[go: up one dir, main page]

CN103561259B - The Web conference visual quality automation appraisal procedure of application-oriented share service - Google Patents

The Web conference visual quality automation appraisal procedure of application-oriented share service Download PDF

Info

Publication number
CN103561259B
CN103561259B CN201310290070.0A CN201310290070A CN103561259B CN 103561259 B CN103561259 B CN 103561259B CN 201310290070 A CN201310290070 A CN 201310290070A CN 103561259 B CN103561259 B CN 103561259B
Authority
CN
China
Prior art keywords
image
web conference
visual quality
picture
quality
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.)
Expired - Fee Related
Application number
CN201310290070.0A
Other languages
Chinese (zh)
Other versions
CN103561259A (en
Inventor
杨小虎
蔡亮
王新宇
鲍凌峰
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.)
Hangzhou Yuan Group Technology Co ltd
Original Assignee
HANGZHOU YUNBEN TECHNOLOGY Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by HANGZHOU YUNBEN TECHNOLOGY Co Ltd filed Critical HANGZHOU YUNBEN TECHNOLOGY Co Ltd
Priority to CN201310290070.0A priority Critical patent/CN103561259B/en
Publication of CN103561259A publication Critical patent/CN103561259A/en
Application granted granted Critical
Publication of CN103561259B publication Critical patent/CN103561259B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Landscapes

  • Compression Or Coding Systems Of Tv Signals (AREA)

Abstract

The invention discloses a kind of Web conference visual quality automation appraisal procedure of application-oriented share service, when the method is by analyzing the multimedia netmeeting collected the image stream data that produces and correlation time information data, can the quality of prediction multimedia conferencing of automation and effect.First the method carries out preliminary treatment to picture stream, then the blocking effect that the Y-PSNR based on healthy and strong principal component analysis on Spatial Dimension respectively between computed image and image produce due to JPG compression, the time serial message difference between the two simultaneously by calculating source image stream and receiving terminal image stream obtains the visual quality evaluation of estimate on time dimension.The finally comprehensive mass loss gone out in each feature calculation, calculates the oeverall quality evaluation of last Web conference.

Description

The Web conference visual quality automation appraisal procedure of application-oriented share service
Technical field
The present invention relates to multi-media image visual quality evaluation areas, particularly relate to a kind of Web conference visual quality automation appraisal procedure for application-oriented share service.
Background technology
Internet era, image and video ubiquitous, but image and video council produce various loss in each stages such as compression, process, transmission, these losses all can have influence on people's impression visually.How correct and effective these losses of assessment become an important problem.General, due to the ultimate recipient of people's image and video, allow the evaluation of going of people's subjectivity be a kind of modal method, but people is often subject to the impact of the factors such as residing environment, mood at that time, cause result not to be often very correct, and this method also waste time and energy very much.Therefore, a lot of researchers proposes the various method for evaluating quality to image and video, and these methods propose different solutions from all angles.Objective automatic visual method for evaluating quality serves important function in a lot, and first, it can monitor that mass change is entertained guests in image or video monitoring system; The second, it can as the standard in an image and processing system for video and algorithm; Three, it can help optimized image and processing system for video and algorithm.
Present most visual quality appraisal procedure mostly all for static images and video.For in the method for evaluating quality of static images, most straightforward procedure is traditional mean square deviation and Y-PSNR etc., these methods are all the pixels based on image, all very simple, easy to understand, also have certain effect, but under many circumstances, these methods well can not reflect the quality between image.Also have a lot of method based on some behaviors of human vision system and the feature extraction appraisal procedure to picture quality, these methods achieve certain effect, but also there are this some problems, calculate all comparatively complicated simultaneously.Also someone proposes a kind of method based on picture structure similitude, achieves good effect.For the quality evaluation of video, because video is relative to still image, add a time dimension, therefore need to consider how to weigh the feature on time dimension, existing a lot of method has all been sampled the method for estimation.
Also there is no now a kind of visual quality appraisal procedure of the Web conference for application-oriented share service.The Web conference of application-oriented share service and video have this very large difference, and the visual quality appraisal procedure therefore for video is not too applicable to this scene.The general mode of Web conference towards sharing application service be a computer or terminal by its Desktop Share to participant, it uses ITUT.128 agreement---multipoint application shares association.General Web conference mainly can share some documents, and then exchanged by document explanation, document mainly contains the kinds such as PPT, WORD, PDF.During meeting, for the computer of share desktop as transmitting terminal, the change of its desktop can be sent to receiving terminal with the form of picture stream by its.Web conference towards sharing application service has following characteristic: when (a) only has the desktop of transmitting terminal to change, new desktop picture just can be sent to recipient by conference applications; B (), in order to reduce the data of Internet Transmission, the region of change is only formed a less picture and sends to receiving terminal by it, allows receiving terminal use this picture to cover the appointed area of current desktop; C the connection of () Web conference is that TCP connects, do not have the loss of wrapping in data transmission procedure, namely image data can not be impaired in transmitting procedure; D the picture great majority produced in () meeting all only have word and block letter, distinguish to some extent with natural image.
Summary of the invention
The object of the invention is to for the deficiencies in the prior art, a kind of Web conference visual quality automation appraisal procedure of application-oriented share service is provided, the method automatically can calculate the visual quality of Web conference, solve human subjective and evaluate various problems in meeting quality, and improve the efficiency of accuracy rate and evaluation, provide cost savings, also this kind of Web conference can be helped to apply the feedback obtaining each meeting situation timely, be conducive to pinpointing the problems fast, dealing with problems.
The present invention is achieved through the following technical solutions:
The data flowchart of one one kinds of application-oriented shared Web conference visual quality automation appraisal procedures, it is characterized in that, the method contains following steps:
(1) picture streaming data that produces in the meeting of various heterogeneous networks condition lower network and temporal information corresponding to these picture stream and image change information is obtained: picture streaming data comprises the source images stream S of transmitting terminal and the image stream P of receiving terminal, and temporal information comprises the time series T of transmitting terminal and the time series T ' of receiving terminal; The form of image change information is (left, top, width, height), represents the coordinate of region of variation and grows tall.
(2) carry out preliminary treatment to picture stream: conversion picture stream being carried out to color space, image is converted into YUV color space from RGB, after conversion, source images S becomes S ', and receiving terminal image P becomes P '.
(3) each width picture lasting time is calculated by the time point of the sequence of pictures of record.
(4) Y-PSNR based on healthy and strong principal component analysis in picture stream between every two width correspondence image is calculated.
(5) blocking effect that in picture stream, image produces due to JPG compression is calculated.
(6) according to the time serial message difference between the two of source image stream and receiving terminal image stream, visual quality evaluation of estimate on time dimension is obtained.
(7) mass loss comprehensively gone out in each feature calculation, calculates the oeverall quality evaluation of last Web conference.
The invention has the beneficial effects as follows, now also not to the Web conference visual quality automation appraisal procedure towards Application share, the present invention is according to the visual quality appraisal procedure of existing picture and video, the monopolizing characteristic of the Web conference that faying face is served to Application share, achieves and a kind ofly automation can assess the method for its visual quality.The data for the treatment of and analysis of the present invention mainly come from the picture streaming data produced in Web conference, corresponding temporal information and picture change information data.The present invention does not need the participation of people, avoids uncertainty and the inaccuracy of the subjective assessment of people, thus accomplishes that the result assessed is more accurate.The assessment of automation also greatly improves in assess effectiveness in addition, decreases this respect institute's time spent and cost; Also can feed back possibility produced problem in the application of this kind of Web conference simultaneously in time, help staff to deal with problems, improve the satisfaction of user.
Accompanying drawing explanation
Fig. 1 is flow chart of the present invention;
Fig. 2 is the design sketch based on healthy and strong principal component analysis;
Fig. 3 is experimental result based on healthy and strong principal component analysis and comparison diagram;
Fig. 4 is image stream time series schematic diagram;
Fig. 5 is time series characterization experiments result figure;
Fig. 6 is that the present invention is to sample evaluation result figure.
Embodiment
The present invention is by obtaining the picture streaming data and information correlation time picture change information that produce in application-oriented shared Web conference, various methods in the assessment of application visual quality, calculate the feature situation that each is relevant to quality, finally comprehensively draw an assessed value to Web conference visual quality.
The present invention can be applied in actual product, also can provide quality assurance more reliably in the development& testing stage of product.The present invention can the critic network meeting visual quality of automation, improves the accuracy of assessment, and also can reflect in time may Problems existing, helps staff to repair problem in time.
Specific works process of the present invention is:
1, the picture streaming data that produces in the meeting of various heterogeneous networks condition lower network and temporal information corresponding to these picture stream and image change information is obtained.
2, preliminary treatment is carried out to picture stream.
3, each width picture lasting time is calculated by the time point of the sequence of pictures of record.
4, the Y-PSNR based on healthy and strong principal component analysis in picture stream between every two width correspondence image is calculated.
5, the blocking effect that in picture stream, image produces due to JPG compression is calculated.
6, according to the time serial message difference between the two of source image stream and receiving terminal image stream, visual quality evaluation of estimate on time dimension is obtained.
7, the mass loss comprehensively gone out in each feature calculation, calculates the oeverall quality evaluation of last Web conference.
The invention has the beneficial effects as follows, now also not to the Web conference visual quality automation appraisal procedure towards Application share, the present invention is according to the visual quality appraisal procedure of existing picture and video, the monopolizing characteristic of the Web conference that faying face is served to Application share, achieves and a kind ofly automation can assess the method for its visual quality.The data for the treatment of and analysis of the present invention mainly come from the picture streaming data produced in Web conference, corresponding temporal information and picture change information data.The present invention does not need the participation of people, avoids uncertainty and the inaccuracy of the subjective assessment of people, thus accomplishes that the result assessed is more accurate.The assessment of automation also greatly improves in assess effectiveness in addition, decreases this respect institute's time spent and cost; Also can feed back possibility produced problem in the application of this kind of Web conference simultaneously in time, help staff to deal with problems, improve the satisfaction of user.
Below in conjunction with accompanying drawing and example, the invention will be described further:
In order to introduce application-oriented shared Web conference visual quality automation appraisal procedure, the characteristic of application-oriented share service first to be understood.The Web conference of application-oriented share service is based on the suggestion 128 (ITU-T.128) of international telecommunication union telecommunication's Standardization Sector, and the name of this agreement is called that multipoint application is shared.During network multimedia conference based on this agreement, the presider of meeting can by the Desktop Share of oneself to participant, then majority of case can say by documents such as PPT, WORD, PDF the content that will pass on, and the image therefore produced in meeting is all the sectional drawing of these documents mostly.The view data produced in meeting comes according to the situation of change of presider's screen, if the such as PPT page turning of the content change of screen, it passes to participant with regard to producing new view data; If but screen does not change for a long time, so it can not produce any view data; If screen upper part has changed, it only may send the view data of this part change in the past, and allow application override this part region of piece image, these are key characters for application-oriented shared Web conference.What have the greatest impact to quality in Web conference should be network condition, the parameters such as network delay, shake, packet loss weigh the important indicator of network condition, this kind of Web conference uses TCP to connect realization mostly in addition, loss is not had in data transmission procedure, so network condition affect the time of key reaction when image stream is transmitted, can cause the delay of image and the change of image persistence, this can affect the effect of meeting, causes visual mass loss.In addition before transmission, generally view data can be carried out the data volume compressed when reducing transmission, compression method mainly contains JPG, JPG2000, ZIP etc., and our method mainly considers the impact that the compression method JPG damaged produces visual quality.
Give the data flowchart of application-oriented shared Web conference visual quality automation appraisal procedure with reference to figure 1, contain following steps:
1, the picture streaming data that produces in the meeting of various heterogeneous networks condition lower network and temporal information corresponding to these picture stream and image change information is obtained.
Initial input of the present invention comprises the data flow picture produced in Web conference, and as shown in Figure 1, comprise the source images stream S of transmitting terminal and the image stream P of receiving terminal, these image sizes are identical, are generally the screen size of transmitting terminal computer or terminal.The data inputted in addition also comprise temporal information corresponding to image stream, and the T in Fig. 1 is the time series of transmitting terminal, the time series that T ' is receiving terminal; In addition also have the information of image change, form is (left, top, width, height), represents the coordinate of region of variation respectively and grows tall.
2, preliminary treatment is carried out to picture stream
Preliminary treatment mainly comprises the conversion of color space, and image is converted into YUV color space from RGB, and shown in Fig. 1, after conversion, source images S becomes S ', and receiving terminal image P becomes P '.
3, each width picture lasting time is calculated by the time point of the sequence of pictures of record.
4, the Y-PSNR based on healthy and strong principal component analysis in picture stream between every two width correspondence image is calculated.
For the image produced in Web conference, mainly to the screenshotss of PPT, WORD, PDF document, generally all contain a large amount of words, and its background is mostly very dull, little on the visual sensory impact of people, therefore the present invention has proposed out the outstanding part of image by healthy and strong principal component analysis, the calculating of this subparticipation Y-PSNR, the better like this visual quality reflecting image.Such as in most PPT document, word is above the part mainly paid close attention to of people and its background is generally all ignored by people, by healthy and strong composition analysis, can obtain the ledge in these PPT pictures.
Y-PSNR method based on healthy and strong principal component analysis finally obtains a numerical result, if quality is good, value is general between 30-40, if poor quality, value may 20 even below.In order to better assessment result, this result is done normalization by following formula by the present invention:
q = 1 1 + e a + b × rpnsr
Wherein q is the mass value after normalization, and numeral is between 0-1, and rpnsr is the value of the Y-PSNR based on healthy and strong composition analysis, a and b is preset parameter, is tentatively set to 8 and-0.25 by experiment.
The validity of this method, can the result explanation of Fig. 3.By producing the picture of a series of corrupted data under different JPG contractive conditions to the first width picture in Fig. 2, wherein the mass figures of JPG is from 5 to 100, is spaced apart 5.Then compare with other two kinds of different images method for evaluating quality.Two kinds of methods are respectively common Y-PSNR and structuring similarity method, and the former is a kind of the most frequently used method weighing picture quality; The latter is that a kind of existing image quality measure method is named as SSIM, and its effect has been proved to be in a lot of experiment and researcher.From the result of Fig. 3, we can see that the discrimination of SSIM here quality is not very large, may be relevant with the content of picture, and the picture in Web conference is mostly relatively simple for structure, all comprises a large amount of word; Y-PSNR based on healthy and strong principal component analysis has obvious discrimination, presents the characteristic of linear correlation from JPG quality 5-100; And common Y-PSNR method discrimination is not obvious especially.From the result of Fig. 3, we can find out that the method that the present invention takes achieves good effect.
5, the blocking effect that in picture stream, image produces due to JPG compression is calculated
Before image transmitting, general needs carries out image the size of data compressing to reduce transmission, thus improves efficiency of transmission.But compression may bring the loss of data of picture, JPG compression is more common a kind of compression method.JPG compression is based on discrete cosine transform, and image is generally divided into the fritter of 8*8 by it, is DCT to each block.The most information of image concentrates on low frequency part, so JPG compression can lose a part of high-frequency information, thus causes the loss of data.The feature that JPG compression produces loss generally shows as blocking effect, and mainly because its piecemeal is, DCT causes.
In order to weigh the blocking effect of image, the present invention takes following computational methods: establish the pixel value that x (m, n) is image, then d h(m, n)=x (m+1, n)-x (m, n) is the direct difference of adjacent horizontal pixels, and so the formula of the blocking effect of calculated level is as follows:
Wherein B hfor the value of horizontal fast effect, M, N are the size of image.In like manner, then overall blocking effect is both averages to vertical account form: source images and receiving terminal image all can calculate by these computational methods, and the blocking effect value of last image is that the blocking effect of source images is divided by receiving terminal blocking effect value.Fig. 4 is the situation of change of picture blocking effect after different JPG quality compression.As can be seen from this result we, method of the present invention can reflect preferably picture through JPG compression after blocking effect.
6, according to the time serial message difference between the two of source image stream and receiving terminal image stream, visual quality evaluation of estimate on time dimension is obtained.
Due in network transmission process, network various factors comprises delay, shake, packet loss, the transmission all affecting data such as out of order.Towards in the Web conference of sharing application, connect owing to employing TCP, so its main manifestations is the change delaying arrival and image persistence of view data.Fig. 5 is the schematic diagram of temporal sequence of images, when Web conference carries out, transmitting terminal produces some row pictures, and it is short that each width picture duration has length to have, picture is through network, after receiving terminal is received, due to the impact of network various factors, generally its time received can be delayed a bit, but this impact is little, but the duration of picture also can change, in Fig. 5, the duration of the i-th width picture is Δ t ibut the picture i duration that receiving terminal receives is Δ t' i, the duration is elongated; Also likely the duration shortens, as i-th in Fig. 5+2 width picture.The change of picture duration is very large on the impact of Web conference, causes the asynchronous of two ends, has an impact to the meeting visual experience of people.In order to calculate the visual transmission of this respect, present invention employs following computational methods: the time series of receiving terminal and transmitting terminal is regarded two vectorial Δ t and Δ t', the similarity degree that then calculating two is vectorial.The calculating of similarity degree, we adopt the Euclidean distance between vector, and COS distance and coefficient correlation are weighed, wherein cosine and coefficient correlation range formula as follows:
By under heterogeneous networks condition, use the data collected to verify, result as shown in Figure 5.Be provided with different values to network delay and network jitter in experiment, the numerical value of delay and jitter is all 0,100,200 and 300 milliseconds, Fig. 5 is the experimental result of Euclidean distance in the present invention, and show generally to postpone larger from figure, the quality of assessment is poorer; Shake is also larger, and quality is poorer, and this and experience at ordinary times meet, and namely network condition is poorer, and the quality of Web conference is poorer.This illustrates that evaluation result effectiveness comparison of the present invention is desirable.
7, the mass loss comprehensively gone out in each feature calculation, calculates the oeverall quality evaluation of last Web conference.
Step has above calculated based on the Y-PSNR of healthy and strong principal component analysis, blocking effect and time serial message feature respectively, wherein the above two are all the method for evaluating quality belonging to still image, two groups of related datas can be obtained, but also need to consider on the whole to draw final quality.The present invention considers that the time that in picture stream, each width picture continues is different, it is also different to the quality contribution of entirety, the picture of one width duration length is large on the picture that the impact of total quality is obviously very short than a width duration, needs to each width picture different weight when therefore calculating total quality.First, for the quality of two kinds of still images, consider the Y-PSNR based on healthy and strong principal component analysis and blocking effect, have
Q i = q i 1 + q i 2 2
Wherein Q ibe the quality that the i-th width image is corresponding, q i1represent the quality based on healthy and strong principal component analysis, q i2represent the value of blocking effect.When then considering total quality, first calculate the ratio tp that piece image accounts for the overall time i=Δ t i/ TotalTime, therefore the visual quality of this respect Web conference has:
Q=∑tp i×Q i
The end value finally assessed the visual quality of primary network meeting is the average of time series quality evaluation value and upper face amount.Fig. 6 is the assessment result of the present invention to some application-oriented shared Web conference samples, and these samples obtain under different parameters, and these 3 parameters are: network delay, network jitter, JPG compression quality.Sample in the result of Fig. 6 when JPG compression quality is 80 general all than JPG compression quality be 35 assessment result good; Network condition becomes positive correlation with assessment result, and namely network condition is poorer, and assessment result is poorer, illustrates that the present invention couple assesses with the visual quality towards the Web conference shared and achieves good effect.

Claims (4)

1. a Web conference visual quality automation appraisal procedure for application-oriented share service, it is characterized in that, the method contains following steps:
(1) picture streaming data that produces in the meeting of various heterogeneous networks condition lower network and temporal information corresponding to these picture stream and image change information is obtained: picture streaming data comprises the source images stream S of transmitting terminal and the image stream P of receiving terminal, and temporal information comprises the time series T of transmitting terminal and the time series T ' of receiving terminal; The form of image change information is (left, top, width, height), represents the coordinate of region of variation and grows tall;
(2) carry out preliminary treatment to picture stream: conversion picture stream being carried out to color space, image is converted into YUV color space from RGB, after conversion, source images S becomes S ', and receiving terminal image P becomes P ';
(3) each width picture lasting time is calculated by the time point of the sequence of pictures of record;
(4) Y-PSNR based on healthy and strong principal component analysis in picture stream between every two width correspondence image is calculated;
(5) blocking effect that in picture stream, image produces due to JPG compression is calculated;
(6) according to the time serial message difference between the two of source image stream and receiving terminal image stream, visual quality evaluation of estimate on time dimension is obtained;
(7) mass loss comprehensively gone out in each feature calculation, calculates the oeverall quality evaluation of last Web conference;
Described step (7) is specially: first, for the quality of two kinds of still images, considers the Y-PSNR based on healthy and strong principal component analysis and blocking effect, has
Q i = q i 1 + q i 2 2
Wherein Q ibe the quality that the i-th width image is corresponding, q i1represent the quality based on healthy and strong principal component analysis, q i2represent the value of blocking effect; When then considering total quality, first calculate the ratio tp that piece image accounts for the overall time i=Δ t i/ TotalTime, therefore, the visual quality value of Web conference is:
Q=∑tp i×Q i
The end value finally assessed the visual quality of primary network meeting is the mean value of the visual quality value of visual quality evaluation of estimate and Web conference on time dimension.
2. the Web conference visual quality automation appraisal procedure of application-oriented share service according to claim 1, it is characterized in that, described step (4) is specially: the Y-PSNR method based on healthy and strong principal component analysis finally obtains a numerical result, then this result is done normalization by following formula:
q = 1 1 + e a + b × r p n s r
Wherein, q is the mass value after normalization, and numerical value is between 0-1, and rpnsr is the value of the Y-PSNR based on healthy and strong principal component analysis, a and b is preset parameter.
3. the Web conference visual quality automation appraisal procedure of application-oriented share service according to claim 1, it is characterized in that, described step (5) is specially: establish the pixel value that x (m, n) is image, then d h(m, n)=x (m+1, n)-x (m, n) is the direct difference of adjacent horizontal pixels, and so the formula of the blocking effect of calculated level is as follows:
Wherein B hfor the value of horizontal block effect, M, N are the size of image; In like manner, then overall blocking effect is both averages to vertical account form:
4. the Web conference visual quality automation appraisal procedure of application-oriented share service according to claim 1, it is characterized in that, described step (6) is specially: the time series of receiving terminal and transmitting terminal is regarded two vectorial Δ t and Δ t', then by the Euclidean distance between vector, COS distance and the vectorial similarity degree of Calculation of correlation factor two; Wherein, cosine and coefficient correlation range formula as follows:
CN201310290070.0A 2013-07-10 2013-07-10 The Web conference visual quality automation appraisal procedure of application-oriented share service Expired - Fee Related CN103561259B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201310290070.0A CN103561259B (en) 2013-07-10 2013-07-10 The Web conference visual quality automation appraisal procedure of application-oriented share service

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201310290070.0A CN103561259B (en) 2013-07-10 2013-07-10 The Web conference visual quality automation appraisal procedure of application-oriented share service

Publications (2)

Publication Number Publication Date
CN103561259A CN103561259A (en) 2014-02-05
CN103561259B true CN103561259B (en) 2016-01-27

Family

ID=50015396

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201310290070.0A Expired - Fee Related CN103561259B (en) 2013-07-10 2013-07-10 The Web conference visual quality automation appraisal procedure of application-oriented share service

Country Status (1)

Country Link
CN (1) CN103561259B (en)

Families Citing this family (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106657989A (en) * 2016-12-30 2017-05-10 西南技术物理研究所 Image transmission data link dynamic real-time image peak signal to noise ratio detection device
CN107395692B (en) * 2017-07-04 2019-09-27 广州云测信息技术有限公司 Terminal screen methods of exhibiting and device
CN107454387A (en) * 2017-08-28 2017-12-08 西安万像电子科技有限公司 Mass parameter acquisition methods and device for image coding and decoding Transmission system
CN111459807A (en) * 2020-03-19 2020-07-28 西安万像电子科技有限公司 Image transmission detection method and device
CN114615491B (en) * 2020-12-04 2025-05-06 中国移动通信集团北京有限公司 A video conference network perception evaluation method, system, electronic device and medium
CN113115347A (en) * 2021-05-10 2021-07-13 游密科技(深圳)有限公司 Network conference visual quality automatic evaluation method oriented to application sharing service
CN113709453B (en) * 2021-09-13 2023-09-08 北京车和家信息技术有限公司 Video quality assessment method, device, equipment and medium
CN115334053B (en) * 2022-08-03 2023-07-18 深圳乐播科技有限公司 Method for realizing associated screen projection in cloud conference and related products
CN118075449A (en) * 2024-04-18 2024-05-24 智广海联(天津)大数据技术有限公司 A method and device for evaluating camera surveillance video quality

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101615290A (en) * 2009-07-29 2009-12-30 西安交通大学 A face image super-resolution reconstruction method based on canonical correlation analysis

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR100669251B1 (en) * 2005-11-25 2007-01-16 한국전자통신연구원 Digital image quality automatic analysis device and method

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101615290A (en) * 2009-07-29 2009-12-30 西安交通大学 A face image super-resolution reconstruction method based on canonical correlation analysis

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
一种新型的基于全参考的视频质量评价方法;田魁等;《现代电信科技》;20120421;正文第41页左栏第5行-第45页第2行,图1 *

Also Published As

Publication number Publication date
CN103561259A (en) 2014-02-05

Similar Documents

Publication Publication Date Title
CN103561259B (en) The Web conference visual quality automation appraisal procedure of application-oriented share service
Narvekar et al. A no-reference image blur metric based on the cumulative probability of blur detection (CPBD)
EP1924101B1 (en) Video communication quality estimation device, method, and program
CN106815839B (en) A Method for Blind Evaluation of Image Quality
CN105678702B (en) A kind of the human face image sequence generation method and device of feature based tracking
CN102655585B (en) Video conference system and time delay testing method, device and system thereof
CN108846800A (en) A kind of non-reference picture quality appraisement method of image super-resolution rebuilding
CN117896552B (en) Video conference processing method, video conference system and related device
Zhang et al. A new no-reference quality metric for JPEG2000 images
CN102984540A (en) Video quality assessment method estimated on basis of macroblock domain distortion degree
CN103297801A (en) No-reference video quality evaluation method aiming at video conference
CN104506968A (en) Method and device for determining video abstract figure
WO2014191462A1 (en) Method and apparatus for predicting the perceptual quality of a signal
Choi et al. Video QoE models for the compute continuum
CN112866683B (en) Quality assessment method based on video preprocessing and transcoding
CN103077514B (en) A kind of visual perception image quality evaluation method based on full variation
CN105491320B (en) Videoconference bandwidth management method and system
CN107465914A (en) Method for evaluating video quality based on Local textural feature and global brightness
CN109889816B (en) Video quality evaluation method, device, equipment and storage medium
Gao et al. A content-based image quality metric
CN105959685A (en) A Compression Bit Rate Prediction Method Based on Video Content and Cluster Analysis
CN103856742A (en) Video and audio information processing method, device and system
CN103345625B (en) A kind of three-dimensional image analysis method and system
Alvarez et al. A flexible QoE framework for video streaming services
CN100518333C (en) Method and device for correcting gamma characteristics of video communication

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
C41 Transfer of patent application or patent right or utility model
TR01 Transfer of patent right

Effective date of registration: 20170120

Address after: 310007 room B, building, Xixi Road, Hangzhou, Zhejiang, 525, China

Patentee after: Hangzhou yuan Group Technology Co.,Ltd.

Address before: 310007 room C, building, Xixi Road, Hangzhou, Zhejiang, 525, China

Patentee before: HANGZHOU YUNBEN TECHNOLOGY Co.,Ltd.

CF01 Termination of patent right due to non-payment of annual fee
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20160127