CN105938603A - Personnel interest degree detection system based on machine vision and personnel interest degree detection method thereof - Google Patents
Personnel interest degree detection system based on machine vision and personnel interest degree detection method thereof Download PDFInfo
- Publication number
- CN105938603A CN105938603A CN201610250120.6A CN201610250120A CN105938603A CN 105938603 A CN105938603 A CN 105938603A CN 201610250120 A CN201610250120 A CN 201610250120A CN 105938603 A CN105938603 A CN 105938603A
- Authority
- CN
- China
- Prior art keywords
- personnel
- information
- intelligence
- interest level
- jettison system
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0242—Determining effectiveness of advertisements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/178—Human faces, e.g. facial parts, sketches or expressions estimating age from face image; using age information for improving recognition
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- General Health & Medical Sciences (AREA)
- Development Economics (AREA)
- Strategic Management (AREA)
- Finance (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a personnel interest degree detection system based on machine vision. The personnel interest degree detection system comprises an signal acquisition unit which is used for acquiring signal information of personnel in a target area; and an signal processing unit which is connected with the signal acquisition unit and used for processing personnel information acquired by the signal acquisition unit and classifying the personnel information according to the interest degree for different advertisements. Therefore, the signal information of the personnel in the target area is acquired by the signal acquisition unit, the personnel information is classified according to the interest degree for different advertisements after processing of the signal processing unit, and the interest degree of the personnel for different information is detected so that targeted corresponding information input can be performed according to the interest degree of the personnel for different information. The invention also provides a personnel interest degree detection method based on machine vision.
Description
Technical field
The information that the present invention relates to throws in field, particularly to a kind of personnel's interest level detecting system based on machine vision and method.
Background technology
Advertisement has vital effect in daily life, and it can not only be putting person's image building, transmit information, additionally it is possible to guide commercial audience to buy and reasonable employment commodity.Advertisement is the tie contacted between putting person and commercial audience.Advertisement is only improved the quantity of target audience and potential audience, can realize increasing of effective audient.
The most centralized input, continuous way are thrown in, or batch (-type) is thrown in, and advertisement is all it cannot be guaranteed that the increase of effective audient in playing the period, because this input mode is not belonging to the most precisely throw in, the quantity of target audience and potential audience is difficult to ensure that.Input object makes a mistake, and also can cause the dislike emotion of commercial audience and resist thought, and this purpose with advertisement putting is runed counter to.Such advertisement does not the most play a positive role, and on the contrary while wasting bigger Financial cost and time cost, compromises the interests of putting person self.
nullAt present，Fast development along with Network Information，The input mode of advertisement presents diversification the most increasingly、Diversified development，Mobile phone、Network multimedia、The appearance of high definition interaction platform，It is increasingly subject to the favor of advertisement operators，And traditional advertisement putting mode is due to by throwing in frequency range，Play、Read period or time and broadcasting、Reading terminal locus，The even restriction of administration means，Far do not reach the requirement that intelligent advertisement is precisely thrown in，Traditional TV、Broadcast、Newspaper、The advertisement putting mode of magazine can not feed back suggestion and the demand of client，Outdoor、The non intelligent platform such as indoor billboard can not be collected or the feedback crowd multidate information to the actual demand of advertisement，Thus cannot realize precisely throwing in advertising business，Seriously govern advertisement accurately and throw in the growth requirement of business，And in advertising business input mode，How testing staff is that those skilled in the art need badly and solve the technical problem that to the interest level of different information.
Summary of the invention
The technical problem to be solved in the present invention is: provide a kind of personnel's interest level detecting system based on machine vision and method, and it can testing staff's interest level to different information.
The solution of the present invention is achieved in that a kind of personnel's interest level detecting system based on machine vision, including:
Signal gathering unit, the signal message of personnel in gathering target area；
Signal processing unit, is connected with described signal gathering unit, for processing the personal information that described signal gathering unit gathers, and personal information is classified by the interest level of different advertisements；
Described signal processing unit includes:
Personnel's demographics unit, utilizes the personal information that described signal gathering unit gathers, and statistics is through personnel's number of intelligence jettison system；
Face identification unit, for identifying the face information in described personal information, and detects human eye and the nose information of each personnel；
Personnel's attention judging unit, utilizes human eye and the nose information of personnel, it is judged that the deflection angle of personnel's face, determines whether personnel's interest level to different information.
So, personnel's interest level detecting system based on machine vision of the present invention, by the signal message of personnel in signal gathering unit collection target area, after signal processing unit processes, personnel are classified by the interest level of different information, by personnel's number is added up, pass through face identification unit, identify the face information in personal information, and detect human eye and the nose information of each personnel, the human eye of recycling personnel and nose information, the deflection angle of judgement personnel's face, determine whether personnel's interest level to different information.After detecting personnel's interest level to different information, according to personnel, the interest level of different information can be thrown in corresponding information targetedly.Thus revolutionize the rough formula put-on method ignoring personnel's interest feedback in the information jettison systems such as existing advertisement or product introduction, improve the effect of the information such as advertisement or demonstration, also improve the Experience Degree of personnel.
Another technical scheme of the present invention is that, on above-mentioned basis, described signal gathering unit includes image collecting device, the picture signal of personnel in gathering target area, and described image collecting device is arranged on shown information and throws on unit.
Another technical scheme of the present invention is on above-mentioned basis, and described image collector is set at least one in photographic instrument, visible image capturing instrument, infrared photography instrument and thermal-imaging camera device.
Another technical scheme of the present invention is that, on above-mentioned basis, described signal gathering unit also includes audio signal sample device, the audio signal of personnel in gathering target area, and described audio signal sample device is arranged on shown information and throws on unit.
Another technical scheme of the present invention is on above-mentioned basis, and described personnel's attention judging unit specifically includes:
Human eye according to personnel and nose information, build the human eye by personnel and nose interconnecting line, the del of composition, assuming that personnel are positioned at the front of intelligence jettison system, when people two arrives nose respectively apart from equal or close, and described del is isosceles triangle, then judge that personnel face intelligence jettison system, attention is concentrated, high to the interest level of information；
When the distance difference that people two arrives nose respectively is bigger, then judge personnel's side-looking intelligence jettison system, absent minded, low to the interest level of information.
Another technical scheme of the present invention is on above-mentioned basis, and described personnel's attention judging unit includes personnel's visual line of sight computing unit:
Before assuming that personnel are positioned at intelligence jettison system, facial orientation is for facing, during sight line zero deflection angle, and the sight angle that people is intrinsicThe intelligence a length of L of jettison system, personnel and distance d of intelligence jettison system, then visual line of sight
Before assuming that personnel are positioned at intelligence jettison system, facial orientation is for facing, during sight line deflection angle θ, and the sight angle that people is intrinsicThe intelligence a length of L of jettison system, personnel and distance d of intelligence jettison system, then visual line of sight
Assuming that personnel have relative sidesway Δ x with intelligence jettison system, facial orientation is for facing, during sight line deflection angle θ, and the sight angle that people is intrinsicThe intelligence a length of L of jettison system, personnel and distance d of intelligence jettison system, then visual line of sight
Wherein, the most closely or too far, visual line of sight is the most invalid for personnel's distance intelligence jettison system, i.e. with d as regional boundary, and d ∈ (dmin,dmax) time effective.
Another technical scheme of the present invention is on above-mentioned basis, and described personnel's attention judging unit also includes personnel's interest level computing unit；Personnel's interest level i.e. visual line of sight is proportion in intelligence jettison system lengthCarry out n sampling, then interest level P is expressed as:
When P is more than or equal to predetermined threshold value, then judge that personnel are interested；When P is less than predetermined threshold value, then judge that personnel lose interest in.
Another technical scheme of the present invention is that, on above-mentioned basis, described signal processing unit also includes:
Personnel's sex recognition unit, the sex of personnel in gathering target area；And/or
Personnel's age recognition unit, the age of personnel in gathering target area；And/or
Personnel identity classification recognition unit, the identity category of personnel in gathering target area；And/or
Personnel's skin color model unit, the colour of skin of personnel in gathering target area.
A kind of personnel's interest level detection method based on machine vision, comprises the following steps:
The signal message of personnel in collection target area；
Process the personal information gathered, and personal information is classified by the interest level of different advertisements；
Process the personal information gathered to specifically include:
Statistics is through personnel's number of intelligence jettison system；
Identify the face information in described personal information, and detect human eye and the nose information of each personnel；
Judgement personnel's attention, utilizes human eye and the nose information of personnel, it is judged that the deflection angle of personnel's face, determines whether personnel's interest level to different information.
Another technical scheme of the present invention is on above-mentioned basis, and described judgement personnel's attention step specifically includes:
Human eye according to personnel and nose information, build the human eye by personnel and nose interconnecting line, the del of composition, assuming that personnel are positioned at the front of intelligence jettison system, when people two arrives nose respectively apart from equal or close, and described del is isosceles triangle, then judge that personnel face intelligence jettison system, attention is concentrated, high to the interest level of information；
When the distance difference that people two arrives nose respectively is bigger, then judge personnel's side-looking intelligence jettison system, absent minded, low to the interest level of information.
Another technical scheme of the present invention is on above-mentioned basis, described judgement personnel's attention step specifically includes: computing staff's visual line of sight, it is assumed that before personnel are positioned at intelligence jettison system, facial orientation is for facing, during sight line zero deflection angle, the sight angle that people is intrinsicThe intelligence a length of L of jettison system, personnel and distance d of intelligence jettison system, then visual line of sight
Before assuming that personnel are positioned at intelligence jettison system, facial orientation is for facing, during sight line deflection angle θ, and the sight angle that people is intrinsicThe intelligence a length of L of jettison system, personnel and distance d of intelligence jettison system, then visual line of sight
Assuming that personnel have relative sidesway Δ x with intelligence jettison system, facial orientation is for facing, during sight line deflection angle θ, and the sight angle that people is intrinsicThe intelligence a length of L of jettison system, personnel and distance d of intelligence jettison system, then visual line of sight
Wherein, the most closely or too far, visual line of sight is the most invalid for personnel's distance intelligence jettison system, i.e. with d as regional boundary, and d ∈ (dmin,dmax) time effective.
Computing staff's interest level again, i.e. visual line of sight be proportion in intelligence jettison system lengthCarry out n sampling, then interest level P is expressed as:
When P is more than or equal to predetermined threshold value, then judge that personnel are interested；When P is less than predetermined threshold value, then judge that personnel lose interest in.
Accompanying drawing explanation
The accompanying drawing of the part constituting the present invention is used for providing a further understanding of the present invention, and the schematic description and description of the present invention is used for explaining the present invention, is not intended that inappropriate limitation of the present invention.
Fig. 1 is the structured flowchart representing the personnel's interest level detecting system based on machine vision involved by present embodiment；
Fig. 2 to Fig. 5 is the calculating schematic diagram of visual line of sight in several embodiments.
Detailed description of the invention
Describing the present invention below in conjunction with the accompanying drawings, the description of this part is only exemplary and explanatory, should not have any restriction effect to protection scope of the present invention.Additionally, those skilled in the art are according to the description of presents, the feature in embodiment in presents and in different embodiment can be carried out respective combination.
The embodiment of the present invention is as follows, refers to Fig. 1, a kind of personnel's interest level detecting system based on machine vision, including:
Signal gathering unit 1, the signal message of personnel in gathering target area；
Signal processing unit 2, is connected with described signal gathering unit, for processing the personal information that described signal gathering unit gathers, and personal information is classified by the interest level of different advertisements；
Described signal processing unit includes:
Personnel's demographics unit, utilizes the personal information that described signal gathering unit gathers, and statistics is through personnel's number of intelligence jettison system；
Face identification unit, for identifying the face information in described personal information, and detects human eye and the nose information of each personnel；
Personnel's attention judging unit, utilizes human eye and the nose information of personnel, it is judged that the deflection angle of personnel's face, determines whether personnel's interest level to different information.
So, personnel's interest level detecting system based on machine vision of the present invention, by the signal message of personnel in signal gathering unit collection target area, after signal processing unit processes, personal information is classified by the interest level of different advertisements, by personnel's number is added up, pass through face identification unit, identify the face information in personal information, and detect human eye and the nose information of each personnel, the human eye of recycling personnel and nose information, the deflection angle of judgement personnel's face, determine whether personnel's interest level to different information.After detecting personnel's interest level to different information, according to personnel, the interest level of different information can be thrown in corresponding information targetedly.Thus revolutionize the rough formula put-on method ignoring personnel's interest feedback in the information jettison systems such as existing advertisement or product introduction, improve the effect of the information such as advertisement or demonstration, also improve the Experience Degree of personnel.
On the basis of above-described embodiment, in another embodiment of the present invention, described signal gathering unit includes image collecting device, the picture signal of personnel in gathering target area, and described image collecting device is arranged on shown information and throws on unit.
On the basis of above-described embodiment, in another embodiment of the present invention, described image collector is set at least one in photographic instrument, visible image capturing instrument, infrared photography instrument and thermal-imaging camera device.
On the basis of above-described embodiment, in another embodiment of the present invention, described signal gathering unit also includes audio signal sample device, the audio signal of personnel in gathering target area, and described audio signal sample device is arranged on shown information and throws on unit.
On the basis of above-described embodiment, in another embodiment of the present invention, described personnel's attention judging unit specifically includes:
Human eye according to personnel and nose information, build the human eye by personnel and nose interconnecting line, the del of composition, assuming that personnel are positioned at the front of intelligence jettison system, when people two arrives nose respectively apart from equal or close, and described del is isosceles triangle, then judge that personnel face intelligence jettison system, attention is concentrated, high to the interest level of information；
When the distance difference that people two arrives nose respectively is bigger, then judge personnel's side-looking intelligence jettison system, absent minded, low to the interest level of information.
On the basis of above-described embodiment, in another embodiment of the present invention, described personnel's attention judging unit includes personnel's visual line of sight computing unit:
Before assuming that personnel are positioned at intelligence jettison system, facial orientation is for facing, during sight line zero deflection angle, and the sight angle that people is intrinsicThe intelligence a length of L of jettison system, personnel and distance d of intelligence jettison system, then visual line of sightAs shown in Figure 2.
Before assuming that personnel are positioned at intelligence jettison system, facial orientation is for facing, during sight line deflection angle θ, and the sight angle that people is intrinsicThe intelligence a length of L of jettison system, personnel and distance d of intelligence jettison system, then visual line of sightAs shown in Figure 3.
Assuming that personnel have relative sidesway Δ x with intelligence jettison system, facial orientation is for facing, during sight line deflection angle θ, and the sight angle that people is intrinsicThe intelligence a length of L of jettison system, personnel and distance d of intelligence jettison system, then visual line of sightAs shown in Figure 4 and Figure 5.
Wherein, the most closely or too far, visual line of sight is the most invalid for personnel's distance intelligence jettison system, i.e. with d as regional boundary, and d ∈ (dmin,dmax) time effective.
On the basis of above-described embodiment, in another embodiment of the present invention, described personnel's attention judging unit also includes personnel's interest level computing unit；Personnel's interest level i.e. visual line of sight is proportion in intelligence jettison system lengthCarry out n sampling, then interest level P is expressed as:
When P is more than or equal to predetermined threshold value, then judge that personnel are interested；When P is less than predetermined threshold value, then judge that personnel lose interest in.
On the basis of above-described embodiment, in another embodiment of the present invention, described signal processing unit also includes:
Personnel's sex recognition unit, the sex of personnel in gathering target area；And/or
Personnel's age recognition unit, the age of personnel in gathering target area；And/or
Personnel identity classification recognition unit, the identity category of personnel in gathering target area；And/or
Personnel's skin color model unit, the colour of skin of personnel in gathering target area.
nullSuch as，According to personnel's interest level analysis to dissimilar advertisement，In conjunction with personnel's sex、Age、The classification information of identity category and the colour of skin etc.，Can speculate and analyze different sexes、All ages and classes、The personnel of different identity classification and the different colour of skin are different with demand to the interest of dissimilar advertisement，According to its statistical information，Make rational planning for adline、Time、The accurate input in place，Such as this period between the 16:30 to 18:00 every afternoon of certain community，Personnel before advertisement display terminal more concentrate、Fixing，By statistical analysis in the personnel of this period，Find the Ms's effective viewing to advertisement、Reading amount is higher，Man is the most slightly lower，Further analyze and find that young Ms is higher to the interest level of the advertisement of toiletries，And in numerous advertisements for cosmetics，The advertisement of skin care item is favored by more Ms，By the result of statistical analysis it is so，The a certain period in Huo Zhe community, this community，The advertisement of the skin care item thrown in cosmetics can be concentrated，Thus according to the situation of the different crowd understood and grasped，Pinpoint、Regularly、The accurate intelligent of orientation throws in advertisement.
On the basis of above-described embodiment, in another embodiment of the present invention, a kind of personnel's interest level detection method based on machine vision, comprise the following steps:
The signal message of personnel in collection target area；
Process the personal information gathered, and personal information is classified by the interest level of different advertisements；
Process the personal information gathered to specifically include:
Statistics is through personnel's number of intelligence jettison system；
Identify the face information in described personal information, and detect human eye and the nose information of each personnel；
Judgement personnel's attention, utilizes human eye and the nose information of personnel, it is judged that the deflection angle of personnel's face, determines whether personnel's interest level to different information.
Another technical scheme of the present invention is on above-mentioned basis, and described judgement personnel's attention step specifically includes:
Human eye according to personnel and nose information, build the human eye by personnel and nose interconnecting line, the del of composition, assuming that personnel are positioned at the front of intelligence jettison system, when people two arrives nose respectively apart from equal or close, and described del is isosceles triangle, then judge that personnel face intelligence jettison system, attention is concentrated, high to the interest level of information；
When the distance difference that people two arrives nose respectively is bigger, then judge personnel's side-looking intelligence jettison system, absent minded, low to the interest level of information.
Another technical scheme of the present invention is on above-mentioned basis, described judgement personnel's attention step specifically includes: computing staff's visual line of sight, it is assumed that before personnel are positioned at intelligence jettison system, facial orientation is for facing, during sight line zero deflection angle, the sight angle that people is intrinsicThe intelligence a length of L of jettison system, personnel and distance d of intelligence jettison system, then visual line of sight
Before assuming that personnel are positioned at intelligence jettison system, facial orientation is for facing, during sight line deflection angle θ, and the sight angle that people is intrinsicThe intelligence a length of L of jettison system, personnel and distance d of intelligence jettison system, then visual line of sight
Assuming that personnel have relative sidesway Δ x with intelligence jettison system, facial orientation is for facing, during sight line deflection angle θ, and the sight angle that people is intrinsicThe intelligence a length of L of jettison system, personnel and distance d of intelligence jettison system, then visual line of sight
Wherein, the most closely or too far, visual line of sight is the most invalid for personnel's distance intelligence jettison system, i.e. with d as regional boundary, and d ∈ (dmin,dmax) time effective.
Computing staff's interest level again, i.e. visual line of sight be proportion in intelligence jettison system lengthCarry out n sampling, then interest level P is expressed as:
When P is more than or equal to predetermined threshold value, then judge that personnel are interested；When P is less than predetermined threshold value, then judge that personnel lose interest in.
Specifically, described personnel's interest level detecting system based on machine vision and method, first with motion detection algorithm, by human bodily form's profile, the flow of personnel amount before intelligence jettison system is added up, such as VIBE algorithm, optical flow algorithm, frame difference method, first moving target area-of-interest in image is extracted, by doing, with background, the foreground target that difference obtains in institute's load image, and utilize pedestrian's body-contoured feature, such as: the occupied area in the ratio of width to height, image, Preliminary screening goes out the pedestrian in target；Then extract HOG pedestrian detection feature, and use support vector machines pedestrian's grader that target is classified, the most accurately identify.
Personnel before emphasis rests on intelligence jettison system to it again carry out the identification at sex, age, the most first facial image is carried out gaussian filtering, it is used for training BP neutral net by pretreated image normalization, to obtain sex recognition classifier, it is carried out sex identification；After extracting the overall situation and the local feature of face, such as face textural characteristics, use SVM regression training to obtain age function, it is thus achieved that age and the corresponding relation of face, thus carry out estimation of Age, thus detect the information at the sex of people, age.And then the number of its effective viewing or reading advertisement is carried out the statistics at a point sex, age.Utilize face recognition algorithms to carry out recognition of face, the method such as such as artificial neural network method, deep approach of learning, Eigenface method, Fisherface, then extract human face region image；Haar feature extraction algorithm and adaboost sorting algorithm is used to detect human eye and nose in its region extracted.
Combine viewing, reading advertisement duration further according to personnel's attention evaluation algorithm, analyze personnel's interest level information to its advertisement.Degree correlation analysis by the deflection angle of face Yu personnel's attention, i.e. when personnel's content situation interested to advertisement, before advertisement display terminal can be rested on, and it is now different, towards difference because of face location, cause occurring that deflection angle is different from advertisement display terminal, then the attention degree shown is different.Utilize the deflection angle of face to judge personnel's attention situation, be segmented into following three types and judge, first the eyes in image and nose are carried out interconnecting line, then constitute the figure of inverted triangular shape；When in the image collected, people two arrives nose respectively apart from equal or close, institute's triangle shape of institute in the picture is similar for isosceles triangle, then be judged as the positive face of face, i.e. show as facing advertisement display terminal, and attention is concentrated, and degree is higher；When the distance difference that in the image collected, two are arrived nose respectively more greatly or not etc. time, the most triangular shaped deform upon, can judge in two kinds of situation, first, when the distance of left eye to nose is compared with right eye to nose, the distance of the del left side length of side constituted is shorter than the right, it is judged as the left face of face, i.e. showing and see towards the advertisement display terminal left side, attention is not good enough, and degree is relatively low；Second, when the distance of left eye to nose is compared with right eye to nose, the distance of the del left side length of side constituted is longer than the right, it is judged that for the right face of face, i.e. showing and see on the right of advertisement display terminal, attention is not good enough, and degree is relatively low；Personnel's attention judges in addition to the judgement of the deflection angle of facial orientation, also relevant with the relative position of advertisement display terminal with personnel, when the facial orientation left side or the right, people is in the right or the left side of advertisement display terminal, and the attention situation showed also differs, such as, be in the right side of advertisement display terminal as personnel, if face is seen to the left side, attention degree is higher, if face is seen to the right, attention degree is relatively low；Otherwise, personnel positions is contrary, and facial orientation is contrary, then personnel's attention degree situation is corresponding on the contrary.
Last total score separates out in different time, different location, the different classes of personnel data effectively watched, read to the certain type of advertisement of its concrete input, it is possible to data uploaded, so that the actual demand of advertising business is understood by advertiser；The intelligence jettison system image information by one or more image collection assemblies, carries out individually or coordinates to analyze, and can be used for recognizer and the judgement of personnel's interest level, the algorithm of man-machine interaction of personnel's sex, age and identity category；Wherein image collection assembly is due to environment difference, and intensity of illumination is different, needs to carry out parameter calibration；Signal processing unit can to advertisement putting particular type and quantity, personnel's viewing or reading time situation, throw in the number of times of different advertisements according to different crowd type, the data such as duration that advertisement is play, frequency number of times are added up, such as, add up all kinds of personnel in effectively the watching of different regions, reading amount；Add up certain class personnel effectively watch in certain time period, reading amount；Add up certain class personnel effectively viewing, reading amount in certain area；Add up certain all kinds of personnel effectively viewing, reading amount in a certain period；Be analyzed by these statistical information, thus different regions, different time and different classes of personnel are effectively pinpointed, regularly, orientation precisely throw in advertisement.
The above is only the preferred embodiment of the present invention; it should be pointed out that, for those skilled in the art, under the premise without departing from the principles of the invention; can also make some improvements and modifications, these improvements and modifications also should be regarded as protection scope of the present invention.
Claims (10)
1. personnel's interest level detecting system based on machine vision, it is characterised in that include
Signal gathering unit, the signal message of personnel in gathering target area；
Signal processing unit, is connected with described signal gathering unit, is used for processing described signal gathering unit and adopts
The personal information of collection, and personal information is classified by the interest level of different advertisements；
Described signal processing unit includes:
Personnel's demographics unit, utilizes the personal information that described signal gathering unit gathers, and statistics is through intelligence
Personnel's number of energy jettison system；
Face identification unit, for identifying the face information in described personal information, and detects each personnel
Human eye and nose information；
Personnel's attention judging unit, utilizes human eye and the nose information of personnel, it is judged that the deflection of personnel's face
Angle, determines whether personnel's interest level to different information.
Personnel's interest level detecting system based on machine vision the most according to claim 1, it is special
Levying and be, described signal gathering unit includes image collecting device, the figure of personnel in gathering target area
Image signal, described image collecting device is arranged on shown information and throws on unit.
Personnel's interest level detecting system based on machine vision the most according to claim 2, it is special
Levying and be, described image collector is set to photographic instrument, visible image capturing instrument, infrared photography instrument and heat
At least one in imaging camera instrument.
Personnel's interest level detecting system based on machine vision the most according to claim 2, it is special
Levying and be, described signal gathering unit also includes audio signal sample device, is used for gathering people in target area
The audio signal of member, described audio signal sample device is arranged on shown information and throws on unit.
Personnel's interest level detecting system based on machine vision the most according to claim 1, it is special
Levying and be, described personnel's attention judging unit specifically includes:
Human eye according to personnel and nose information, build the human eye by personnel and nose interconnecting line, composition
Del, it is assumed that personnel are positioned at the front of intelligence jettison system, when people two arrives nose respectively apart from equal
Or close, described del is isosceles triangle, then judge that personnel face intelligence jettison system, attention
Concentrate, high to the interest level of information；
When the distance difference that people two arrives nose respectively is bigger, then judge personnel's side-looking intelligence jettison system,
Absent minded, low to the interest level of information.
Personnel's interest level detecting system based on machine vision the most according to claim 1, it is special
Levying and be, described personnel's attention judging unit includes personnel's visual line of sight computing unit:
Before assuming that personnel are positioned at intelligence jettison system, facial orientation is for facing, during sight line zero deflection angle, and people
Intrinsic sight angleThe intelligence a length of L of jettison system, personnel and distance d of intelligence jettison system, then regard
Line scope
Before assuming that personnel are positioned at intelligence jettison system, facial orientation is for facing, during sight line deflection angle θ, and people
Intrinsic sight angleThe intelligence a length of L of jettison system, personnel and distance d of intelligence jettison system, then regard
Line scope
Assume personnel with intelligence jettison system have relative sidesway Δ x, facial orientation for facing, sight line deflection angle
During degree θ, the sight angle that people is intrinsicIntelligence a length of L of jettison system, personnel with intelligence jettison system away from
From d, then visual line of sight
Wherein, the most closely or too far, visual line of sight is the most invalid, i.e. with d as district for personnel's distance intelligence jettison system
Boundary, d ∈ (dmin,dmax) time effective.
Personnel's interest level detecting system based on machine vision the most according to claim 6, it is special
Levying and be, described personnel's attention judging unit also includes personnel's interest level computing unit；Personnel feel emerging
Interest degree i.e. visual line of sight is proportion in intelligence jettison system lengthCarry out n sampling, then feel
Level of interest P is expressed as:
When P is more than or equal to predetermined threshold value, then judge that personnel are interested；When P is little
When predetermined threshold value, then judge that personnel lose interest in.
Personnel's interest level based on machine vision the most according to any one of claim 1 to 7 is examined
Examining system, it is characterised in that described signal processing unit also includes:
Personnel's sex recognition unit, the sex of personnel in gathering target area；And/or
Personnel's age recognition unit, the age of personnel in gathering target area；And/or
Personnel identity classification recognition unit, the identity category of personnel in gathering target area；And/or
Personnel's skin color model unit, the colour of skin of personnel in gathering target area.
9. personnel's interest level detection method based on machine vision, it is characterised in that include following
Step:
The signal message of personnel in collection target area；
Process the personal information gathered, and personal information is classified by the interest level of different advertisements；
Process the personal information gathered to specifically include:
Statistics is through personnel's number of intelligence jettison system；
Identify the face information in described personal information, and detect human eye and the nose information of each personnel；
Judgement personnel's attention, utilizes human eye and the nose information of personnel, it is judged that the deflection angle of personnel's face,
Determine whether personnel's interest level to different information.
Personnel's interest level detection method based on machine vision the most according to claim 9, its
Being characterised by, described judgement personnel's attention step specifically includes:
Human eye according to personnel and nose information, build the human eye by personnel and nose interconnecting line, composition
Del, it is assumed that personnel are positioned at the front of intelligence jettison system, when people two arrives nose respectively apart from equal
Or close, described del is isosceles triangle, then judge that personnel face intelligence jettison system, attention
Concentrate, high to the interest level of information；
When the distance difference that people two arrives nose respectively is bigger, then judge personnel's side-looking intelligence jettison system,
Absent minded, low to the interest level of information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610250120.6A CN105938603A (en) | 2016-04-20 | 2016-04-20 | Personnel interest degree detection system based on machine vision and personnel interest degree detection method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610250120.6A CN105938603A (en) | 2016-04-20 | 2016-04-20 | Personnel interest degree detection system based on machine vision and personnel interest degree detection method thereof |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105938603A true CN105938603A (en) | 2016-09-14 |
Family
ID=57151509
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610250120.6A Pending CN105938603A (en) | 2016-04-20 | 2016-04-20 | Personnel interest degree detection system based on machine vision and personnel interest degree detection method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105938603A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106778676A (en) * | 2016-12-31 | 2017-05-31 | 中南大学 | A kind of notice appraisal procedure based on recognition of face and image procossing |
| CN108875678A (en) * | 2018-06-29 | 2018-11-23 | 上海小蚁科技有限公司 | The concern number of subject matter determines method and device, storage medium, terminal |
| CN108921829A (en) * | 2018-06-20 | 2018-11-30 | 广州晖恒广告策划有限公司 | A kind of advertisement design method for objectively evaluating of view-based access control model attention mechanism |
| CN109767273A (en) * | 2019-04-10 | 2019-05-17 | 深兰人工智能芯片研究院（江苏）有限公司 | A kind of advertisement sending method, device, advertisement playback terminal and storage medium |
| CN111160945A (en) * | 2019-02-27 | 2020-05-15 | 浙江诸暨美数信息科技有限公司 | Advertisement accurate delivery method and system based on video technology |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101335864A (en) * | 2007-06-28 | 2008-12-31 | 当代天启技术（北京）有限公司 | Method and system for number of outdoor video receiving people statistic |
| CN102446270A (en) * | 2010-10-15 | 2012-05-09 | 汉王科技股份有限公司 | Monitoring device and method based on face recognition |
| CN102622596A (en) * | 2012-02-20 | 2012-08-01 | 哈尔滨工业大学 | Interactive positioning method based on biometric recognition |
| CN202472689U (en) * | 2012-03-06 | 2012-10-03 | 上海骏聿数码科技有限公司 | Face posture detection system |
| CN102881239A (en) * | 2011-07-15 | 2013-01-16 | 鼎亿数码科技(上海)有限公司 | Advertisement playing system and method based on image identification |
| CN103300815A (en) * | 2012-03-15 | 2013-09-18 | 凹凸电子（武汉）有限公司 | Eyeball focus determination method, device and system |
| CN103996033A (en) * | 2014-05-29 | 2014-08-20 | 厦门瑞为信息技术有限公司 | Human identification method based on tracking of human face five-sense-organ coordinates |
-
2016
- 2016-04-20 CN CN201610250120.6A patent/CN105938603A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101335864A (en) * | 2007-06-28 | 2008-12-31 | 当代天启技术（北京）有限公司 | Method and system for number of outdoor video receiving people statistic |
| CN102446270A (en) * | 2010-10-15 | 2012-05-09 | 汉王科技股份有限公司 | Monitoring device and method based on face recognition |
| CN102881239A (en) * | 2011-07-15 | 2013-01-16 | 鼎亿数码科技(上海)有限公司 | Advertisement playing system and method based on image identification |
| CN102622596A (en) * | 2012-02-20 | 2012-08-01 | 哈尔滨工业大学 | Interactive positioning method based on biometric recognition |
| CN202472689U (en) * | 2012-03-06 | 2012-10-03 | 上海骏聿数码科技有限公司 | Face posture detection system |
| CN103300815A (en) * | 2012-03-15 | 2013-09-18 | 凹凸电子（武汉）有限公司 | Eyeball focus determination method, device and system |
| CN103996033A (en) * | 2014-05-29 | 2014-08-20 | 厦门瑞为信息技术有限公司 | Human identification method based on tracking of human face five-sense-organ coordinates |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106778676A (en) * | 2016-12-31 | 2017-05-31 | 中南大学 | A kind of notice appraisal procedure based on recognition of face and image procossing |
| CN106778676B (en) * | 2016-12-31 | 2020-02-18 | 中南大学 | Attention assessment method based on face recognition and image processing |
| CN108921829A (en) * | 2018-06-20 | 2018-11-30 | 广州晖恒广告策划有限公司 | A kind of advertisement design method for objectively evaluating of view-based access control model attention mechanism |
| CN108875678A (en) * | 2018-06-29 | 2018-11-23 | 上海小蚁科技有限公司 | The concern number of subject matter determines method and device, storage medium, terminal |
| CN111160945A (en) * | 2019-02-27 | 2020-05-15 | 浙江诸暨美数信息科技有限公司 | Advertisement accurate delivery method and system based on video technology |
| CN109767273A (en) * | 2019-04-10 | 2019-05-17 | 深兰人工智能芯片研究院（江苏）有限公司 | A kind of advertisement sending method, device, advertisement playback terminal and storage medium |
| CN109767273B (en) * | 2019-04-10 | 2019-07-16 | 深兰人工智能芯片研究院（江苏）有限公司 | A kind of advertisement sending method, device, advertisement playback terminal and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105938603A (en) | Personnel interest degree detection system based on machine vision and personnel interest degree detection method thereof | |
| US8401248B1 (en) | Method and system for measuring emotional and attentional response to dynamic digital media content | |
| CN106294489B (en) | Content recommendation method, device and system | |
| US20190340780A1 (en) | Engagement value processing system and engagement value processing apparatus | |
| Liciotti et al. | Person re-identification dataset with rgb-d camera in a top-view configuration | |
| US8462996B2 (en) | Method and system for measuring human response to visual stimulus based on changes in facial expression | |
| US9443144B2 (en) | Methods and systems for measuring group behavior | |
| CN106557937B (en) | Advertisement pushing method and device | |
| US20120140069A1 (en) | Systems and methods for gathering viewership statistics and providing viewer-driven mass media content | |
| JP4603974B2 (en) | Content attention evaluation apparatus and content attention evaluation method | |
| US20160191995A1 (en) | Image analysis for attendance query evaluation | |
| WO2021031600A1 (en) | Data collection method and apparatus, computer device, and storage medium | |
| CN102930454A (en) | Intelligent 3D (Three Dimensional) advertisement recommendation method based on multiple perception technologies | |
| CN106296307A (en) | Electronic stop plate advertisement delivery effect based on recognition of face analyzes method | |
| CN102881239A (en) | Advertisement playing system and method based on image identification | |
| CN102081918A (en) | Video image display control method and video image display device | |
| CN101588443A (en) | Statistical device and detection method for television audience ratings based on human face | |
| TW201033908A (en) | System and method for counting people flow | |
| CN103914691A (en) | Target group analysis system and method based on face recognition and height recognition method | |
| JP6498900B2 (en) | Advertisement evaluation system, advertisement evaluation method | |
| CN108694627A (en) | Moving advertising show robot system | |
| CN110889719A (en) | Advertisement pushing system and method based on user consumption characteristics | |
| Yu et al. | AI-based targeted advertising system | |
| CN102368295A (en) | Method for extracting age characteristics under complex environment | |
| CN114746882A (en) | Systems and methods for interaction awareness and content presentation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160914 |
|
| RJ01 | Rejection of invention patent application after publication |