CN106027291B - QoE evaluation method for BE service based on Weber Fechner theorem - Google Patents
QoE evaluation method for BE service based on Weber Fechner theorem Download PDFInfo
- Publication number
- CN106027291B CN106027291B CN201610320052.6A CN201610320052A CN106027291B CN 106027291 B CN106027291 B CN 106027291B CN 201610320052 A CN201610320052 A CN 201610320052A CN 106027291 B CN106027291 B CN 106027291B
- Authority
- CN
- China
- Prior art keywords
- webpage
- qoe
- visual information
- picture
- information content
- 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
Links
- 238000011156 evaluation Methods 0.000 title claims abstract description 34
- 230000000007 visual effect Effects 0.000 claims abstract description 69
- 238000013210 evaluation model Methods 0.000 claims abstract description 24
- 238000012549 training Methods 0.000 claims abstract description 18
- 239000000284 extract Substances 0.000 claims abstract description 12
- 238000000605 extraction Methods 0.000 claims description 23
- 238000002474 experimental method Methods 0.000 claims description 22
- 239000011159 matrix material Substances 0.000 claims description 8
- 238000010606 normalization Methods 0.000 claims description 6
- 230000009466 transformation Effects 0.000 claims description 3
- 238000000034 method Methods 0.000 abstract description 33
- 230000008901 benefit Effects 0.000 abstract description 5
- 238000004891 communication Methods 0.000 abstract description 4
- 230000000694 effects Effects 0.000 abstract description 4
- 238000012854 evaluation process Methods 0.000 abstract 1
- 238000013441 quality evaluation Methods 0.000 abstract 1
- 238000012360 testing method Methods 0.000 description 6
- 238000005314 correlation function Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000004088 simulation Methods 0.000 description 3
- 230000000638 stimulation Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- YBMRDBCBODYGJE-UHFFFAOYSA-N germanium dioxide Chemical compound O=[Ge]=O YBMRDBCBODYGJE-UHFFFAOYSA-N 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- 241001269238 Data Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000009194 climbing Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000001303 quality assessment method Methods 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 238000000611 regression analysis Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 241000894007 species Species 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/50—Testing arrangements
- H04L43/55—Testing of service level quality, e.g. simulating service usage
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/50—Network service management, e.g. ensuring proper service fulfilment according to agreements
- H04L41/508—Network service management, e.g. ensuring proper service fulfilment according to agreements based on type of value added network service under agreement
- H04L41/509—Network service management, e.g. ensuring proper service fulfilment according to agreements based on type of value added network service under agreement wherein the managed service relates to media content delivery, e.g. audio, video or TV
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Multimedia (AREA)
- Quality & Reliability (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a QoE (quality of service) evaluation method for BE (BE-service) services based on the Weber-Fechner theorem, belonging to the technical field of multimedia service quality evaluation of communication networks. The method comprises the steps of establishing a QoE evaluation model for BE service by utilizing visual information based on WFL, extracting webpage visual information quantity and training parameters, and finally predicting an MOS value. The method and the device respectively extract the visual content information quantity of the main composition element pictures and texts in the webpage by considering the influence of the visual information quantity generated by the webpage content on the QoE of the terminal user, so that the visual content information of the webpage is quantized and is convenient to calculate, the technical effect of accurately evaluating and predicting the QoE of the BE service is achieved, the method and the device have the advantages of simple and rapid evaluation process and accurate evaluation result, and have the significance of optimizing an objective QoE evaluation method, accurately reflecting the QoE of the user and predicting the QoE of the BE service.
Description
Technical field
The invention belongs to communication network multimedia service quality assessment technique field, be related in multimedia service as possible and
For evaluation side user experience quality (QoE, Quality of Experience) of type service (BE, Best-Effort service)
A kind of method, and in particular to psychophysics theorem weber Fechner theorem.
Background technique
BE business be IEEE 802.16 define a kind of basic business class (bibliography [1]: C.Cicconetti,
L.Lenzini,E.Mingozzi,and C.Eklund.“Quality of service support in IEEE
802.16networks. " IEEE Netw., vol.20, no.2, pp.50-55, Mar./Apr.2006), this basic business class
The main business including web browsing, Email browsing, file download and short message transmission these types type.Because of user
It is more likely to selection and is able to satisfy the network operator of high QoE demand, which results in the competition between operator is more and more fierce, respectively
Operator, family is based on oneself existing network capacity, falls over each other to provide the BE business with higher QoE.In order to attract and user bound
Using a kind of specific BE business, the real-time accurate QoE of BE business is assessed and becomes more and more important (ginseng for operator
It examines document [2]: Jeevan Pokhrel, Felipe Lalanne, Ana Cavalli, Wissam Mallouli. " QoE
estimation for web service selection using a Fuzzy-Rough hybrid
expertsystem.”Advanced Information Networking and Applications(AINA),May
2014).Numerous studies at this stage are concentrated mainly in the exploration to the QoE appraisal procedure of video traffic, BE business it is real-time
There is no mature standards for QoE assessment.
Currently, the method for assessment QoE is broadly divided into two classes: subjective evaluation method and objective evaluation method.Subjective evaluation side
Method means that terminal user directly assesses the QoE of business.Although this method is relatively intuitive more accurate, this method
Cost is higher, needs to carry out a large amount of stringent subjective testings, and test the test environment for needing true user and harshness,
It expends considerable time and effort.Compared to subjective evaluation method, objective evaluation method is although inaccurate, but wins in method letter
It is single, it is easily operated.Furthermore, it is possible to which the result using subjective experiment is modified and adjusts to objective evaluation model, so that objective
The accuracy of appraisal procedure increases.Therefore, very multi-disciplinary method is applied in objective evaluation method, for improving
The accuracy of objective evaluation method.Bibliography [3] (F.Lalanne, A.Cavalli, and S.Maag. " Quality of
experience as a selection criterion for web services.”Signal Image Technology
and Internet Based Systems(SITIS),2012Eighth International Conference on,
Pp.519-526,2012) by establishing between Web qos parameter (business deadline, reliability, validity) and QoE
Correlation function assess the QoE of Web service, and calculate the index of correlation function using a regression analysis tool.Although this
Kind method is relatively simple convenient, but the result assessed but shows correlation function and non-optimal and error span is higher.With reference to text
It offers in [2], author uses a kind of QoE that Web service is assessed based on the novel method of Fuzzy and Rough mixed model.This side
Method eliminates inaccurate and ambiguous situation in subjective test using fuzzy theory combining rough set theory, can be to avoid information
It is lost, but the experimental result obtained based on the method is but shown between subjectivity QoE score value and the QoE score value of the model prediction
Difference is not small enough, and prediction result is undesirable.
In recent years, with the development of psychophysics research, more and more researchers start to draw in QoE assessment
Enter the principle weber Fechner theorem (WFL, Weber-Fechner Law) in terms of psychophysics.The theorem describes one
Relationship between the magnitude of physical stimulation and its feeling intensity being perceived.Bibliography [4] (P.Reichl, S.Egger,
R.Schatz,and A.D’Alconzo.“The Logarithmic Nature of QoE and the Role of the
Weber-Fechner Law in QoE Assessment.”IEEE International Conference
Communications (ICC), Cape Town, South Afric, pp.1-5,2010) in, this relationship is employed for the first time
In QoE assessment.But this method only accounts for single stimulus, this defect results in the different of some MOS values
Often, bibliography [5] (Le Thu Nguyen, Richard Harris, Amal Punchihewa. " Assessment of
Quality of Experience for Web Browsing--As Function of Quality of Service and
Content Factors. " Ubiquitous and Future Networks (ICUFN) .July2013) to this abnormal
Explanation is that the influence of how many couples of terminal users of business tine information is not accounted in QoE evaluation.In order to business carry out compared with
For comprehensively and accurately QoE evaluation, need to consider influence of the business tine information to terminal user in QoE evaluation method.Although
Bibliography [5] considers service content information in QoE evaluation method, but not to the accurate definition of content information,
Also without the accurate extraction of progress content information, only very schematically the richness of the information content is divided, therefore information
There is no intuitively, accurately expressed to the influence that QoE is evaluated for content.
Summary of the invention
The present invention is directed to cannot accurately be truly reflected user's sense for BE business QoE evaluation method in current communication system
The case where knowing experience, by considering influence of the visual information amount of web page contents generation to terminal user QoE, to the master in webpage
It wants component picture and text to carry out the extraction of vision content information content respectively, quantifies the vision content information of webpage
And convenient for calculating, the technical effect for carrying out accurate evaluation and prediction to the QoE of BE business is reached, has had evaluation procedure simple
Rapidly, the accurate advantage of evaluation result, and there is optimization QoE method for objectively evaluating, accurate response user QoE and carry out BE industry
The meaning of business QoE prediction.
BE business QoE evaluation method provided by the invention based on weber Fechner theorem, comprising the following steps:
Step 1: proposing that one kind is novel on the basis of QoE evaluation model of the tradition based on WFL and utilize vision to believe based on WFL
Breath carries out BE business QoE evaluation model.
Step 2: based on the QoE evaluation model in step 1, establishing webpage visual information content and extract model, complete QoE
The extraction of webpage visual information content in evaluation model.
Step 3: based on the QoE evaluation model in step 1, carrying out subjective experiment, extract in QoE evaluation model
Training parameter.
Step 4: according to the visual information amount and training parameter extracted in step 2 and step 3, utilizing the QoE in step 1
Evaluation model obtains the QoE evaluation result of BE business, that is, the MOS value predicted.
In the step 1, only regard business total waiting time as single one physical in traditional QoE evaluation model based on WFL
Stimulation carries out QoE prediction, and proposed by the invention based on WFL is in WFL using the BE business QoE evaluation model of visual information
Middle to regard the webpage visual information content of unit waiting time as physical stimulation, QoE evaluation model can be indicated with following formula:
I therein indicates webpage visual information content, and T indicates the webpage unit waiting time, and α, β, k, ω are from subjective experiment
The training parameter of middle extraction.
In the step 2, two parts are broadly divided into the extraction of webpage visual information content, calculate separately picture in webpage
Visual information amount and textual visual information content.The picture visual information amount calculation method are as follows: picture is subjected to piecemeal processing,
It carries out 8*8DCT transformation (also referred to as cosine discrete transform), calculates picture using the probability of occurrence of each block of information after dct transform
Visual information amount；The textual visual information computing method are as follows: calculate the number that different letters occur in every section of text, benefit
The calculating of textual visual information content is completed with the letter frequency in natural language.
In the step 3, reasonable subjective experiment, the MOS value of counting user and waiting time are designed, calculates webpage visual
Information content.Based on the QoE evaluation model in step 1, in the case where known I, T and QoE value, nonlinear regression and fitting is carried out,
Obtain the specific value of each training parameter.
In the step 4, the training parameter obtained in step 3 is substituted into the QoE evaluation model formula of step 1, to known
Waiting time, the business for calculating webpage visual information content carry out the assessment prediction of QoE.
As can be seen that technical solution of the present invention utilizes visual information by establishing the novel WFL that is based on from above-mentioned steps
The QoE evaluation model for carrying out BE business considers influence of the visual information to terminal user QoE to utilize subjectivity in QoE evaluation
It tests the training parameter obtained to be modified QoE evaluation model, realizes quick, accurate and effective BE business QoE evaluation.
A kind of the advantages of BE business QoE evaluation method based on weber Fechner theorem of the invention, is:
(1) influence by visual information to user QoE is taken into account, can more accurately carry out BE business QoE evaluation and
Prediction.
(2) the visual information abstracting method of webpage is divided into picture information quantity calculating and text information meter in webpage
It calculates, can easily and fast and accurately extract the visual information of webpage.
Detailed description of the invention
Fig. 1 is a kind of system framework figure of the BE business QoE evaluation method based on weber Fechner theorem of the present invention；
Fig. 2 is the subjective experiment setting schematic diagram that training parameter is obtained in the present invention；
Fig. 3 is the building flow chart that webpage visual information content extracts model in the present invention；
Fig. 4 is the deviation journey for the subjective MOS value that the prediction MOS value of tradition WFL model and subjective experiment provide in embodiment
Spend effect picture；
Fig. 5 is the inclined of the subjective MOS value that the prediction MOS value of model provided by the present invention is provided with subjective experiment in embodiment
From degree effect picture.
Specific embodiment
The present invention is described in detail with reference to the accompanying drawings and examples.
The present invention provides a kind of BE business QoE evaluation method based on weber Fechner theorem.Main thought of the invention
It is influence of the vision content information to QoE provided by consideration business on the basis of QoE prediction model of the tradition based on WFL, leads to
Cross the quantity of stimulus that the visual information amount of unit waiting time is regarded as and influences user experience quality QoE；By webpage visual information content
Extraction is divided into the extraction to the extraction of picture visual information amount and textual visual information content in webpage, reaches quantization visual information pair
The influence of QoE, the accurate purpose for carrying out QoE evaluation and foreca.
It is clearer to describe technical solution of the present invention advantage, with reference to the accompanying drawings and examples to of the invention
Specific embodiment is described in further detail.In the following description, technology unrelated to the invention is only done brief
Technology illustrates or directly skips over.
Fig. 1 is a kind of system framework figure of the BE business QoE evaluation method based on weber Fechner theorem of the present invention, is such as schemed
Shown in 1, the QoE evaluation method includes three parts: record qos parameter, webpage visual information extraction, subjective experiment extract
Training parameter.
The qos parameter includes the business waiting time, and the record qos parameter, which refers to, uses web page browsing in user
When business, record user is from web-page requests are issued to the total time for opening entire webpage.
Webpage visual information content extracts the extraction including picture visual information amount and the extraction of textual visual information content, mainly
It is to calculate the webpage that user accesses to can be provided to the visual information amount of user.
Subjective experiment extraction training parameter refers to carries out suitable subjectivity under the current network environment of user experience business
Experiment, the subjective scoring that Binding experiment personnel provide obtain the training under current network environment using the method for nonlinear fitting
Parameter.
Three above part provides input quantity for the QoE of assessment business, it is known that training parameter can determine current network ring
The BE business QoE evaluation model formula used under border, for the particular webpage that user is experienced, recorded the waiting time,
After the visual information of webpage has been calculated, input QoE evaluation model formula respectively, can assessment prediction go out the MOS value of the business.
Fig. 2 is the subjective experiment setting schematic diagram that training parameter is extracted in the present invention.As shown in Fig. 2, carrying out subjective experiment
Need the computer that can be surfed the Internet, a web page server and a network simulation server.
Web page server is accessed for placing webpage for user, and the type of webpage of placement mainly includes news category webpage, purchase
Species webpage and searching class webpage.Meanwhile a MOS database is constructed on the web page server, for recording experimenter
Experience the MOS score value provided when web page browsing.
Network simulation server is used to control the waiting time of access bandwidth and web page access, and when waiting described in record
Between and web page server will be sent the waiting time to and be recorded into database.
Computer browses webpage for user, and gives a mark, and MOS value is transmitted to web page server after the completion of marking, records
In MOS database.
The embodiment of the present invention deploys 10 webpages on web page server, this 10 webpages have different pictorial informations
Accounting and text information accounting.The energy of the waiting time of access bandwidth and web page access is controlled using network simulation server simultaneously
5 different waiting time conditions are arranged in power.50 laymans are selected to participate in subjective experiment as experimenter altogether, this
50 experimenters are divided into 10 groups, every group of 5 people.This 10 groups of experimenters are (different in 5 different conditions respectively as unit of group
Waiting time) under, it gives a mark to 10 webpages, lower 5 people of identical conditions is taken into average work to the MOS value that same webpage provides
For the subjective experience mass value under this condition to this webpage.Every group will provide 50 MOS values, and 10 groups there are out 500 MOS values.
Corresponding webpage information amount and waiting time are recorded in MOS database by this 500 MOS values together, share 500 groups of realities
Test data.This 500 groups of experimental datas in present example, 80% experimental data are used to construct subjective experiment library, extract instruction
Practice parameter, 20% experimental data is used to carry out performance verification to this method.
Fig. 3 is the method flow diagram of webpage visual information extraction in the present invention.Two are divided into webpage visual information extraction
Part, the extraction to the extraction of picture visual information amount and textual visual information content in webpage, it is real that this two parts is based on MATLAB
It is existing.
Mainly there are following steps to the extraction of picture visual information amount:
Step 301: webpage being stored in local, extracts picture from webpage, and using the sentence in MATLAB by webpage
In picture batch read in.
Step 302: the picture of reading being divided into the block of information B (x, y) of 8*8, x, y indicate the position coordinates of block of information, right
The corresponding picture numerical matrix f (i, j) of each block of information carries out 8*8DCT transformation (discrete cosine transform), transforms in DCT domain
Numerical matrix F (u, v).
Step 303: using MATLAB calculate dct transform after obtain numerical matrix F (u, v) probability of occurrence P (F (u,
V)), the probability of occurrence of numerical matrix F (u, v) is the probability of occurrence P (B (x, y)) of block of information B (x, y), i.e.,.
P (B (x, y))=P (F (u, v))
Step 304: picture visual information amount is summation after each block of information information content normalization.
The information computing of single block of information B (x, y):
I (B (x, y))=- log2P(B(x,y))
Normalization:
The visual information amount of i-th of picture:
In formulaIndicate region area shared by i-th of picture in Web page picture region.
Mainly there are following steps to the extraction of textual visual information content:
Step 305: webpage being stored in local, text information is extracted from webpage .TXT text is saved as unit of paragraph
Part, and read in the files in batch comprising text information using the sentence in MATLAB.
Step 306: counting the times N that each English alphabet occurs in each file using the cyclic program in MATLABi。
Step 307: textual visual information content is summation after each alphabetical information amount normalization.
According to the letter frequency p in English languageiCalculate alphabetical information amount Ii:
Ii=-pi log2(pi) i ∈ { a, b, c... } (26 alphabetical letter frequencies are shown in Table 1)
The gross information content of jth section text
Step 308: comprehensive.
The page area A of webpagewebpageIt mainly include three kinds of parts: picture region Aimage, text filed Atext, blank area
Domain Ablank.Picture region is made of many pictures, and region area shared by i-th of picture isIt is text filed by very much
Duan Wenben is formed, and region area shared by jth section text isWhite space also divides the white space of different location, and k-th
The region area of blank parts isWherein, white space does not provide visual information amount, picture region and text filed
Visual information percentage contribution is related with the area coverage in the region.Using the area of entire webpage as 1, accounting for for each region is calculated
Plate area.
Step 309: calculating webpage visual information content.
Using visible elements (picture and text) between occupied area and overall page area ratio, that is, plate rate as should
The vision weight of visual zone calculates the visual information amount of entire webpage using following formula.
The corresponding frequency of occurrences of each letter in 1 English of table
As a result it indicates:
Based on the subjective experiment library constructed in the embodiment of the present invention, in the totally 500 groups of data generated to subjective experiment
80% i.e. 400 groups of data carry out Multiple non-linear, and the value of each training parameter obtained is successively are as follows:
α=1.0558, β=1.5648, k=1.3989, ω=2.8019
The value of these training parameters is brought into obtain in the QoE evaluation model of this method and is suitable under current network environment
User's QoE judgement schematics:
It to the performance verification of this method is commented by comparing this method and QoE of the tradition based on WFL in the embodiment of the present invention
Estimate the QoE estimated performance and accuracy of method.Calculate separately RMSE (the Root Mean Square of two kinds of prediction techniques
Error, root-mean-square error more level off to 0, and estimated performance is better) and SROCC (Spearman Rank Order
Correlation Coefficient, history more level off to 1 than Germania coefficient of rank correlation, and estimated performance is better), and it is right respectively
Than the difference between the two methods prediction QoE value obtained and subjectivity QoE value.Table 2 shows the performance comparison of both models
As a result, from Table 2, it can be seen that compared to traditional QoE prediction model based on WFL for only considering the waiting time, the present invention
The RMSE of the model of offer more levels off to 0, SROCC and more levels off to 1, i.e., this method propose using visual information based on WFL's
QoE model has better estimated performance.
The comparison of 2 model performance of table
Based on the subjective experiment in the embodiment of the present invention, remaining 100 data in subjective experiment library are constructed for verifying this
The prediction accuracy of model, Fig. 4 and Fig. 5 provide the prediction accuracy comparison of two kinds of models.Abscissa in Fig. 4 and Fig. 5 is
The subjective MOS value that subjective experiment provides, Fig. 4 ordinate are the prediction MOS value of tradition WFL model, and Fig. 5 ordinate is institute of the present invention
The prediction MOS value of model is provided, straight line is prediction MOS value and the identical situation of subjectivity MOS value, and the distribution of point is closer to straight line table
Show prediction MOS value closer to subjectivity MOS value, departure degree is bigger, and expression prediction is more inaccurate.It can be seen from the figure that in Fig. 4
Point distribution deviate straight line degree be apparently higher than the midpoint Fig. 5 distribution, using this method assessment prediction QoE closer to user master
MOS value is seen, this indicates that the prediction result of institute's climbing form type of the present invention is more accurate compared with traditional WFL model.
Claims (1)
1. the type service BE business QoE evaluation method of doing one's best based on weber Fechner theorem WFL, it is characterised in that: including
Following steps,
Step 1: establishing and BE business QoE evaluation model is carried out using visual information based on WFL；The QoE evaluation model formula
It indicates:
I therein indicates webpage visual information content, and T indicates the webpage unit waiting time, and α, β, k, ω are mentioned from subjective experiment
The training parameter taken；
Step 2: establishing webpage visual information content and extract model, complete the extraction of webpage visual information content in QoE evaluation model；Institute
The extraction for the webpage visual information content stated includes the extraction of picture visual information amount and the extraction of textual visual information content in webpage；
Step 3: carrying out subjective experiment, extract the training parameter in QoE evaluation model；
Step 4: according to the webpage visual information content and training parameter extracted, the QoE of BE business is obtained using QoE evaluation model
Evaluation result, that is, the MOS value predicted；
The picture visual information amount extract the following steps are included:
Step 301: webpage being stored in local, picture is extracted from webpage, and will be in webpage using the sentence in MATLAB
Picture batch is read in；
Step 302: the picture of reading being divided into the block of information B (x, y) of 8*8, x, y indicate the position coordinates of block of information, to each
The corresponding picture numerical matrix f (i, j) of block of information carries out 8*8DCT transformation, transforms to the numerical matrix F (u, v) in DCT domain:
Step 303: the probability of occurrence P (F (u, v)) of the numerical matrix F (u, v) obtained after dct transform, number are calculated using MATLAB
The probability of occurrence of value matrix F (u, v) is the probability of occurrence P (B (x, y)) of block of information B (x, y), that is,
P (B (x, y))=P (F (u, v))
Step 304: picture visual information amount is summation after each block of information information content normalization；
The information computing of single block of information B (x, y):
I (B (x, y))=- log2P(B(x,y))
Normalization:
The visual information amount of i-th of picture:
In formulaIndicate region area shared by i-th of picture in Web page picture region；
Extraction to textual visual information content the following steps are included:
Step 305: webpage is stored in local, text information is extracted from webpage .TXT file is saved as unit of paragraph, and
The files in batch comprising text information is read in using the sentence in MATLAB；
Step 306: counting the times N that each English alphabet occurs in each file using the cyclic program in MATLABi；
Step 307: textual visual information content is summation after each alphabetical information amount normalization；
According to the letter frequency p in English languageiCalculate alphabetical information amount Ii:
Ii=-pilog2(pi) i∈{a,b,c...}
The gross information content of jth section text
Step 308: comprehensive；
The page area A of webpagewebpageInclude three kinds of parts: picture region Aimage, text filed Atext, white space Ablank,
Picture region is made of many pictures, and region area shared by i-th of picture isIt is text filed by many section texts
It forms, region area shared by jth section text isWhite space also divides the white space of different location, k-th of gutter
Point region area beUsing the area of entire webpage as 1, calculate each region accounts for plate area:
Step 309: calculating webpage visual information content；
Using plate rate as the vision weight of visual zone, the visual information amount of entire webpage is calculated using following formula:
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510828384 | 2015-11-25 | ||
| CN2015108283840 | 2015-11-25 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106027291A CN106027291A (en) | 2016-10-12 |
| CN106027291B true CN106027291B (en) | 2019-03-12 |
Family
ID=57096853
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610320052.6A Active CN106027291B (en) | 2015-11-25 | 2016-05-13 | QoE evaluation method for BE service based on Weber Fechner theorem |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106027291B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110430440B (en) * | 2019-06-19 | 2021-11-30 | 鹏城实验室 | Video transmission method, system, computer device and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1391865A1 (en) * | 2002-08-23 | 2004-02-25 | Deutsche Thomson-Brandt Gmbh | Plasma display panel (PDP) - Reduction of dithering noise while displaying less video levels than required |
| CN102087652A (en) * | 2009-12-08 | 2011-06-08 | 百度在线网络技术（北京）有限公司 | Method for screening images and system thereof |
| CN102630038A (en) * | 2012-04-13 | 2012-08-08 | 北京邮电大学 | Mapping method from video objective parameters to QoE (Quality of Experience) based on video contents |
| CN104112064A (en) * | 2014-07-01 | 2014-10-22 | 河南科技大学 | Touch comfort level model based on Weber-Fechner law |
-
2016
- 2016-05-13 CN CN201610320052.6A patent/CN106027291B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1391865A1 (en) * | 2002-08-23 | 2004-02-25 | Deutsche Thomson-Brandt Gmbh | Plasma display panel (PDP) - Reduction of dithering noise while displaying less video levels than required |
| CN102087652A (en) * | 2009-12-08 | 2011-06-08 | 百度在线网络技术（北京）有限公司 | Method for screening images and system thereof |
| CN102630038A (en) * | 2012-04-13 | 2012-08-08 | 北京邮电大学 | Mapping method from video objective parameters to QoE (Quality of Experience) based on video contents |
| CN104112064A (en) * | 2014-07-01 | 2014-10-22 | 河南科技大学 | Touch comfort level model based on Weber-Fechner law |
Non-Patent Citations (1)
| Title |
|---|
| The Logarithmic Nature of QoE and the Role of;Peter Reichl，Sebastian Egger,Raimund Schatz;《IEEE ICC 2010》;20101231;全文 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106027291A (en) | 2016-10-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109033408B (en) | Information pushing method and device, computer readable storage medium and electronic equipment | |
| CN1936893B (en) | Method and system for generating input-method word frequency base based on internet information | |
| CN112104642B (en) | Abnormal account number determination method and related device | |
| US20170235726A1 (en) | Information identification and extraction | |
| CN104834731A (en) | Recommendation method and device for self-media information | |
| CN101819585A (en) | Device and method for constructing forum event dissemination pattern | |
| CN111882224B (en) | Method and device for classifying consumption scenarios | |
| CN106294718A (en) | Information processing method and device | |
| CN111143688B (en) | Evaluation method and system based on mobile news client | |
| CN112150094A (en) | Model training method, model-based evaluation method and device | |
| US11269896B2 (en) | System and method for automatic difficulty level estimation | |
| US20150088876A1 (en) | Forensic system, forensic method, and forensic program | |
| CN106027291B (en) | QoE evaluation method for BE service based on Weber Fechner theorem | |
| US10853429B2 (en) | Identifying domain-specific accounts | |
| CN108733672A (en) | The method and apparatus for realizing network information quality evaluation | |
| CN106575418A (en) | Suggested keywords | |
| CN108259588A (en) | A kind of method for pushing and device of the cultural cloud platform based on big data | |
| Hong et al. | The structural complexity of Chinese words and its relationship with word frequency | |
| Hambley et al. | Optimising web accessibility evaluation: Population sourcing methods for web accessibility evaluation | |
| CN104252411B (en) | A kind of system pressure analysis method and equipment | |
| CN110598960B (en) | Entity-level emotion assessment method and device | |
| Pukas et al. | Intelligent Analyzing Module in the Academic Staff Performance Appraisal System. | |
| Ishikawa et al. | Detection of users suspected of using multiple user accounts and manipulating evaluations in a community site | |
| CN103699602B (en) | A kind of method and apparatus for setting up model essay webpage database | |
| CN118885672B (en) | Employment recommendation method and device and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |