CN111274884A - Intelligent advertisement pushing system based on integration of face recognition and behavior recognition - Google Patents
Intelligent advertisement pushing system based on integration of face recognition and behavior recognition Download PDFInfo
- Publication number
- CN111274884A CN111274884A CN202010030490.5A CN202010030490A CN111274884A CN 111274884 A CN111274884 A CN 111274884A CN 202010030490 A CN202010030490 A CN 202010030490A CN 111274884 A CN111274884 A CN 111274884A
- Authority
- CN
- China
- Prior art keywords
- advertisement
- behavior
- face
- video
- module
- 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
Images
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
- G06Q30/0251—Targeted advertisements
-
- 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
- G06Q30/0251—Targeted advertisements
- G06Q30/0269—Targeted advertisements based on user profile or attribute
-
- 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
- G06V40/20—Movements or behaviour, e.g. gesture recognition
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Game Theory and Decision Science (AREA)
- Psychiatry (AREA)
- Social Psychology (AREA)
- Image Analysis (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Computer Vision & Pattern Recognition (AREA)
Abstract
The invention relates to an intelligent advertisement pushing system based on face recognition and behavior recognition, which comprises the following modules: the system comprises a video acquisition module, a face detection module, a face feature analysis module, a video stream coding and decoding module, a video stream forwarding module, an advertisement material management module, an advertisement playing interaction module and a behavior analysis module; the corresponding background services are: the system comprises a behavior analysis server, a behavior model management background, an advertisement material management background service and a database service which are responsible for behavior analysis; the corresponding client side comprises: an advertiser app installed on the advertiser. Compared with the method of simply pushing the advertisement by age and gender, the method is more targeted.
Description
Technical Field
The invention relates to the field of face recognition, behavior recognition and video monitoring, in particular to an intelligent advertisement pushing system based on the integration of face recognition and behavior recognition.
Background
Most of the advertisement machines in the current market randomly play advertisement video contents prestored in the advertisement machines, and are single and have no pertinence. This may result in that the advertiser or the store does not achieve the desired advertisement delivery effect, and effective feedback information cannot be obtained, thereby increasing the advertisement delivery cost. For the situation, a relatively large number of solutions in the market at present integrate a face recognition detection algorithm with an advertising machine, then analyze attributes such as gender, age, expression and the like of a face by detecting the face, and then push related advertisements. However, this solution has its drawbacks, first: the distance between the human face and the advertising machine is required to be not too far, the human face is difficult to detect when the distance exceeds 1 m in practice, and secondly: the recognition of the facial expression generally requires a relatively clear frontal face, and thus the angle requirement on the face is relatively high. Thirdly, the method comprises the following steps: the scheme of pushing advertisements by light from age, gender and expression is also less targeted. It is difficult to meet the accurate delivery of advertisements in different scenes.
Object of the Invention
The invention aims to provide an intelligent advertisement pushing system based on face recognition and behavior recognition, which mainly comprises the following modules: the system comprises a video acquisition module, a face detection module, a face feature analysis module, a video stream coding and decoding module, a video stream forwarding module, an advertisement material management module, an advertisement playing interaction module and a behavior analysis module; the corresponding background services are: the system comprises a behavior analysis server, a behavior model management background, an advertisement material management background service and a database service which are responsible for behavior analysis; the corresponding client side comprises: an advertisement machine app installed on the advertisement machine; wherein
The advertising machine is provided with a built-in camera;
the advertisement player is provided with an app client of a software system, and the app client software is used for face recognition, encoding and sending video data collected by a camera and playing advertisements;
the advertisement management background service and the database service are used for managing advertisement materials and can add, delete and modify the advertisement materials;
the behavior model management background is responsible for managing and issuing various behavior models;
the behavior analysis server is mainly based on video analysis, has higher performance on the server, and is used for analyzing the video stream uploaded by the client of the advertising player in real time and returning the identification result to the web background;
the advertisement management background is used for storing advertisement material content; the database is a MySql database and is used for storing relevant data information of other advertising machines; the management background service is an http-based service;
the behavior analysis server is based on a tcp network protocol, the advertiser client can establish long connection with the behavior analysis server, and the behavior analysis server can pull different trained behavior models from a behavior model management background in advance;
the face recognition algorithm adopts a face-reading face algorithm sdk, the sdk provides face detection, face tracking and face recognition api interfaces, and the sdk can return face-related information such as age, gender and the like;
the behavior recognition algorithm adopts R-C3D (local 3D convolution neural network), R-C3D extracts video features by using 3D convolution, and adopts the idea of fast-RCNN form, namely, proposal is generated firstly, then roi-posing is performed, finally classification and boundary regression are performed, and the open source project caffe is used by network reasoning;
the resolution ratio collected by the built-in camera is 1280 x 720, the collected video format is yuv, the yuv video is mainly used for two places, one is used for face detection, the other is used for coding transmission and behavior analysis, the coding format adopts H264, and the coding mode uses hard coding of an advertising machine.
Furthermore, the invention relates to an intelligent advertisement pushing method based on the integration of face recognition and behavior recognition, which comprises the following steps:
(1) the advertisement player is provided with an app application;
(2) the client of the advertisement player pulls the advertisement content of the advertisement player from the background service of the advertisement player management through an http request and stores the advertisement content on a storage medium of the advertisement player;
(3) the client of the advertising player starts the built-in camera to start to collect the video;
(4) the advertiser client will start the face detection service and the video coding service;
(5) the face detection service can carry out face detection on the video acquired by the camera frame by frame;
(6) if the face is detected, extracting face characteristic value information, age and gender;
(7) the advertisement player finds out the corresponding advertisement ID according to the age and gender information;
(8) the advertisement player judges the current advertisement playing state, if the current state is playing, the detected advertisement ID is deleted, no processing is carried out, the steps 4-8 are repeated, and if the current state is idle, the corresponding advertisement is played;
(9) the video coding service codes videos collected by the camera in real time and sends the coded video data to the behavior analysis center server, and the behavior analysis center server sends the videos to each behavior model server;
(10) each behavior model server reports the behavior analysis result to a behavior analysis center server, and the behavior analysis center server reports the behavior result ID to a background web server;
(11) the background web server issues the advertisement ID corresponding to the behavior analysis result ID to the advertisement machine;
(12) the advertisement machine judges according to the current advertisement playing state, if the current state is playing, the detected advertisement ID is deleted, no processing is carried out, and the advertisement machine continues to wait for the next advertisement ID issued by the web server; and if the current state is idle, playing the advertisement corresponding to the advertisement ID.
Technical effects
Compared with the prior art, the invention has the technical effects that:
1. aiming at the problems that the existing advertisement putting is not targeted and the limitation of the existing technical scheme of pushing the advertisement by simply accessing face recognition, the invention intelligently pushes the advertisement by combining the face recognition and the human behavior recognition.
2. The behavior of the person contains a lot of useful information, such as the preference and habit of the person, and the information is more targeted than simply pushing the advertisement by age and gender.
3. In practice, most scenes are that people are away from an advertising machine for a certain distance, so that the people are difficult to capture by simply using face detection, and the aim of targeted pushing is not achieved in 70% -80% of the time of the advertising machine; but the requirement of human behavior recognition on the distance is not so high, and the targeted analysis can be carried out as long as the action behaviors of the user are captured within the visual angle range of the camera of the advertising machine, so that the utilization rate of the advertising machine can be greatly improved.
Drawings
Fig. 1 is a schematic diagram of an intelligent advertising player pushing system based on integration of face recognition and behavior recognition.
Fig. 2 is a flow chart of the intelligent advertising player pushing system based on the integration of face recognition and behavior recognition.
Detailed Description
The following further describes the embodiments of the present invention with reference to the drawings:
referring to fig. 1, an intelligent advertisement push system based on face recognition and behavior recognition, includes:
1. the vertical advertising machine based on the android system is provided with a built-in camera;
2. the android client running on the advertising machine is used for recognizing human faces, coding and sending video data acquired by the camera, and playing advertisements;
3. the advertisement management background service is used for managing advertisement materials and can add, delete and modify the advertisement materials;
4. the behavior model management background is responsible for managing and issuing various behavior models;
5. the behavior analysis server is mainly based on video analysis and has higher performance on the server, so that the server with the GPU is adopted for analyzing the video stream uploaded by the client of the advertising machine in real time and returning the identification result to the web background;
6. in the embodiment, the advertisement machine mainboard is an android version of optoetat 3288A, the android version is 5.1.1, the built-in resolution of the camera is 1920 × 1080, and in the invention, the resolution acquired by the camera is 1280 × 720;
7. in the embodiment, the advertisement management background mainly uses the object storage of the Tencent cloud for storing the content of the advertisement material; the database is a MySql database used for storing relevant data information of other advertising machines; the management background service is mainly based on http;
8. in this embodiment, the behavior analysis server is mainly based on a tcp network protocol, the advertiser client establishes a long connection with the behavior analysis server, and the behavior analysis server pulls different trained behavior models from a behavior model management background in advance;
9. in this embodiment, the face recognition algorithm mainly uses a face-reading face algorithm sdk, the sdk provides a face detection interface, a face tracking interface, and a face recognition api interface, and the sdk can return face-related information such as age, gender, and the like;
10. in the embodiment, the behavior recognition algorithm mainly uses R-C3D (local 3D convolution neural network), R-C3D uses 3D convolution to extract video features, and the idea of the fast-RCNN form is adopted, namely proposal is generated firstly, then roi-posing is performed, finally classification and boundary regression are performed, and the open source project caffe is used for network reasoning.
10. In this embodiment, the resolution of the built-in camera acquisition is 1280 × 720, the acquired video format is yuv, the yuv video is mainly used in two places, one is used for face detection, the other is used for encoding transmission and behavior analysis, the encoding format adopts H264, and the encoding mode uses hard encoding of an advertising machine;
referring to fig. 1 and 2, the specific process of the intelligent advertisement push system based on the integration of face recognition and behavior recognition of the present invention is as follows:
1. the advertisement player is powered on, and app applications installed on the advertisement player are started;
2. the client of the advertising machine can pull the advertising content of the advertising machine from the advertising machine management background service through an http request and store the advertising content on a storage medium of the advertising machine;
3. the client of the advertising player starts the built-in camera to start to collect the video;
4. the advertising player client can start a face detection service and a video coding service;
5. the face detection service can carry out face detection on the video acquired by the camera frame by frame;
6. if the face is detected, extracting face characteristic value information, age and gender;
7. the advertising machine finds out the corresponding advertising ID according to the age and gender information;
8. the advertisement player judges the current advertisement playing state, if the current state is playing, the detected advertisement ID is deleted, no processing is carried out, the step 4-8 is repeated, and if the current state is idle, the corresponding advertisement is played;
9. the video coding service can code videos collected by the camera in real time and send the coded video data to the behavior analysis center server, and the behavior analysis center server can send the videos to each behavior model server;
10. each behavior model server reports the behavior analysis result to a behavior analysis center server, and the behavior analysis center server reports the behavior result ID to a background web server;
11. the background web server issues the advertisement ID corresponding to the behavior analysis result ID to the advertisement machine;
12. the advertisement player judges according to the current advertisement playing state, if the current state is playing, the detected advertisement ID is deleted, no processing is carried out, and the advertisement player continues to wait for the next advertisement ID issued by the web server; if the current state is idle, the advertisement corresponding to the advertisement ID is played.
Claims (4)
1. The utility model provides an intelligent advertisement push system based on face identification and action recognition which characterized in that, the module that contains has: the system comprises a video acquisition module, a face detection module, a face feature analysis module, a video stream coding and decoding module, a video stream forwarding module, an advertisement material management module, an advertisement playing interaction module and a behavior analysis module; the corresponding background services are: the system comprises a behavior analysis server, a behavior model management background, an advertisement material management background service and a database service which are responsible for behavior analysis; the corresponding client side comprises: an advertisement machine app installed on the advertisement machine; wherein
The advertising machine is provided with a built-in camera;
the advertisement player is provided with an app client of a software system, and the app client software is used for face recognition, encoding and sending video data collected by a camera and playing advertisements;
the advertisement management background service and the database service are used for managing advertisement materials and can add, delete and modify the advertisement materials;
the behavior model management background is responsible for managing and issuing various behavior models;
the behavior analysis server is based on video analysis, has higher performance on the server, and is used for analyzing the video stream uploaded by the client of the advertising player in real time and returning the identification result to the web background;
the advertisement management background is used for storing advertisement material content; the database is a MySql database and is used for storing relevant data information of other advertising machines; the management background service is an http-based service;
the behavior analysis server is based on a tcp network protocol, the advertiser client can establish long connection with the behavior analysis server, and the behavior analysis server can pull different trained behavior models from a behavior model management background in advance;
the face recognition algorithm adopts a face-reading face algorithm sdk, the sdk provides face detection, face tracking and face recognition api interfaces, and the sdk can return face related information;
the behavior recognition algorithm adopts R-C3D, R-C3D extracts video features by using 3D convolution, and adopts the idea of a fast-RCNN form, namely, proposal is generated firstly, then roi-posing is performed, finally classification and boundary regression are performed, and the open source project caffe is used by network reasoning;
and the video collected by the built-in camera is used for face detection, encoding and sending to behavior analysis.
2. The intelligent advertisement pushing system based on face recognition and behavior recognition is characterized in that the resolution of the built-in camera acquisition is 1280 x 720, and the acquired video format is yuv.
3. The intelligent advertisement delivery system based on face recognition and behavior recognition as claimed in claim 1, wherein the video coding format is H264, and the coding mode uses hard coding of an advertisement machine.
4. An intelligent advertisement pushing method based on integration of face recognition and behavior recognition comprises the following steps:
(1) the advertisement player is provided with an app application;
(2) the client of the advertisement player pulls the advertisement content of the advertisement player from the background service of the advertisement player management through an http request and stores the advertisement content on a storage medium of the advertisement player;
(3) the client of the advertising player starts the built-in camera to start to collect the video;
(4) the advertiser client will start the face detection service and the video coding service;
(5) the face detection service can carry out face detection on the video acquired by the camera frame by frame;
(6) if the face is detected, extracting face characteristic value information, age and gender;
(7) the advertisement player finds out the corresponding advertisement ID according to the age and gender information;
(8) judging the current advertisement playing state by the advertisement player, if the current state is playing, deleting the detected advertisement ID, not performing any processing, repeating the steps 4-8, and if the current state is idle, playing the corresponding advertisement;
(9) the video coding service codes videos collected by the camera in real time and sends the coded video data to the behavior analysis center server, and the behavior analysis center server sends the videos to each behavior model server;
(10) each behavior model server reports the behavior analysis result to a behavior analysis center server, and the behavior analysis center server reports the behavior result ID to a background web server;
(11) the background web server issues the advertisement ID corresponding to the behavior analysis result ID to the advertisement machine;
(12) the advertisement player judges according to the current advertisement playing state, if the current state is playing, the detected advertisement ID is deleted, no processing is carried out, and the advertisement player continues to wait for the next advertisement ID issued by the web server; and if the current state is idle, playing the advertisement corresponding to the advertisement ID.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010030490.5A CN111274884A (en) | 2020-01-11 | 2020-01-11 | Intelligent advertisement pushing system based on integration of face recognition and behavior recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010030490.5A CN111274884A (en) | 2020-01-11 | 2020-01-11 | Intelligent advertisement pushing system based on integration of face recognition and behavior recognition |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111274884A true CN111274884A (en) | 2020-06-12 |
Family
ID=70997063
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010030490.5A Pending CN111274884A (en) | 2020-01-11 | 2020-01-11 | Intelligent advertisement pushing system based on integration of face recognition and behavior recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111274884A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112184314A (en) * | 2020-09-29 | 2021-01-05 | 福州东方智慧网络科技有限公司 | Popularization method based on equipment side visual interaction |
| CN112256969A (en) * | 2020-10-28 | 2021-01-22 | 北京字节跳动网络技术有限公司 | Media material pushing method and device, electronic equipment and storage medium |
| CN113591821A (en) * | 2021-10-08 | 2021-11-02 | 广州洛克韦陀安防科技有限公司 | Image identification security system based on big data screening |
| CN114125026A (en) * | 2021-11-24 | 2022-03-01 | 四川长虹电器股份有限公司 | Communication method for advertising player and advertising player |
-
2020
- 2020-01-11 CN CN202010030490.5A patent/CN111274884A/en active Pending
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112184314A (en) * | 2020-09-29 | 2021-01-05 | 福州东方智慧网络科技有限公司 | Popularization method based on equipment side visual interaction |
| CN112256969A (en) * | 2020-10-28 | 2021-01-22 | 北京字节跳动网络技术有限公司 | Media material pushing method and device, electronic equipment and storage medium |
| CN112256969B (en) * | 2020-10-28 | 2024-04-26 | 抖音视界有限公司 | Media material pushing method and device, electronic equipment and storage medium |
| CN113591821A (en) * | 2021-10-08 | 2021-11-02 | 广州洛克韦陀安防科技有限公司 | Image identification security system based on big data screening |
| CN113591821B (en) * | 2021-10-08 | 2021-12-10 | 广州洛克韦陀安防科技有限公司 | Image identification security system based on big data screening |
| CN114125026A (en) * | 2021-11-24 | 2022-03-01 | 四川长虹电器股份有限公司 | Communication method for advertising player and advertising player |
| CN114125026B (en) * | 2021-11-24 | 2023-09-05 | 四川长虹电器股份有限公司 | Communication method for advertising machine and advertising machine |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111274884A (en) | Intelligent advertisement pushing system based on integration of face recognition and behavior recognition | |
| CN109377293B (en) | Advertisement putting effect monitoring method and system | |
| US8503523B2 (en) | Forming a representation of a video item and use thereof | |
| CN109934625A (en) | A kind of artificial intelligence display screen advertisement dynamic throwing system and method | |
| CN101420595B (en) | Method and equipment for describing and capturing video object | |
| CN108985845A (en) | Broadcast monitoring method, apparatus and system | |
| CN112347941B (en) | Motion video collection intelligent generation and distribution method based on 5G MEC | |
| CN110633669A (en) | Mobile terminal face attribute identification method based on deep learning in home environment | |
| JP2012108916A (en) | Data mining for specifying reaction of online user to broadcast message | |
| CN109543575B (en) | Method for feeding back excitation big data based on behavior track and dynamic multi-dimensional information and regional multi-dimensional detection feedback combined equipment | |
| CN202998337U (en) | Video program identification system | |
| CN114679607B (en) | Video frame rate control method and device, electronic equipment and storage medium | |
| CN105160550A (en) | Intelligent advertisement delivery method and apparatus | |
| CN113393275B (en) | Intelligent medium management system based on VOC (volatile organic compound) vehicle owner big data platform | |
| CN110599129A (en) | Campus attendance checking method, device, identification terminal and system based on image tracking | |
| CN103187083B (en) | A kind of storage means based on time domain video fusion and system thereof | |
| CN112101304A (en) | Data processing method, device, storage medium and equipment | |
| CN104766230A (en) | Advertising effect evaluation method based on human skeletal tracking | |
| CN111724199A (en) | Intelligent community advertisement accurate delivery method and device based on pedestrian active perception | |
| CN110866553A (en) | User behavior classification method and system based on encrypted camera flow statistical characteristics | |
| CN113377327A (en) | Huge curtain MAX intelligent terminal in garage that possesses intelligent voice interaction function | |
| CN107766796A (en) | A kind of facial-recognition security systems and method based on cloud computing | |
| CN111191498A (en) | Behavior recognition method and related product | |
| CN110634035B (en) | Face recognition popup advertisement system | |
| JP2003319421A (en) | Image management method, device, image management program and recording medium with the program recorded thereon |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |