∈[0.5,2]，a5∈[-0.5,0)。
(6) Determining perceptual sensitivity factors for LCUs
Perceptual sensitivity factor theta of nth LCU in mth I frame of videom,nDetermined according to equation (5):
θm,n＝μ×dm,n+(1-μ)×tm,n (5)
μ＝b1×Dm,n+b2 (6)
wherein μ, b1、b2Is a model parameter, b1∈[-0.001,0)，b2∈[0.3,1]。
(7) Determining target bits for LCUs
Figure BDA0002788227980000041
Where round () is a rounding function, BmIs the target bit number of the mth I frame of the video, LN is the number of LCUs in the mth I frame of the video, omegam,nThe weight of the nth LCU in the mth I frame of the video is determined according to the code rate control method based on the lambda field of HEVC. Target bit B 'of LCU as coding advances'm,nUpdating is performed according to equation (8):
Figure BDA0002788227980000042
wherein R isres,tarIs the target number of bits remaining in the mth I frame of the video, Am,kAnd sw represents the size of a sliding window for the number of coding bits actually output by the kth LCU in the mth I frame of the video.
(8) Encoding subsequent I-frames of surveillance video
From obtained B'm,nAnd performing code rate control on the nth LCU in the mth I frame of the monitoring video to complete the coding process of the nth LCU in the mth I frame.
In the step (3) of determining motion perception sensitivity factor of LCU of the present invention, said p1The optimum value is 0.2984, q1Most preferably 0.6182, p2The value is preferably-0.0187, q2The optimum value is 4.7163.
In the step (5) of determining the texture perception sensitivity factor of the LCU, a is1The optimum value is-1.4749 × 10-5，a2The optimum value is 0.0017, a3The optimal value is-0.0702, a4The optimum value is 0.9954, a5The most preferred value is-0.2183.
Determining perceptual sensitivity factors of LCU in the present inventionIn step (6), b is1Most preferably-0.0004, b2And most preferably 0.85.
Because the invention adopts the motion perception sensitive factor for determining the LCU and the texture perception sensitive factor for determining the LCU, considers the content characteristic of the monitoring video and the perception characteristic of the human visual system, adds the perception sensitive factors into the bit distribution weight of the LCU level, distributes more code rates for the visual sensitive areas with medium texture complexity, proper motion speed and low motion speed, uses relatively less code rates in areas with flat texture, excessive complexity and high motion speed, well represents the relation between the visual perception quality and the coding code rate, solves the technical problems that the perception characteristic of the human visual system and the parameter calculation are not fully considered in the prior code rate control technology, and particularly can improve the perception quality of the reconstructed video under medium and low code rates. The method has the advantages of simple parameter calculation, high bit control precision, consistency of the reconstructed video quality and the human visual system perception characteristic and the like, and can be used in the technical field of monitoring video code rate control.
Drawings
FIG. 1 is a flowchart of example 1 of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the drawings and examples, but the present invention is not limited to the examples.
Example 1
In fig. 1, the HEVC intra rate control method for optimizing perceptual quality of a surveillance video in this embodiment includes the following steps:
(1) coding surveillance video frame 1
And (4) giving a target code rate, and coding the 1 st frame of the monitoring video according to the HEVC code rate control method based on the lambda field.
(2) Determining average frame difference for LCU
Average frame difference D of nth LCU in mth I frame of videom,nDetermined according to equation (1):
Figure BDA0002788227980000051
wherein m is more than 1, m and n are finite positive integers, w and h are the width and height of the nth LCU in the mth I frame of the video respectively, and xm,nAnd (I, j) is the brightness component value of the nth LCU in the mth I frame of the video at the pixel point (I, j).
(3) Determining motion perception sensitivity factors for LCUs
Motion perception sensitivity factor d of nth LCU in mth I frame of videom,nDetermined according to equation (2):
Figure BDA0002788227980000052
wherein p is1、q1、p2、q2Is a model parameter, p1∈(0,1]P of the present embodiment1Value 0.2984, q1∈[0.5,1]Q of the present embodiment1Value 0.6182, p2E [ -0.02, 0)), p of the present embodiment2The value is-0.0187, q2∈[2,5]Q of the present embodiment2The value is 4.7163.
The step considers the content characteristics of the monitoring video and the time domain covering effect, and solves the technical problems that the perception characteristics of the human visual system are not fully considered and the parameter calculation is complex in the existing code rate control technology.
(4) Determining average gradient of LCU
Average gradient T of nth LCU in mth I frame of videom,nDetermined according to equation (3):
Figure BDA0002788227980000061
(5) determining texture perception sensitivity factors for LCUs
Texture perception sensitivity factor t of nth LCU in mth I frame of videom,nDetermined according to equation (4):
Figure BDA0002788227980000062
wherein a is1、a2、a3、a4、a5Is a model parameter, a1∈[-2×10-5,-1×10-5]A of the present embodiment1The value is-1.4749 x 10-5，a2∈[0.001,0.002]A of the present embodiment2A value of 0.0017, a3∈[-0.1,-0.05]A of the present embodiment3The value is-0.0702, a4∈[0.5,2]A of the present embodiment4Value 0.9954, a5E [ -0.5,0), a of the present embodiment5The value is-0.2183.
The steps adopt the characteristics of monitoring video content and the spatial masking effect, and solve the technical problems that the existing code rate control technology does not fully consider the perception characteristics of a human visual system and the parameter calculation is complex.
(6) Determining perceptual sensitivity factors for LCUs
Perceptual sensitivity factor theta of nth LCU in mth I frame of videom,nDetermined according to equation (5):
θm,n＝μ×dm,n+(1-μ)×tm,n (5)
μ＝b1×Dm,n+b2 (6)
wherein μ, b1、b2Is a model parameter, b1E [ -0.001, 0)), b of the present embodiment1The value is-0.0004, b2∈[0.3,1]B of the present embodiment2The value is 0.85.
(7) Determining target bits for LCUs
Figure BDA0002788227980000063
Where round () is a rounding function, BmIs the target bit number of the mth I frame of the video, LN is the number of LCUs in the mth I frame of the video, omegam,nThe weight of the nth LCU in the mth I frame of the video is determined according to the code rate control method based on the lambda field of HEVC. As the encoding progresses, the target bits of the LCUB′m,nUpdating is performed according to equation (8):
Figure BDA0002788227980000071
wherein R isres,tarIs the target number of bits remaining in the mth I frame of the video, Am,kAnd sw represents the size of a sliding window for the number of coding bits actually output by the kth LCU in the mth I frame of the video.
The perception sensitivity factors are added into LCU-level bit distribution weights, more code rates are distributed for the visual sensitive areas with medium texture complexity, proper movement speed and low movement speed, relatively few code rates are used in areas with flat texture, excessive complex texture and high movement speed, the relation between the visual perception quality and the coding code rate is represented, the technical problems that the perception characteristics of a human visual system and the calculation of parameters are not fully considered in the existing code rate control technology are solved, and especially the perception quality of a reconstructed video can be improved under medium and low code rates.
(8) Encoding subsequent I-frames of surveillance video
From obtained B'm,nAnd performing code rate control on the nth LCU in the mth I frame of the monitoring video to complete the coding process of the nth LCU in the mth I frame.
And completing the HEVC intra-frame code rate control method for optimizing the perceived quality of the monitoring video.
Example 2
The HEVC intra rate control method for optimizing the perceptual quality of a surveillance video of the present embodiment comprises the following steps:
(1) coding surveillance video frame 1
This procedure is the same as in example 1.
(2) Determining average frame difference for LCU
This procedure is the same as in example 1.
(3) Determining motion perception sensitivity factors for LCUs
Motion perception sensitivity factor d of nth LCU in mth I frame of videom,nDetermined according to equation (2):
Figure BDA0002788227980000072
wherein p is1、q1、p2、q2Is a model parameter, p1∈(0,1]P of the present embodiment1Value of 0.1, q1∈[0.5,1]Q of the present embodiment1A value of 0.5, p2E [ -0.02, 0)), p of the present embodiment2The value is-0.02, q2∈[2,5]Q of the present embodiment2The value is 2.
(4) Determining average gradient of LCU
This procedure is the same as in example 1.
(5) Determining texture perception sensitivity factors for LCUs
Texture perception sensitivity factor t of nth LCU in mth I frame of videom,nDetermined according to equation (4):
Figure BDA0002788227980000081
wherein a is1、a2、a3、a4、a5Is a model parameter, a1∈[-2×10-5,-1×10-5]A of the present embodiment1The value is-2 x 10-5，a2∈[0.001,0.002]A of the present embodiment2A value of 0.001, a3∈[-0.1,-0.05]A of the present embodiment3The value is-0.1, a4∈[0.5,2]A of the present embodiment4A value of 0.5, a5E [ -0.5,0), a of the present embodiment5The value is-0.5.
(6) Determining perceptual sensitivity factors for LCUs
Perceptual sensitivity factor theta of nth LCU in mth I frame of videom,nDetermined according to equation (5):
θm,n＝μ×dm,n+(1-μ)×tm,n (5)
μ＝b1×Dm,n+b2 (6)
wherein μ, b1、b2Is a model parameter, b1E [ -0.001, 0)), b of the present embodiment1The value is-0.001, b2∈[0.3,1]B of the present embodiment2The value is 0.3.
The other steps are the same as in example 1.
And completing the HEVC intra-frame code rate control method for optimizing the perceived quality of the monitoring video.
Example 3
The HEVC intra rate control method for optimizing the perceptual quality of a surveillance video of the present embodiment comprises the following steps:
(1) coding surveillance video frame 1
This procedure is the same as in example 1.
(2) Determining average frame difference for LCU
This procedure is the same as in example 1.
(3) Determining motion perception sensitivity factors for LCUs
Motion perception sensitivity factor d of nth LCU in mth I frame of videom,nDetermined according to equation (2):
Figure BDA0002788227980000091
wherein p is1、q1、p2、q2Is a model parameter, p1∈(0,1]P of the present embodiment1Value of 1, q1∈[0.5,1]Q of the present embodiment1Value of 1, p2E [ -0.02, 0)), p of the present embodiment2The value is-0.001, q2∈[2,5]Q of the present embodiment2The value is 5.
(4) Determining average gradient of LCU
This procedure is the same as in example 1.
(5) Determining texture perception sensitivity factors for LCUs
Texture perception sensitivity factor t of nth LCU in mth I frame of videom,nDetermined according to equation (4):
Figure BDA0002788227980000092
wherein a is1、a2、a3、a4、a5Is a model parameter, a1∈[-2×10-5,-1×10-5]A of the present embodiment1The value is-1 x 10-5，a2∈[0.001,0.002]A of the present embodiment2The value is 0.002, a3∈[-0.1,-0.05]A of the present embodiment3The value is-0.05, a4∈[0.5,2]A of the present embodiment4A value of 2, a5E [ -0.5,0), a of the present embodiment5The value is-0.1.
(6) Determining perceptual sensitivity factors for LCUs
Perceptual sensitivity factor theta of nth LCU in mth I frame of videom,nDetermined according to equation (5):
θm,n＝μ×dm,n+(1-μ)×tm,n (5)
μ＝b1×Dm,n+b2 (6)
wherein μ, b1、b2Is a model parameter, b1E [ -0.001, 0)), b of the present embodiment1The value is-0.0001, b2∈[0.3,1]B of the present embodiment2The value is 1.
The other steps are the same as in example 1.
And completing the HEVC intra-frame code rate control method for optimizing the perceived quality of the monitoring video.
To verify the beneficial effects of the present invention, the inventor performed experiments on test videos by using the method of embodiment 1 of the present invention, the experiments are as follows:
selecting 4 videos in the PKU-SVD-A data set as test sequences, wherein the names, the resolutions and the frame rates of the 4 videos are respectively as follows: campus, 720x576, 30 frames/sec; classlover, 720x576, 30 frames/sec; mainload, 1600x1200, 30 frames/sec; interaction, 1600x1200, 30 frames/sec. A comparison test was performed by using the HEVC intra-frame rate control method (hereinafter referred to as the present invention) for optimizing and monitoring video perceptual quality in embodiment 1 of the present invention and the rate control method in the encoder HM16.0 (hereinafter referred to as the comparison method 1) for which the high definition video coding standard is formulated and organized to recommend. Bit error BE is used for measuring bit control precision corresponding to the two methods, and the bit error BE is calculated according to the following formula:
Figure BDA0002788227980000101
wherein B isactIs the actual bitrate of the coded video, BtarIs the target bitrate for coding the video. The smaller the BE, the more accurate the bit control is represented. The perceptual quality of the reconstructed video is measured by the average subjective score MOS. The larger the MOS, the higher the perceptual quality of the reconstructed video.
Using a test model HM16.0 recommended by the international standard organization, adopting a full intra-frame coding structure, namely encoder _ intra _ main.cfg, in the configuration file, under a given target code rate, starting the code rate control coding of each test sequence according to the method of the invention and the comparison method 1, and obtaining a coding result, as shown in table 1.
Table 1 inventive and comparative method 1 coding results
Figure BDA0002788227980000102
As can BE seen from Table 1, for all the tested sequences, the average BE of comparative method 1 was 0.0074%, the average MOS was 2.61, the average BE of the inventive method was 0.0039%, and the average MOS was 2.93. Compared with the comparison method 1, the method has higher bit control precision, provides the perception quality more in line with the vision of human eyes and has higher coding performance.

Claims (4)

1. An HEVC intra-frame rate control method for optimizing and monitoring video perception quality is characterized by comprising the following steps of:
(1) coding surveillance video frame 1
A target code rate is given, and the 1 st frame of the monitoring video is coded according to an HEVC code rate control method based on a lambda field;
(2) determining average frame difference for LCU
Average frame difference D of nth LCU in mth I frame of videom,nDetermined according to equation (1):
Figure FDA0002788227970000011
wherein m is more than 1, m and n are finite positive integers, w and h are the width and height of the nth LCU in the mth I frame of the video respectively, and xm,n(I, j) is the brightness component value of the nth LCU in the mth I frame of the video at the pixel point (I, j);
(3) determining motion perception sensitivity factors for LCUs
Motion perception sensitivity factor d of nth LCU in mth I frame of videom,nDetermined according to equation (2):
Figure FDA0002788227970000012
wherein p is1、q1、p2、q2Is a model parameter, p1∈(0,1]，q1∈[0.5,1]，p2∈[-0.02,0)，q2∈[2,5]；
(4) Determining average gradient of LCU
Average gradient T of nth LCU in mth I frame of videom,nDetermined according to equation (3):
Figure FDA0002788227970000013
(5) determining texture perception sensitivity factors for LCUs
Texture perception sensitivity factor t of nth LCU in mth I frame of videom,nDetermined according to equation (4):
Figure FDA0002788227970000014
wherein a is1、a2、a3、a4、a5Is a model parameter, a1∈[-2×10-5,-1×10-5]，a2∈[0.001,0.002]，a3∈[-0.1,-0.05]，a4∈[0.5,2]，a5∈[-0.5,0)；
(6) Determining perceptual sensitivity factors for LCUs
Perceptual sensitivity factor theta of nth LCU in mth I frame of videom,nDetermined according to equation (5):
θm,n＝μ×dm,n+(1-μ)×tm,n (5)
μ＝b1×Dm,n+b2 (6)
wherein μ, b1、b2Is a model parameter, b1∈[-0.001,0)，b2∈[0.3,1]；
(7) Determining target bits for LCUs
Target bit B of nth LCU in mth I frame of videom,nDetermined according to equation (7):
Figure FDA0002788227970000021
where round () is a rounding function, BmIs the target bit number of the mth I frame of the video, LN is the number of LCUs in the mth I frame of the video, omegam,nThe weight of the nth LCU in the mth I frame of the video is determined according to the code rate control method of HEVC based on the lambda field; target bit B 'of LCU as coding advances'm,nUpdating is performed according to equation (8):
Figure FDA0002788227970000022
wherein R isres,tarFor the target bit remained in the mth I frame of the videoNumber, Am,kThe actual output coding bit number of a kth LCU in an mth I frame of the video is shown as sw, and the sw represents the size of a sliding window;
(8) encoding subsequent I-frames of surveillance video
From obtained B'm,nAnd performing code rate control on the nth LCU in the mth I frame of the monitoring video to complete the coding process of the nth LCU in the mth I frame.
2. The HEVC intra-frame rate control method for optimizing and monitoring video perceptual quality as recited in claim 1, wherein: in the step (3) of determining the motion perception sensitivity factor of the LCU, p is1Value 0.2984, q1Value 0.6182, p2The value is-0.0187, q2The value is 4.7163.
3. The HEVC intra-frame rate control method for optimizing and monitoring video perceptual quality as recited in claim 1, wherein: in the step (5) of determining the texture perception sensitivity factor of the LCU, a is1The value is-1.4749 x 10-5，a2A value of 0.0017, a3The value is-0.0702, a4Value 0.9954, a5The value is-0.2183.
4. The HEVC intra-frame rate control method for optimizing and monitoring video perceptual quality as recited in claim 1, wherein: in the step (6) of determining the perception sensitivity factor of the LCU, b is1The value is-0.0004, b2The value is 0.85.
CN202011305568.6A 2020-11-20 2020-11-20 HEVC intra-frame code rate control method for optimizing and monitoring video perception quality Active CN112291564B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011305568.6A CN112291564B (en) 2020-11-20 2020-11-20 HEVC intra-frame code rate control method for optimizing and monitoring video perception quality

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011305568.6A CN112291564B (en) 2020-11-20 2020-11-20 HEVC intra-frame code rate control method for optimizing and monitoring video perception quality

Publications (2)

Publication Number Publication Date
CN112291564A CN112291564A (en) 2021-01-29
CN112291564B true CN112291564B (en) 2021-09-14

Family

ID=74399309

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011305568.6A Active CN112291564B (en) 2020-11-20 2020-11-20 HEVC intra-frame code rate control method for optimizing and monitoring video perception quality

Country Status (1)

Country Link
CN (1) CN112291564B (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114666585A (en) * 2022-02-23 2022-06-24 翱捷科技股份有限公司 Code rate control method and device based on visual perception
CN114866773B (en) * 2022-05-09 2023-02-28 西安邮电大学 Quantization parameter cascade method for eliminating time domain jitter effect
CN115361571A (en) * 2022-08-04 2022-11-18 武汉依迅北斗时空技术股份有限公司 Playing method and device of cloud storage video data

Citations (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1685734A (en) * 2002-09-05 2005-10-19 新加坡科技研究局 Method and apparatus for controlling rate of video sequence, video encoding device
CN103918271A (en) * 2011-06-01 2014-07-09 王舟 Method and system for structural similarity based perceptual video coding
CN104079925A (en) * 2014-07-03 2014-10-01 中国传媒大学 Ultrahigh definition video image quality objective evaluation method based on visual perception characteristic
CN104602016A (en) * 2015-01-12 2015-05-06 宁波大学 HEVC video information hiding method based on intra-frame prediction mode difference
CN105959707A (en) * 2016-03-14 2016-09-21 合肥工业大学 Static state background video compression algorithm based on motion perception
CN106303521A (en) * 2016-08-15 2017-01-04 华侨大学 A kind of HEVC Rate-distortion optimization method based on sensitivity of awareness
CN106303530A (en) * 2016-10-20 2017-01-04 北京工业大学 A kind of bit rate control method merging vision perception characteristic
CN106331707A (en) * 2016-08-30 2017-01-11 上海大学 Asymmetric perceptual video coding system and method based on just noticeable distortion model
CN107197235A (en) * 2017-06-26 2017-09-22 杭州当虹科技有限公司 A kind of HDR video pre-filterings method
CN107197266A (en) * 2017-06-26 2017-09-22 杭州当虹科技有限公司 A kind of HDR method for video coding
CN108111857A (en) * 2017-12-29 2018-06-01 广州海昇计算机科技有限公司 A kind of target bit allocation method, system and device for the control of H.265 code check
CN108810530A (en) * 2018-07-12 2018-11-13 珠海亿智电子科技有限公司 A kind of AVC bit rate control methods based on human visual system
CN109451316A (en) * 2018-11-21 2019-03-08 北京航空航天大学 A kind of QP selection algorithm based on CU conspicuousness
CN109982082A (en) * 2019-05-05 2019-07-05 山东大学深圳研究院 A kind of more distortion criterion Rate-distortion optimization methods of HEVC based on local grain characteristic
CN110166796A (en) * 2018-05-16 2019-08-23 腾讯科技（深圳）有限公司 Processing method, device, computer-readable medium and the electronic equipment of video frame
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

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11178407B2 (en) * 2012-11-19 2021-11-16 Texas Instruments Incorporated Adaptive coding unit (CU) partitioning based on image statistics

Patent Citations (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1685734A (en) * 2002-09-05 2005-10-19 新加坡科技研究局 Method and apparatus for controlling rate of video sequence, video encoding device
CN103918271A (en) * 2011-06-01 2014-07-09 王舟 Method and system for structural similarity based perceptual video coding
CN104079925A (en) * 2014-07-03 2014-10-01 中国传媒大学 Ultrahigh definition video image quality objective evaluation method based on visual perception characteristic
CN104602016A (en) * 2015-01-12 2015-05-06 宁波大学 HEVC video information hiding method based on intra-frame prediction mode difference
CN105959707A (en) * 2016-03-14 2016-09-21 合肥工业大学 Static state background video compression algorithm based on motion perception
CN106303521A (en) * 2016-08-15 2017-01-04 华侨大学 A kind of HEVC Rate-distortion optimization method based on sensitivity of awareness
CN106331707A (en) * 2016-08-30 2017-01-11 上海大学 Asymmetric perceptual video coding system and method based on just noticeable distortion model
CN106303530A (en) * 2016-10-20 2017-01-04 北京工业大学 A kind of bit rate control method merging vision perception characteristic
CN107197235A (en) * 2017-06-26 2017-09-22 杭州当虹科技有限公司 A kind of HDR video pre-filterings method
CN107197266A (en) * 2017-06-26 2017-09-22 杭州当虹科技有限公司 A kind of HDR method for video coding
CN108111857A (en) * 2017-12-29 2018-06-01 广州海昇计算机科技有限公司 A kind of target bit allocation method, system and device for the control of H.265 code check
CN110166796A (en) * 2018-05-16 2019-08-23 腾讯科技（深圳）有限公司 Processing method, device, computer-readable medium and the electronic equipment of video frame
CN108810530A (en) * 2018-07-12 2018-11-13 珠海亿智电子科技有限公司 A kind of AVC bit rate control methods based on human visual system
CN109451316A (en) * 2018-11-21 2019-03-08 北京航空航天大学 A kind of QP selection algorithm based on CU conspicuousness
CN109982082A (en) * 2019-05-05 2019-07-05 山东大学深圳研究院 A kind of more distortion criterion Rate-distortion optimization methods of HEVC based on local grain characteristic
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

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
Fast HEVC intra coding algorithm based on Otsu"s method and gradient;Xuanjing Wang;《2016 IEEE International Symposium on Broadband Multimedia Systems and Broadcasting (BMSB)》;20160728;全文 *
Surveillance Source Compression with Background Modeling for Video Big Data;Ling Tian;《2016 IEEE International Conferences on Big Data and Cloud Computing (BDCloud), Social Computing and Networking (SocialCom), Sustainable Computing and Communications (SustainCom) (BDCloud-SocialCom-SustainCom)》;20161031;全文 *
分级预测视频编码感知质量优化研究;公衍超;《中国优秀博硕士学位论文全文数据库(博士)信息科技辑》;20190115;全文 *
基于HEVC的视频编码快速算法研究;荆倩;《中国优秀硕士学位论文全文数据库 信息科技辑》;20170915;全文 *

Also Published As

Publication number Publication date
CN112291564A (en) 2021-01-29

Similar Documents

Publication Publication Date Title
CN112291564B (en) HEVC intra-frame code rate control method for optimizing and monitoring video perception quality
CN1098599C (en) System for coding video signal in presence of image intensity gradient
CN103179394B (en) A kind of based on area video quality stable I frame bit rate control method
US20090067493A1 (en) Method and apparatus for controlling bitrate of scalable video stream
CN108989802B (en) HEVC video stream quality estimation method and system by utilizing inter-frame relation
CN108924554B (en) Panoramic video coding rate distortion optimization method based on spherical weighting structure similarity
CN104378636B (en) A kind of video encoding method and device
CN101710993A (en) Block-based self-adaptive super-resolution video processing method and system
CN104994382B (en) A kind of optimization method of perception rate distortion
CN110677654B (en) Quantization parameter cascade method of high-efficiency video coding standard low-delay coding structure
Yuan et al. Visual JND: A perceptual measurement in video coding
CN108200431B (en) Bit allocation method for video coding code rate control frame layer
CN104994387B (en) A kind of bit rate control method of fused images feature
CN107343202B (en) Feedback-free distributed video coding and decoding method based on additional code rate
Afsana et al. Efficient scalable uhd/360-video coding by exploiting common information with cuboid-based partitioning
CN110493597B (en) Efficient perceptual video coding optimization method
CN107592535B (en) H.265/HEVC image layer code rate control method
Cen et al. Joint decoding of independently encoded compressive multi-view video streams
CN112040231B (en) Video coding method based on perceptual noise channel model
CN103313064B (en) Based on the time domain error hidden method of inter-frame mode and movement repair
Wang et al. Quality assessment for MPEG-2 video streams using a neural network model
CN113259662B (en) Code rate control method based on three-dimensional wavelet video coding
CN114173131A (en) Video compression method and system based on inter-frame correlation
CN111464805B (en) Three-dimensional panoramic video rapid coding method based on panoramic saliency
CN114339241A (en) Video code rate control method

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