CN104484651B - Portrait dynamic contrast method and system - Google Patents
Portrait dynamic contrast method and system Download PDFInfo
- Publication number
- CN104484651B CN104484651B CN201410763088.2A CN201410763088A CN104484651B CN 104484651 B CN104484651 B CN 104484651B CN 201410763088 A CN201410763088 A CN 201410763088A CN 104484651 B CN104484651 B CN 104484651B
- Authority
- CN
- China
- Prior art keywords
- video
- portrait
- information
- database
- server
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
- G06V20/53—Recognition of crowd images, e.g. recognition of crowd congestion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
-
- 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
- G06V40/18—Eye characteristics, e.g. of the iris
- G06V40/193—Preprocessing; Feature extraction
-
- 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
- G06V40/18—Eye characteristics, e.g. of the iris
- G06V40/197—Matching; Classification
Abstract
The present invention relates to portrait correlation technique field, particularly a kind of portrait dynamic contrast method gathers the portrait video of each site by video acquisition device;Then after the portrait video compress collected each site by streaming media server streaming media video file is formed to store;The real-time portrait of video compares server by accessing streaming media server, obtains the camera video of Identification of Images, carries out portrait in graphical analysis, with database and is compared in real time;Then judge whether similarity meets, personal information in database is then carried out if satisfaction and is inquired about.Using the above method, the present invention is by setting streaming media server preferably to obtain dynamic video information, so as to further obtain figure information from video information, carries out portrait comparison;On the other hand, database portrait can be upgraded in time according to video daily record and critical event situation by streaming media server and database update server.
Description
Technical field
The present invention relates to portrait correlation technique field, particularly a kind of portrait dynamic contrast method and system.
Background technology
It can be very good to recognize piece identity by Identification of Images contrast, many fields are all used for portrait at this stage
Identifying system, for identification to improve safety.But be due to daily video information content it is larger, it is impossible to accomplish to adopt well
Collection;In addition, it is necessary to extract portrait data, and portrait data in database from database when carrying out Identification of Images contrast
Need timely to update and adjust.
The A of Chinese invention patent CN 102682283 disclose a kind of dynamic face recognition system, including real-time interconnection monitoring
Identifying system, non real-time interconnected monitoring system and real-time interconnection monitoring application module system, real-time interconnection monitoring identifying system bag
Comprehensive monitoring platform and application platform are included, the input of comprehensive monitoring platform is connected with front end human image collecting server, comprehensive prison
Control platform output end is connected with day net video monitoring site and social unit video monitoring site, the input connection of application platform
There is Identification of Images to compare server and background data base server, the output end of application platform is connected with portrait search comparison subsystem
System, data maintenance Query Subsystem and alarm center, background data base server input are connected with identification contrast portrait storehouse, people
It is connected as identification compares server with front end human image collecting server.
The content of the invention
The method for the portrait dynamic contrast that the technical problem to be solved in the invention speed is fast, comparative analysis effect is good and it is
System.
To solve above-mentioned technical problem, the invention discloses a kind of portrait dynamic contrast method, comprise the following steps,
Step S101:The portrait video of each site is gathered by video acquisition device;
Step S102:After the portrait video compress for each site being collected by streaming media server form streaming media video
File is stored;
Step S103:The real-time portrait of video compares server by accessing streaming media server, obtains Identification of Images
Camera video, carries out portrait in graphical analysis, with database and is compared in real time;
Step S104:Judge whether similarity meets, enter step S105 if similarity is met;If similarity is not
Meet and then enter step S106;
Step S105:Personal information is inquired about in database;
Step S106:Terminate.
Further, the step S103 is further comprising the steps of:
Step S1031:Video acquisition, the real-time portrait of video compares server by accessing streaming media server, is used
In the camera video of Identification of Images;
Step S1032:Corresponding facial image is obtained in face extraction, the camera video obtained from step S1031;
Step S1033:Feature extraction, obtains the face characteristic of corresponding site from the facial image of acquisition;
Step S1034:Similarity comparison, similarity pair is carried out by facial image feature in above-mentioned face characteristic and database
Than.
Further, files in stream media is additionally operable to the renewal of database figure information, including following step in the step S102
Suddenly:
S201:Video information is stored, and the video data information of files in stream media is stored in video information memory cell,
And flag bit setting is carried out to video information memory cell, it is then store in the scratchpad area (SPA) of database update server, institute
The sign position setting stated includes unique sign index of the sender of portrait video, and all sender with respectively as
The recipient of portrait video general unique sign index in the range of data delivery area, such sender is just by obtaining
Unique sign index convection media server of sender with portrait video exports claimant as the instruction of video, described
Claimant includes the unique sign index for the sender for carrying portrait video as the instruction of video;
S202:Information sifting is weighted, and the video information memory cell added with flag bit is weighted according to significance level, then
Weighting bit value is screened;
S203:Figure information is obtained, and the video data information in video information memory cell left to screening enters pedestrian
Face is extracted and feature extraction, obtains figure information;
S204:Figure information is stored, and it is single that the figure information of acquisition is put into figure information storage by database update server
In member;
S205:Database figure information is updated, and figure information memory cell in database update server is set newly
Flag bit, then will update into database added with the figure information memory cell of flag bit.
Further, the flag bit of the step S205 figure information memory cell is according to the qualitative classification of figure information
Setting.
The invention further relates to a kind of portrait dynamic contrast system, including each site video signal acquisition device, the video
Signal pickup assembly is connected with streaming media server, and the streaming media server compares server with the real-time portrait of video respectively
Be connected with database update server, the real-time portrait of the video compare server, database update server all with data
Storehouse is connected, and the real-time portrait of video compares server and is connected with application platform.
Further, the video acquisition device according to the difference of different positions and angle be divided into channel-style camera device and
Panorama camera device.
After the above method and structure, the present invention is by setting streaming media server preferably to obtain dynamic video
Information, so as to further obtain figure information from video information, carries out portrait comparison;On the other hand, Streaming Media can be passed through
Server and database update server upgrade in time according to video daily record and critical event situation to database portrait.
Brief description of the drawings
Below in conjunction with the drawings and specific embodiments to being originally described in further detail.
Fig. 1 a are the present inventor as the flow chart of dynamic contrast method.
Fig. 1 b are the flow chart that database figure information of the present invention updates.
Fig. 2 is the present inventor as the structured flowchart of dynamic contrast system.
Fig. 3 a are the structural representation of video information memory cell of the present invention.
Fig. 3 b are the structural representation of database purchase of the present invention.
In figure:1 is video signal acquisition device, and 2 be streaming media server, and 3 be that the real-time portrait of video compares server, 4
It is database update server for database, 5,6 be application platform, and 7 be video information memory cell
401 be portrait information memory cell, and 402 be data bit, and 403 be flag bit
701 be data bit, and 702 be summation of weighted bits, and 703 be flag bit
Embodiment
As shown in Fig. 2 a kind of portrait dynamic contrast system, including each site video signal acquisition device 1, the video letter
Number harvester 1 is connected with streaming media server 2, and video signal acquisition device 1 updates to streaming media server 2 in real time to be regarded
Frequency signal.Video signal acquisition device 1 uses camera in the present embodiment, is arranged on light rail transit), square, Internet bar's public place of entertainment
The sites such as institute, cell, supermarket are deployed to ensure effective monitoring and control of illegal activities region.In addition to cost-effective, the video acquisition device in present embodiment is pressed
Difference according to different positions and angle is divided into channel-style camera device and panorama camera device, in the flow of personnel such as square or cell
Nondirectional place uses full-view camera;The place of staircase or other staff's one-way flow is taken the photograph using channel-style above and below supermarket
As head.
The streaming media server 2 compares server 3 with the real-time portrait of video respectively and database update server 5 is connected
Connect, the real-time portrait of video compares server 3, database update server 5 and is all connected with database 4, and the video is real
When portrait compare server 3 be connected with application platform 6.The present inventor is as follows as the operation principle of dynamic contrast system:Flow matchmaker
Be put into after the video information compression that body server 2 can collect camera on server, user can with while download while watch,
So it is easy to the current intelligence of timely more new video.The real-time portrait of video contrasts server 3 by accessing streaming media server 2
The camera shooting and video for Identification of Images is obtained, video is compared in real time and carries out graphical analysis, by being carried including face extraction, feature
Take, feature similarity comparison carries out real-time portrait contrast.If the real-time portrait contrast server 3 of video is carried out after real-time portrait contrast
Similarity, which is met, to be required then by data base querying personal information, then by related personnel's information transfer to user platform.In addition,
Database update server 5 obtains log information or dependent event information also by streaming media server 2 is accessed, then to these
Information carries out collecting screening, obtains the video information for needing to store, and then obtains related figure information storage into database, makes
Database update is obtained, to carry out portrait dynamic contrast in the future.
The invention further relates to portrait dynamic contrast method, comprise the following steps, as shown in Figure 1a,
Step S101, the channel-style camera and full-view camera for being arranged on each site by the system gathers regarding for each net
Frequency information, for carrying out portrait dynamic contrast.
After step S102, the portrait video compress for each site being collected by streaming media server form streaming media video
File is stored, and files in stream media helps so that user can watch video in time.On the one hand streaming media video file is used to regard
The access of the frequency server of portrait comparison in real time, so as to carry out real-time portrait comparison.On the other hand it is used for database update server
Access, for updating the data the figure information in storehouse, consequently facilitating portrait dynamic comparison later.
Wherein, real-time portrait comparison is carried out to realize by step S103, it is specific as follows:The real-time portrait of video compares server
By accessing streaming media server, the camera video of Identification of Images is obtained, portrait in graphical analysis, with database is carried out and enters
Row is compared in real time.As shown in Figure 1 b, present embodiment step S103 is further comprising the steps of:
Step S1031, video acquisition, the real-time portrait of video compares server by accessing streaming media server, is used
In the camera video of Identification of Images;
Corresponding facial image is obtained in step S1032, face extraction, the camera video obtained from step S1031;
Step S1033, feature extraction obtains the face characteristic of corresponding site from the facial image of acquisition;
Facial image feature in above-mentioned face characteristic and database is carried out similarity pair by step S1034, similarity comparison
Than.
Enter step S104 after the completion of the comparison of portrait dynamic realtime, judge whether similarity meets, if similarity is full
It is sufficient then into step S105;Enter step S106 if similarity is unsatisfactory for;
Personal information is inquired about in step S105, database;
Step S106, terminates.
On the other hand, database figure information updates is realized by following steps,
The video data information of files in stream media is stored in video information storage single by step S201, video information storage
In member, and flag bit setting is carried out to video information memory cell, be then store in the interim storage of database update server
Area, described sign position setting includes unique sign index of the sender of portrait video, and all senders are with each
From unique sign index that the recipient as portrait video is general in the range of data delivery area, such sender just passes through
Unique sign index convection media server of the obtained sender with portrait video exports claimant as the instruction of video,
Described claimant includes the unique sign index for the sender for carrying portrait video as the instruction of video.
Step S202, information sifting weighting, the video information memory cell added with flag bit is weighted according to significance level,
Then weighting bit value is screened.
Above step S201 and step S202 as shown in Figure 3 a, by video information memory cell 7 be divided into data bit 701,
Summation of weighted bits 702 and flag bit 703, wherein video information is stored in data bit 701.Flag bit 703 is used for setting video to believe
The position of the scratchpad area (SPA) in database update server of memory cell 7 is ceased, flag bit 703 can be set in order here
Numerical value, so can in order to find correlation video information memory cell in video information.Summation of weighted bits 702 is for being easy to
Screen the video information of video information memory cell.Because each site passes through regarding that video signal acquisition device is collected daily
Frequency information is more, it is impossible to all retain storage;Cheese, event are few, so by adding to unit of video information
Power and position 702 is weighted mark, determines the significance level of associated video information, then deletes that weighted value is less regards by screening
Video information in frequency information memory cell, retains the information in the big video information memory cell of weighted value in addition.
Step S203, figure information is obtained, the video in the big video information memory cell of the weighted value left to screening
Data message carries out face extraction and feature extraction, obtains figure information.
The figure information of acquisition is put into figure information and deposited by step S204, figure information storage, database update server
In storage unit.
Step S205, database figure information is updated, and figure information memory cell in database update server is set
New flag bit, then will update into database added with the figure information memory cell of flag bit.
As shown in Figure 3 b, wherein figure information memory cell 401 includes data bit 402 by above step S204 and step S205
With flag bit 403, here data bit 402 be used for deposit correlation figure information;Flag bit 403 also for be easy to search portrait
Position of the information memory cell 401 in database 4, but flag bit 403 is not simple sequence here, is deposited according to portrait
Determined in storage unit 401 after the qualitative classification of figure information, figure information can be according to various features such as age, sexes here
Property point is come, and so can improve portrait dynamic in order to the figure information in frequency in real time portrait comparison server 3 searching data storehouse
The speed and analytical effect of contrast.In addition, needed after figure information memory cell 401 is stored in database add with wherein
The related information of portrait, such as identity, educational background, resume.
Although the foregoing describing the embodiment of the present invention, those skilled in the art should be appreciated that this
It is merely illustrative of, various changes or modifications can be made to present embodiment, without departing from the principle and essence of the present invention,
Protection scope of the present invention is only limited by the claims that follow.
Claims (3)
1. a kind of portrait dynamic contrast method, comprises the following steps,
Step S101:The portrait video of each site is gathered by video acquisition device;
Step S102:Form streaming media video file after the portrait video compress for each site being collected by streaming media server
Storage;
Step S103:The real-time portrait of video compares server by accessing streaming media server, obtains the shooting of Identification of Images
Machine video, carries out portrait in graphical analysis, with database and is compared in real time;
Step S104:Judge whether similarity meets, enter step S105 if similarity is met;If similarity is unsatisfactory for
Then enter step S106;
Step S105:Personal information is inquired about in database;
Step S106:Terminate;
Characterized in that, files in stream media is additionally operable to the renewal of database figure information in the step S102, comprise the following steps:
Step S201:Video information is stored, and the video data information of files in stream media is stored in video information memory cell,
And the setting of sign position is carried out to video information memory cell, it is then store in the scratchpad area (SPA) of database update server, institute
The sign position setting stated includes unique sign index of the sender of portrait video, and all sender with respectively as
The recipient of portrait video general unique sign index in the range of data delivery area, such sender is just by obtaining
Unique sign index convection media server of sender with portrait video exports claimant as the instruction of video, described
Claimant includes the unique sign index for the sender for carrying portrait video as the instruction of video;
Step S202:Information sifting is weighted, and the video information memory cell added with sign position is weighted according to significance level, then
Weighting bit value is screened;
Step S203:Figure information is obtained, and the video data information in video information memory cell left to screening enters pedestrian
Face is extracted and feature extraction, obtains figure information;
Step S204:Figure information is stored, and it is single that the figure information of acquisition is put into figure information storage by database update server
In member;
Step S205:Database figure information is updated, and figure information memory cell in database update server is set newly
Position is indicated, then updates the figure information memory cell added with sign position into database.
2. portrait dynamic contrast method according to claim 1, it is characterised in that the step S103 also includes following step
Suddenly:
Step S1031:Video acquisition, the real-time portrait of video compares server by accessing streaming media server, obtains being used for people
As the camera video of identification;
Step S1032:Corresponding facial image is obtained in face extraction, the camera video obtained from step S1031;
Step S1033:Feature extraction, obtains the face characteristic of corresponding site from the facial image of acquisition;
Step S1034:Similarity comparison, similarity comparison is carried out by facial image feature in above-mentioned face characteristic and database.
3. portrait dynamic contrast method according to claim 2, it is characterised in that the step S205 figure informations storage
The sign position of unit is set according to the qualitative classification of figure information.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410763088.2A CN104484651B (en) | 2014-12-12 | 2014-12-12 | Portrait dynamic contrast method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410763088.2A CN104484651B (en) | 2014-12-12 | 2014-12-12 | Portrait dynamic contrast method and system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104484651A CN104484651A (en) | 2015-04-01 |
CN104484651B true CN104484651B (en) | 2017-08-04 |
Family
ID=52759192
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410763088.2A Active CN104484651B (en) | 2014-12-12 | 2014-12-12 | Portrait dynamic contrast method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104484651B (en) |
Families Citing this family (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105868719A (en) * | 2016-03-31 | 2016-08-17 | 中山艾华企业管理咨询有限公司 | Face recognition and trace guiding system |
CN106407953A (en) * | 2016-10-14 | 2017-02-15 | 蔡璟 | Intelligent people searching system based on multi-image big data identification |
CN106776838A (en) * | 2016-11-24 | 2017-05-31 | 深圳明创自控技术有限公司 | A kind of massive video analysis and quick retrieval system based on cloud computing |
CN110825891B (en) * | 2019-10-31 | 2023-11-14 | 北京小米移动软件有限公司 | Method and device for identifying multimedia information and storage medium |
CN111594271A (en) * | 2020-06-02 | 2020-08-28 | 脑谷人工智能研究院(南京)有限公司 | Coal mine safety monitoring system based on portrait acquisition and processing |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101324919A (en) * | 2007-06-15 | 2008-12-17 | 上海银晨智能识别科技有限公司 | Photograph video contrast method |
CN102510478A (en) * | 2011-10-28 | 2012-06-20 | 唐玉勇 | Intelligent distribution control system and method used for 'Safe City' project |
CN102682283A (en) * | 2012-04-09 | 2012-09-19 | 重庆市行安电子科技有限公司 | Dynamic face recognition system |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20130216107A1 (en) * | 2012-02-20 | 2013-08-22 | Chih-Hsung Huang | Method of surveillance by face recognition |
-
2014
- 2014-12-12 CN CN201410763088.2A patent/CN104484651B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101324919A (en) * | 2007-06-15 | 2008-12-17 | 上海银晨智能识别科技有限公司 | Photograph video contrast method |
CN102510478A (en) * | 2011-10-28 | 2012-06-20 | 唐玉勇 | Intelligent distribution control system and method used for 'Safe City' project |
CN102682283A (en) * | 2012-04-09 | 2012-09-19 | 重庆市行安电子科技有限公司 | Dynamic face recognition system |
Also Published As
Publication number | Publication date |
---|---|
CN104484651A (en) | 2015-04-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106998444B (en) | Big data face monitoring system | |
CN104484651B (en) | Portrait dynamic contrast method and system | |
Chen et al. | A survey on an emerging area: Deep learning for smart city data | |
CN105022835B (en) | A kind of intelligent perception big data public safety recognition methods and system | |
CN101655860B (en) | Video aided analytical method for criminal investigation and case detection and system | |
Briggs et al. | Acoustic classification of multiple simultaneous bird species: A multi-instance multi-label approach | |
Katakis et al. | An ensemble of classifiers for coping with recurring contexts in data streams | |
CN103678417A (en) | Man-machine interaction data processing method and device | |
CN104091176A (en) | Technology for applying figure and head portrait comparison to videos | |
CN110263613A (en) | Monitor video processing method and processing device | |
CN115272037A (en) | Smart city region public security management early warning method and system based on Internet of things | |
CN106529373A (en) | Human face recognition monitoring alarm method and system | |
JP2022518459A (en) | Information processing methods and devices, storage media | |
CN108062366B (en) | Public culture information recommendation system | |
CN104142995A (en) | Social event recognition method based on visual attributes | |
CN107230267A (en) | Intelligence In Baogang Kindergarten based on face recognition algorithms is registered method | |
CN108898051A (en) | A kind of face identification method and system based on video flowing | |
CN102314916A (en) | Video processing method and system | |
CN112836657B (en) | Pedestrian detection method and system based on lightweight YOLOv3 | |
CN106537387B (en) | Retrieval/storage image associated with event | |
CN109492604A (en) | Faceform's characteristic statistics analysis system | |
CN107358190A (en) | A kind of image key area management method and device | |
CN109830277A (en) | A kind of rope skipping monitoring method, electronic device and storage medium | |
CN202120402U (en) | Mass video analysis based on cloud computing and rapid retrieval system | |
CN106713859A (en) | Image visual monitoring search system and search method thereof |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CP03 | Change of name, title or address |
Address after: Room 701, 6th Floor, Hengrun Building, No.201 Zhuqiao Road, Suzhou City, Jiangsu Province, 215000 Patentee after: Suzhou Jinnaodai Intelligent System Engineering Co.,Ltd. Address before: Room 510, Ruiji Building, No. 879 Ganjiang Road, Suzhou City, Jiangsu Province, 215000 Patentee before: SUZHOU GOLDEN BRAIN INTELLIGENT SYSTEM ENGINEERING CO.,LTD. |
|
CP03 | Change of name, title or address |