CN201467351U - Television audience rating statistical device based on human face detection - Google Patents
Television audience rating statistical device based on human face detection Download PDFInfo
- Publication number
- CN201467351U CN201467351U CN2009200767751U CN200920076775U CN201467351U CN 201467351 U CN201467351 U CN 201467351U CN 2009200767751 U CN2009200767751 U CN 2009200767751U CN 200920076775 U CN200920076775 U CN 200920076775U CN 201467351 U CN201467351 U CN 201467351U
- Authority
- CN
- China
- Prior art keywords
- module
- face
- people
- top box
- audience rating
- 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
Links
Images
Abstract
The utility model a television audience rating statistical device based on human face detection, which comprises a double camera, an infrared lamp, an infrared lamplight filter, a set-top box and a back-stage server, wherein the double camera is connected with the infrared lamp and the infrared lamplight filter in parallel and is connected with the set-top box, and the set-top box is connected with the back-stage server in a wired or wireless manner and is also connected with a television set. The television audience rating statistical device has the advantages that by utilizing the biological feature identification and identity confirmation as well as watching status, the device performs the statistics on the number of people watching certain advertisements and the degree of preference as well as age, gender and living standard of televiewers, and the detected human face is tracked real-timely to avoid double counting, thereby improving the accuracy degree of the television audience rating.
Description
Technical field
The utility model relates to a kind of audience rating statistical device, particularly a kind of TV audience rating statistic device that detects based on people's face.
Background technology
As everyone knows, the value of advertisement putting and effect directly affect the interests of businessman.
Newspaper and magazine advertisement-printing are according to the circulation defrayment, and the web advertisement is to come defrayment according to pageview, need carry out quantity statistics by technological means at the mode of these advertisement deliveries.
Television advertising is the most effective now a kind of advertisement putting channel, for television advertising, usually broadcaster can not know have how many beholders can see specific advertisement in advance accurately, can not accomplish that more accurate statistics watches information such as the people's of particular advertisement sex, age, living standard.Therefore a large amount of work is used for predicting beholder's quantity, and the corresponding setting price.Guarantee that minimum audient's scale is unrare,, will play advertisement again up to reaching required quantity if can not guarantee.
Usually television advertising divides prime time and non-prime time, and the location of prime time is that the estimation that a large amount of prediction work obtain has been done by broadcaster, but not accurate statistics.The method of carrying out these statistics adds a statistics box often in the centre of cable television line, and the data that these obtain are the number of times that advertisement is play, and really has how many people seeing that this advertisement can not get adding up.The input effect of an advertisement depends on the quantity of watching it, and businessman tends to select the same advertisement of repeat playing in order to reach good effect, so not only will bear huge advertising expenditure, and not necessarily get a desired effect.
Audience ratings is to weigh a kind of statistical indicator that TV programme has the attendance.Being the common transaction currency of television advertising, is the important indicator of assessment spectator attendance.As analysing in depth the data basis that television program receiving market provides, be not only the important indicator of Promgramming, assessment, also be the strong instrument of formulating advertisement putting plan, assessment advertising results.Audience ratings both had been TV station's service, also was advertising company, advertiser's service, also was the decision-making foundation of broadcasting and TV administrative department simultaneously.
But problem is: one, this not exact figure of audience ratings is used for accuracy and requires high advertising cost to calculate, judge in view of the above whether TV station has fulfiled the rating point of promising to undertake in the contract, and definite compensation scheme; Two, determine that wanting the clutter of after-culture is so accurately, and there is inexactness in the data of the standard that affords redress for it; Three, because the difference that computational methods and target group define also often causes the dispute to effective rating point of TV station and advertising company.
Adopt two kinds of statistical methods at present: 1) diary chucking method: carry out hand-kept; 2) measuring instrument method: need the user to press the button in beginning with when finishing.
But, the pure manual method of diary card, the possibility that can make a mistake even forget, subjectivity is very big; And individual rating measuring instrument method is that each rating personnel of dependence are forward and backward in rating, initiatively confirms the observer identity by the mode of corresponding button.This " initiative " requires just inevitably can to cause the problem of " forget by ".Although research firm claims that all this part data can revise, still can cause the query of data accuracy.
The utility model content
The utility model is that a kind of TV audience rating statistic device based on the detection of people's face that improves the audience rating statistical accuracy will be provided.
In order to solve above technical problem, the utility model provides a kind of TV audience rating statistic device that detects based on people's face, this device comprises dual camera, infrared lamp, infrared lamp optical filter, set-top box and background server, wherein, described dual camera and infrared lamp, infrared lamp optical filter are in parallel, dual camera is connected with set-top box in addition, and set-top box is connected with background server by wired or wireless mode, and set-top box also is connected with television set.
Described set-top box comprises people's face detection module, face recognition module, face tracking module, demographics module, data analysis processing module and wireless transport module, described people's face detection module is connected with face recognition module, face tracking module, demographics module respectively, face recognition module, face tracking module, demographics module are connected with the data analysis processing module respectively, and the data analysis processing module is connected with background server by wireless transport module; People's face detection module receives the human face image information of being obtained by dual camera, and the people's face information that detects in the image is sent to face recognition module; Face recognition module is discerned people's face information, determines beholder's age, sex information; Simultaneously people's face detection module is sent to face tracking module and demographics module with people's face information respectively, and the face tracking module is carried out real-time tracking to people's face information, to people's face information carry out local towards with the analysis of whether sleeping; The demographics module is determined beholder's quantity information; Face recognition module, face tracking module, demographics module are sent to the data analysis processing module with beholder's age, sex information, trace information, quantity information respectively and carry out analyzing and processing, information after handling is sent to background server by wireless transport module, and background server counts this period beholder's statistics.
Superior effect of the present utility model is:
1) utilizes living things feature recognition and affirmation identity and viewed status; Can discern viewer identity automatically, need not any active behavior to take place with spectators; Whether the energy real-time judge is left or is returned; Can judge whether spectators fall asleep or wake up;
2) age-sex's living standard of number, preference degree and the beholder of the utility model by being watched particular advertisement each time period is added up, and detected people's face carried out real-time tracking, avoid double counting, advertising company can select to play the advertisement that is fit to the beholder or select to play the higher advertisement of beholder's preference degree at current slot; For commercials company has brought accurate statistics, for later ad distribution time and price provide accurate foundation from new location;
3) the utility model is given background server by the transfer of data of wireless mode after with the set-top box analyzing and processing, the possibility of being attacked and obtain data by the hacker by cable network reduces greatly, the network security performance is good, is convenient to install, and does not need the user installation broadband device.
4) the utility model is not preserved and transmission video signal, just with the transfer of data after the analyzing and processing to the background server end, can guarantee user's personal secrets.
Description of drawings
Fig. 1 is a structural representation of the present utility model;
Fig. 2 is the theory diagram of set-top box;
The number in the figure explanation
The 1-dual camera; The 2-infrared lamp;
3-infrared lamp optical filter;
The 4-set-top box;
401-people's face detection module; The 402-face recognition module;
403-face tracking module; 404-demographics module;
405-data analysis processing module; The 406-wireless transport module;
The 5-television set; The 6-background server;
The 7-beholder.
Embodiment
See also shown in the accompanying drawing, the utility model will be further described.
Shown in Fig. 1 structural representation of the present utility model, the utility model provides a kind of TV audience rating statistic device that detects based on people's face, this device comprises dual camera 1, infrared lamp 2, infrared lamp optical filter 3, set-top box 4 and background server 6, described dual camera 1 and infrared lamp 2, infrared lamp optical filter 3 are in parallel, dual camera 1 is connected with set-top box 4 in addition, set-top box 4 is connected with background server 6 by wired or wireless mode, and set-top box 4 also is connected with television set 5.
Dual camera 2 can carry out the IMAQ of people's face to 180 degree wide before the video screen, facial image by 3 couples of beholders 7 under dark surrounds of infrared lamp is gathered, 3 pairs of Infrared of infrared lamp optical filter effectively filter, and eliminate the sense of discomfort of 7 pairs of infrared light of beholder.
Shown in the theory diagram of Fig. 2 set-top box, described set-top box 4 comprises people's face detection module 401, face recognition module 402, face tracking module 403, demographics module 404, data analysis processing module 405 and wireless transport module 406, described people's face detection module 401 respectively with face recognition module 402, face tracking module 403, demographics module 404 connects, face recognition module 402, face tracking module 403, demographics module 404 is connected with data analysis processing module 405 respectively, and data analysis processing module 405 is connected with background server 6 by wireless transport module 406; People's face detection module 403 receives the human face image information of being obtained by dual camera 1, and the people's face information that detects in the image is sent to face recognition module 402; 402 pairs of people's faces of face recognition module information is discerned, and determines beholder 7 age, sex information; Simultaneously people's face detection module 401 is sent to face tracking module 403 and demographics module 404 with people's face information respectively, and 403 pairs of people's faces of face tracking module information is carried out real-time tracking, to people's face information carry out local towards with the analysis of whether sleeping; Demographics module 404 is determined beholder 7 quantity information; Face recognition module 402, face tracking module 403, demographics module 404 are sent to data analysis processing module 405 with beholder 7 age, sex information, trace information, quantity information respectively and carry out analyzing and processing, information after handling is sent to background server 6 by wireless transport module 406, and background server 6 counts this period beholder 7 statistics.
The image that 401 pairs of dual cameras 1 of people's face detection module collect carries out people's face and detects, the main eyes location algorithm that adopts in the human face detection tech.Be the image processing that collects gray level image at first, former gray level image done edge gray scale reinforcement; In conjunction with the position of human eye decision criteria of setting up according to people's face geometric properties priori, in the process that segmentation threshold increases progressively, searching can be partitioned into the optimum segmentation threshold value of eyes eye piece then; The authenticity of coming the eyes that inspection goes out as symmetrical similarity at last with the bidimensional coefficient correlation, and utilize the gray integration projection of the binocular images vertical direction that finds, accurately locate pupil center.After human eye was accurately positioned, then other features of face as eyebrow, nose, mouth etc., can be located more exactly by potential distribution relation.For the image that collects, utilize the colour of skin to set up complexion model at the distribution character of YCbCr color space, find out area of skin color roughly, carry out greyscale transformation after, adopt said method to detect people's face again.
404 pairs of detected people's face numbers of demographics module are added up.
405 pairs of face recognition module 402 of data analysis processing module, demographics module 404, face tracking module 403 gained results carry out statistical analysis, analyze the current number that collects, age, sex, to preference degree of program or the like, send analysis result to wireless transport module 406.
Detection method of the present utility model comprises following step:
(1) start, the user logins;
(2) start dual camera 1; If be in dark surrounds, then start infrared lamp 2,3 pairs of Infrared of infrared lamp optical filter filter, and eliminate the sense of discomfort of beholder to infrared light;
(3) people's face of 1 couple of beholder 7 of dual camera carries out IMAQ;
(4) image information of obtaining is sent to people's face detection module 401 of set-top box 4;
(5) people's face detection module 401 detects the people's face information in the image;
(6) people's face detection module 401 is sent to face recognition module 402 with people's face information, sees for (7) step;
Simultaneously, people's face detection module 401 is sent to face tracking module 403 with people's face information, sees for (8) step;
Simultaneously, people's face detection module 401 is sent to demographics module 404 with people's face information, sees for (9) step;
(7) face recognition module 402 is carried out recognition of face, identifies beholder 7 age, sex information, and this information is sent to data analysis processing module 405; Changeed for (10) step;
(8) 403 pairs of detected people's faces of face tracking module carry out real-time tracking, analyze beholder 7 face orientation, the information of whether sleeping, and this information is sent to data analysis processing module 405; Changeed for (10) step;
(9) 404 pairs of detected people's faces of demographics module carry out quantity statistics, and the result is sent to data analysis processing module 405;
(10) data analysis processing module 405 will come from the information of face recognition module 402, face tracking module 403 and demographics module 404 and carry out comprehensive analysis processing, and result is sent to wireless transport module 406;
(11) wireless transport module 406 sends data to background server 6 by wireless mode;
(12) background server 6 draws the beholder's 7 of this period statistics according to result.
Family with some five mouthfuls of people is an example. after turning on television set 5 and registering, dual camera 1 begins 180 degree wide before the video screen are carried out the IMAQ of people's face, people's face detection module 401 detects four people's faces, then thinking has four people watching current TV programme. and 403 pairs of detected people's faces of face tracking module are followed the tracks of, confirm beholder 7 number, and analysis spectators' preference degree. for example, detected four philtrums, there is a people before television set 5, to leave, confirms that then current attendance is three people; Detected a people in sleep, two people's face orientation is not towards television set, think that then the preference degree of 7 pairs of current programs of beholder is lower. information such as face recognition module 402 analysis beholders' 7 age, sex, for example, if current beholder 7 has three male sex, a women thinks that then current program relatively attracts the spectators male sex; Perhaps current beholder 7 has three people more than 50 years old, three people about 20 years old, think that then current program relatively attracts old beholder's 7. data analysis processing modules 405 that above gained data are carried out statistical analysis, analysis result is sent to background server 6 by wireless transport module 406, background server 6 is according to statistic analysis result, accurately count beholder's 7 numbers of current television program, and the range of age of beholder 7, sex etc.
The utility model adopts the detection of people's face, tracking and recognition technology, not only counts beholder's number of TV programme comparatively exactly, and can also count beholder's age, sex, to the information such as preference degree of program.For commercials company, can the arrangement of program be shot the arrow at the target, select to play the advertisement that is fit to the beholder or select to play the higher advertisement of beholder's preference degree at current slot.For commercials company has brought accurate statistics, reorientating for later ad distribution time and price provides foundation.For advertiser, can be chosen in the advertisement that the suitable period plays suitable number of times, thereby save advertising expenditure effectively.And the utility model is not preserved transmission of video, only gives background server with the data analysis result transmission, has not only ensured the interests of businessman, has ensured user's privacy yet.Therefore, the utlity model has high social and economic benefit.
Claims (2)
1. TV audience rating statistic device that detects based on people's face, this device comprises dual camera, infrared lamp, infrared lamp optical filter, set-top box and background server, it is characterized in that:
Described dual camera and infrared lamp, infrared lamp optical filter are in parallel, and dual camera is connected with set-top box in addition, and set-top box is connected with background server by wired or wireless mode, and set-top box also is connected with television set.
2. by the described TV audience rating statistic device that detects based on people's face of claim 1, it is characterized in that:
Described set-top box comprises people's face detection module, face recognition module, face tracking module, demographics module, data analysis processing module and wireless transport module, described people's face detection module is connected with face recognition module, face tracking module, demographics module respectively, face recognition module, face tracking module, demographics module are connected with the data analysis processing module respectively, and the data analysis processing module is connected with background server by wireless transport module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009200767751U CN201467351U (en) | 2009-06-22 | 2009-06-22 | Television audience rating statistical device based on human face detection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009200767751U CN201467351U (en) | 2009-06-22 | 2009-06-22 | Television audience rating statistical device based on human face detection |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN201467351U true CN201467351U (en) | 2010-05-12 |
Family
ID=42394697
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2009200767751U Expired - Fee Related CN201467351U (en) | 2009-06-22 | 2009-06-22 | Television audience rating statistical device based on human face detection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN201467351U (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102129644A (en) * | 2011-03-08 | 2011-07-20 | 北京理工大学 | Intelligent advertising system having functions of audience characteristic perception and counting |
| CN102231813A (en) * | 2011-01-17 | 2011-11-02 | 康佳集团股份有限公司 | Television with function of face recognition child lock |
| CN104581383A (en) * | 2014-12-12 | 2015-04-29 | 百视通网络电视技术发展有限责任公司 | System and method for analyzing program watching populations through imager |
| CN106067008A (en) * | 2016-06-15 | 2016-11-02 | 汤美 | Student's statistical method of network courses and system |
-
2009
- 2009-06-22 CN CN2009200767751U patent/CN201467351U/en not_active Expired - Fee Related
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102231813A (en) * | 2011-01-17 | 2011-11-02 | 康佳集团股份有限公司 | Television with function of face recognition child lock |
| CN102129644A (en) * | 2011-03-08 | 2011-07-20 | 北京理工大学 | Intelligent advertising system having functions of audience characteristic perception and counting |
| CN104581383A (en) * | 2014-12-12 | 2015-04-29 | 百视通网络电视技术发展有限责任公司 | System and method for analyzing program watching populations through imager |
| CN106067008A (en) * | 2016-06-15 | 2016-11-02 | 汤美 | Student's statistical method of network courses and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101588443A (en) | Statistical device and detection method for television audience ratings based on human face | |
| US10204264B1 (en) | Systems and methods for dynamically scoring implicit user interaction | |
| US8462996B2 (en) | Method and system for measuring human response to visual stimulus based on changes in facial expression | |
| Sugano et al. | Calibration-free gaze sensing using saliency maps | |
| CN108234591B (en) | Content data recommendation method and device based on identity authentication device and storage medium | |
| US8401248B1 (en) | Method and system for measuring emotional and attentional response to dynamic digital media content | |
| CN102779274B (en) | Intelligent television face recognition method based on binocular camera | |
| CN110532970B (en) | Age and gender attribute analysis method, system, equipment and medium for 2D images of human faces | |
| CN104680121B (en) | Method and device for processing face image | |
| CN107146096B (en) | Intelligent video advertisement display method and device | |
| CN104112209A (en) | Audience statistical method of display terminal, and audience statistical system of display terminal | |
| JP4159159B2 (en) | Advertising media evaluation device | |
| CN105160321A (en) | Vision-and-wireless-positioning-based mobile terminal identity verification method | |
| US20100232644A1 (en) | System and method for counting the number of people | |
| CN109544754B (en) | Access control method and system based on face recognition and computer equipment | |
| CN103098079A (en) | Personalized program selection system and method | |
| CN109784274A (en) | Identify the method trailed and Related product | |
| US20170249670A1 (en) | Targeted advertising and facial extraction and analysis | |
| CN103914691A (en) | Target group analysis system and method based on face recognition and height recognition method | |
| CN109345305A (en) | A kind of elevator electrical screen advertisement improvement analysis method based on face recognition technology | |
| CN201467351U (en) | Television audience rating statistical device based on human face detection | |
| US20150249871A1 (en) | Information Push Method And System, Digital Television Receiving Terminal, And Computer Storage Medium | |
| CN112132616B (en) | Mobile multimedia advertisement intelligent pushing management system based on big data | |
| CN105938603A (en) | Personnel interest degree detection system based on machine vision and personnel interest degree detection method thereof | |
| CN104766230A (en) | Advertising effect evaluation method based on human skeletal tracking |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20100512 Termination date: 20110622 |