CN104581081B - Passenger flow analysing method based on video information - Google Patents
Passenger flow analysing method based on video information Download PDFInfo
- Publication number
- CN104581081B CN104581081B CN201510039687.4A CN201510039687A CN104581081B CN 104581081 B CN104581081 B CN 104581081B CN 201510039687 A CN201510039687 A CN 201510039687A CN 104581081 B CN104581081 B CN 104581081B
- Authority
- CN
- China
- Prior art keywords
- video
- passenger flow
- video information
- camera
- information
- 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
Abstract
Lack the weak point lacked to density of stream of people monitoring in the prior art to make up, and improve the degree of accuracy of the passenger flow statisticses based on video and provide necessary support for the security of public arena, the invention provides a kind of passenger flow analysing method based on video information, including：Obtain the first video information of the first video camera；First video information is transmitted to the first analytic unit；First analytic unit carries out the first passenger flow analysing to the first video information；According to the result of the first passenger flow analysing, start the second video camera and the 3rd video camera, and obtain the second video information from the second video camera and the 3rd video information from the 3rd video camera；Second video information and the 3rd video information are transmitted to steps such as the second analytic units.The present invention realizes the passenger flow accurate statistics based on video, and can provide and be automatically reminded to when density of stream of people is too high.In addition, by the prediction of movement locus etc., also technical support is provided for security work.
Description
Technical field
The present invention relates to safety and traffic monitoring technical field, more particularly, to use in public places based on regarding
The passenger flow analysing method of frequency information.
Background technology
At present, statistical management, the emergency evacuation of the public place due to itself such as domestic various supermarket, large-scale sales field and shops
Or the requirement of regulations for fire fighting etc., increasingly pay attention to the detection and analysis of passenger flow quantity.
On the one hand, existing detection of passenger flow system is typically to install video camera in specific position, is being controlled by special messenger
The image information of center monitors shooting, it is mainly used in safety monitoring, and the potential accident of trampling come for passenger flow dense band is endangered
Danger can not analyze precognition.
Traditional passenger flow statisticses mode mainly has two kinds, first, manually being counted by monitoring work personnel to realize；Second, make
The passenger flow number by a certain fixed position is counted with infrared photoelectric sensor.Observed using artificial or video monitoring indirect labor
To carry out the mode of passenger flow statisticses not only labor intensive cost, and the human eye that works long hours is easily tired, when the volume of the flow of passengers is intensive
When, observer's reaction does not come, and the accuracy rate of statistics is relatively low；With infrared emission block triggering come the method for statistical number of person often by
Be formed on site environment, equipment is not easy to install, it is impossible to distinguish people and object, when occur more people in parallel through or intersect pass in and out when,
Statistical error will be produced.Therefore, no matter counting accuracy rate or installing and using on convenience, the passenger flow based on video
Statistics has very big advantage than traditional passenger flow statisticses, and the passenger flow demographics based on video can be according to application scenario
Difference adjusts detection range, and hardware device is easy for installation, and has storage playback function, can be preferably applied for various multiple
Miscellaneous occasion.
Guest flow statistics based on video is mainly to analyze the pedestrian movement track in video, accumulative to enter and leave two
Passenger flow number on direction, be currently based on video analysis passenger flow statisticses technically there is also some problems and deficiency, mainly
Robustness is not good enough, is influenceed by environment, light condition, may there is preferable effect in some occasions, but in some occasions
Under statistical accuracy do not reach requirement.Do not have for pedestrian with cap, white hair, knapsack and the special circumstances such as floor background is more black
There is good separating capacity, produce statistical error.In addition, the passenger flow statisticses based on video are mainly the video by rear end
Analysis server to carry out Treatment Analysis to video image, and video surveillance point is more, background server while analyzing multiple video
When computing pressure it is bigger, or even need increase server apparatus, so as to increase cost input, and be not easy to extend.
On the other hand, in festivals or holidays, famous sites and tourist district or the public area stream of people of great celebration etc. is held
Gather density is high, it is easy to tread event occurs.And lack the effective ways to density of stream of people monitoring in the prior art.
The content of the invention
Lack the weak point lacked to density of stream of people monitoring in the prior art to make up, and improve based on video
The degree of accuracy of passenger flow statisticses and security for public arena provides necessary support, the invention provides one kind to be based on video information
Passenger flow analysing method, including：
(1) the first video information of the first video camera is obtained；
(2) first video information to the first analytic unit is transmitted；
(3) first analytic units carry out the first passenger flow analysing to the first video information；
(4) according to the result of the first passenger flow analysing, the second video camera and the 3rd video camera are started, and obtain and taken the photograph from second
Second video information of camera and the 3rd video information from the 3rd video camera；
(5) second video information and the 3rd video information to the second analytic unit are transmitted；
(6) second analytic units carry out the second passenger flow analysing to the second video information and the 3rd video information；
(7) according to the second passenger flow analysing result, sound detection unit and the 4th video camera are started；
(8) acoustic information of the sound detection unit detection and the 4th video letter of the 4th shooting machine testing are transmitted
Cease to the 3rd analytic unit；
(9) the 3rd passenger flow analysing is carried out to the acoustic information and the 4th video information；
(10) according to the result of the 3rd passenger flow analysing, the result of the 3rd passenger flow analysing is sent and to visitor to security department
Stream provides prompt message.
Further, the first video information is by cable-network transmission to the first analytic unit.
Further, the second video information and the 3rd video information are by wireless network transmissions to the second analytic unit
's.
Further, the prompt message includes warning message.
Further, first analysis includes：Intensity of passenger flow is analyzed.
Further, second analysis includes：Movement velocity is analyzed and volume of the flow of passengers analysis.
Further, the 3rd analysis includes recognition of face analysis.
Further, second video camera and the 3rd video camera use thermal camera.
Beneficial effects of the present invention are as follows：The present invention realizes the passenger flow accurate statistics based on video, and can be in people
Provide and be automatically reminded to when current density is too high.In addition, by the prediction of movement locus etc., also technology branch is provided for security work
Hold.
Brief description of the drawings
Fig. 1 shows the flow chart of the passenger flow analysing method based on video information of the present invention.
Embodiment
The passenger flow analysing method based on video information of the present invention is as shown in Figure 1.This method includes：Utilize 1080p high definitions
Clear degree video camera obtains the video information of passenger flow position from the surface at passenger flow place, then without compressing, and it is straight
Connect and the information is transferred to neighbouring Control Room by special video delivery network line, by the first server in Control Room
The frame for carrying out video image is selected, and selects the frame of definition highest and fog-level minimum, and passes therethrough image procossing calculating
Number of people number in the frame, so that it is determined that the passenger flow number in the frame.Then, by divided by the video that pre-sets cover
The area in region, calculate intensity of passenger flow.When intensity of passenger flow is less than default intensity of passenger flow threshold value, such as the area shared by everyone is low
When 0.75 square metre, start the first thermal camera and carry out human body infrared thermal imaging.First thermal camera, which is arranged on, to be treated
Oblique 45 degree of tops of the passenger flow of analysis, and the area coincidence that the region being directed to is directed to described 1080p high-definition cameras.
By the infrared information that the thermal camera obtains by the second server in 4G network transmissions to Control Room, then image is entered
The processing of row picture element matrixization, that is, the rgb value of all pixels in each frame of video is calculated, and to wherein representing red area
RGB areas are added up one by one, if red area is beyond predetermined threshold value, (through experiment, human body is in infrared imaging image kind
Red area is substantially at 0.2 square metre), then it is assumed that the red area represents a people.Further analyze in this way
Passenger flow number in the video area.Then, intensity of passenger flow is calculated by method same as above.If pass through first service
The intensity of passenger flow of device and second server is below default intensity of passenger flow threshold value, then it is assumed that intensity of passenger flow now has exceeded police
Limit is guarded against, to security department's alert.
While above-mentioned first thermal camera is opened, the second thermal camera is also opened.Second thermal camera
Oblique 45 degree of tops in above-mentioned region to be analyzed are arranged on, and the region absorbed is arranged to absorb region in the first thermal camera
(i.e. above-mentioned region to be analyzed) is separated by 100 meters of region.Preferably, the adjacent area is in above-mentioned region to be analyzed
In the direction of advance of the movement locus of passenger flow.The video information of second infrared camera scan is passed through 4G network transmissions to institute
State second server.Second server according to the infrared information from the first thermal camera and the second thermal camera, according to
Object of reference (for example, the special person of the bodily form chosen in passenger flow is object of reference) in the image of the first thermal camera intake occurs
The spent time in the image of the second thermal camera intake, the motion of passenger flow is obtained using 100 meters divided by the time
Velocity information.When the movement velocity of passenger flow exceedes predetermined threshold value, identification now may have occurred accident.Now, it is sharp
The image shot with above-mentioned 1080p video cameras carries out recognition of face to owner, and the image result that recognition of face is obtained is protected
Memory cell is stored to, and is recorded in the sound transducer for opened while recognition of face areas adjacent to be analyzed, and
Recorded message is sent to the 3rd server of Control Room by 4G networks.3rd server to the sound carry out low frequency filtering and
Noise reduction process, if the mean frequency value of obtained audio-frequency information is beyond default threshold value (being greater than 1500Hz), then it is assumed that
Now there occurs accident (for example, fearing time etc. cruelly).
Now, the 4th video camera is then turned on, the 4th video camera is using three-dimensional camera and preferably with adjustable shooting
Angle three-dimensional camera.The video camera instructs the video of shooting specific location according to caused by the 3rd server.3rd clothes
The face characteristic of the business device a bad actor saved to database in the image and the 3rd server of the recognition of face is carried out
Contrast and provide similarity and regarded more than the face (determining such a artificial doubtful personnel) of predetermined threshold (such as 70%) in newest
The location of in frequency frame.If the relevant position for the image that the 4th video camera photographs includes above-mentioned doubtful personnel, to
Security department sends the positional information and provides prompt message to passenger flow, and preferably the information is warning message.
Although embodiments of the invention have been shown and described above, it is to be understood that above-described embodiment is example
Property, it is impossible to limitation of the present invention is interpreted as, one of ordinary skill in the art is not departing from the principle and objective of the present invention
In the case of above-described embodiment can be changed within the scope of the invention, change, replace and modification.
Claims (5)
- A kind of 1. passenger flow analysing method based on video information, it is characterised in that including：(1) video information of passenger flow position is obtained from the surface at passenger flow place using 1080p high-definition cameras, As the first video information；(2) to the first video information without compression, and directly the information is transmitted by special video delivery network line To neighbouring Control Room, the Control Room is as the first analytic unit；(3) first analytic units carry out the first passenger flow analysing to the first video information, and first passenger flow analysing includes：By monitoring The frame that first server in room carries out video image is selected, and selects the frame of definition highest and fog-level minimum, and therefrom The number of people number in the frame is calculated by image procossing, so that it is determined that the passenger flow number in the frame；Then, by divided by set in advance The area in the region for the video covering put, calculates intensity of passenger flow；(4) according to the result of the first passenger flow analysing, the second video camera and the 3rd video camera are started, and obtain and come from the second video camera The second video information and the 3rd video information from the 3rd video camera, the step include：When the area shared by everyone is less than 0.75 square metre constantly, starts the first thermal camera and carries out human body infrared thermal imaging；First thermal camera, which is arranged on, to be treated Oblique 45 degree of tops of the passenger flow of analysis, and the area coincidence that the region being directed to is directed to described 1080p high-definition cameras, Wherein the first thermal camera starts the second thermal camera, the second thermal camera conduct as the second video camera 3rd video camera, second thermal camera is arranged on oblique 45 degree of tops in above-mentioned region to be analyzed, and the region absorbed is set It is set to being separated by 100 meters of region in the first thermal camera intake region；(5) second video information and the 3rd video information are transmitted to the second analytic unit, including：First thermal camera The video information of second infrared camera scan is by by 4G network transmissions to second server, and the second server is as Two analytic units；(6) second analytic units carry out the second passenger flow analysing, second passenger flow to the second video information and the 3rd video information Analysis includes：Picture element matrix processing is carried out to image, that is, calculates the rgb value of all pixels in each frame of video, and to it The middle RGB areas for representing red area are added up one by one, if red area exceeds predetermined threshold value, then it is assumed that the red color area Domain represents a people；The passenger flow number in the region of video covering is further analyzed in this way；Then, by with Method as above calculates intensity of passenger flow；Second server is according to from the first thermal camera and the second thermal camera Infrared information, the object of reference in the image absorbed in the first thermal camera appear in the figure of the second thermal camera intake The spent time as in, the motion velocity information of passenger flow is obtained using 100 meters divided by the time；When the movement velocity of passenger flow During more than predetermined threshold value, identification now may have occurred accident；(7) when the movement velocity of passenger flow exceedes predetermined threshold value, using the image of above-mentioned 1080p video cameras shooting to owner Recognition of face is carried out, and the image result that recognition of face is obtained is saved in memory cell, and while recognition of face is carried out The sound transducer for opening areas adjacent to be analyzed is recorded, and recorded message is sent to the of Control Room by 4G networks Three servers；3rd server carries out low frequency filtering and noise reduction process to the sound, and the mean frequency value of obtained audio-frequency information is such as Fruit is beyond default threshold value, then it is assumed that now there occurs accident；(8) if now there occurs accident, the 4th video camera is then turned on, the 4th video camera uses three-dimensional camera and tool There is adjustable shooting angle three-dimensional camera；The video camera instructs regarding for shooting specific location according to caused by the 3rd server Frequently；The people of the 3rd server a bad actor saved to database in the image and the 3rd server of the recognition of face Face feature is contrasted and provides face of the similarity more than predetermined threshold location in newest frame of video；If the The relevant position for the image that four video cameras photograph includes doubtful personnel, then the knot of the 3rd passenger flow analysing is sent to security department Fruit simultaneously provides prompt message to passenger flow, and the doubtful personnel are the personnel with the face for meeting following condition：3rd server The face characteristic of image to the recognition of face a bad actor saved with database in the 3rd server contrasts And provide the face that similarity exceedes predetermined threshold.
- 2. the passenger flow analysing method according to claim 1 based on video information, it is characterised in that the first video information is to pass through Cable-network transmission is to the first analytic unit.
- 3. the passenger flow analysing method according to claim 1 based on video information, it is characterised in that the second video information and Three video informations are by wireless network transmissions to the second analytic unit.
- 4. the passenger flow analysing method according to claim 1 based on video information, it is characterised in that the prompt message includes report Alert information.
- 5. the passenger flow analysing method according to claim 1 based on video information, it is characterised in that second video camera and Three video cameras use thermal camera.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510039687.4A CN104581081B (en) | 2015-01-26 | 2015-01-26 | Passenger flow analysing method based on video information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510039687.4A CN104581081B (en) | 2015-01-26 | 2015-01-26 | Passenger flow analysing method based on video information |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104581081A CN104581081A (en) | 2015-04-29 |
| CN104581081B true CN104581081B (en) | 2018-03-30 |
Family
ID=53096158
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510039687.4A Active CN104581081B (en) | 2015-01-26 | 2015-01-26 | Passenger flow analysing method based on video information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104581081B (en) |
Families Citing this family (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106934356B (en) * | 2017-02-28 | 2023-12-01 | 苏州清研微视电子科技有限公司 | Bus stop night passenger flow statistical method and system based on thermal imaging |
| CN107483894B (en) * | 2017-09-13 | 2019-06-04 | 浙大网新系统工程有限公司 | The high-speed rail station video monitoring system of realization passenger transportation management is judged based on scene |
| CN108154108A (en) * | 2017-12-22 | 2018-06-12 | 触景无限科技（北京）有限公司 | The monitoring method and device of a kind of emergency |
| CN108334829B (en) * | 2018-01-23 | 2021-08-06 | 浙江大仓信息科技股份有限公司 | Dynamic passenger car number counting system |
| CN108419014B (en) * | 2018-03-20 | 2020-02-21 | 北京天睿空间科技股份有限公司 | Method for capturing human face by linkage of panoramic camera and multiple capturing cameras |
| CN108932485A (en) * | 2018-06-21 | 2018-12-04 | 芜湖宝文信息科技有限公司 | The statistics equipment for the people information statistics being easily installed |
| CN108921063A (en) * | 2018-06-21 | 2018-11-30 | 芜湖宝文信息科技有限公司 | The statistics equipment of simple installation people information statistics |
| CN108921273A (en) * | 2018-06-21 | 2018-11-30 | 芜湖宝文信息科技有限公司 | Statistics equipment for people information statistics |
| CN110706459A (en) * | 2019-10-22 | 2020-01-17 | 山东国惠安创智能物联发展有限公司 | Situation perception cabinet air conditioner system |
| CN110909607B (en) * | 2019-10-24 | 2023-06-09 | 北京中润惠通科技发展有限公司 | Passenger flow sensing device system in intelligent subway operation |
| CN111785030A (en) * | 2020-08-06 | 2020-10-16 | 展讯通信（深圳）有限公司 | Traffic monitoring terminal and traffic monitoring system |
| CN112839198A (en) * | 2020-12-22 | 2021-05-25 | 安徽东方风景建设有限公司 | Scenic spot security monitoring method and monitoring system based on big data processing |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2005039181A1 (en) * | 2003-10-21 | 2005-04-28 | Matsushita Electric Industrial Co., Ltd. | Monitoring device |
| CN101795395A (en) * | 2009-02-04 | 2010-08-04 | 深圳市先进智能技术研究所 | System and method for monitoring crowd situation |
| CN101959060A (en) * | 2010-10-26 | 2011-01-26 | 镇江科大船苑计算机网络工程有限公司 | Video passenger flow monitoring system and method |
| CN102724478A (en) * | 2012-05-25 | 2012-10-10 | 张亮 | Embedded mobile intelligent monitoring method and system facing public safety service |
| CN203872286U (en) * | 2014-05-26 | 2014-10-08 | 北京中电兴发科技有限公司 | Intelligent monitoring device |
-
2015
- 2015-01-26 CN CN201510039687.4A patent/CN104581081B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2005039181A1 (en) * | 2003-10-21 | 2005-04-28 | Matsushita Electric Industrial Co., Ltd. | Monitoring device |
| CN101795395A (en) * | 2009-02-04 | 2010-08-04 | 深圳市先进智能技术研究所 | System and method for monitoring crowd situation |
| CN101959060A (en) * | 2010-10-26 | 2011-01-26 | 镇江科大船苑计算机网络工程有限公司 | Video passenger flow monitoring system and method |
| CN102724478A (en) * | 2012-05-25 | 2012-10-10 | 张亮 | Embedded mobile intelligent monitoring method and system facing public safety service |
| CN203872286U (en) * | 2014-05-26 | 2014-10-08 | 北京中电兴发科技有限公司 | Intelligent monitoring device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104581081A (en) | 2015-04-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104581081B (en) | Passenger flow analysing method based on video information | |
| US11823492B2 (en) | Technique for providing security | |
| KR101215948B1 (en) | Image information masking method of monitoring system based on face recognition and body information | |
| CN102521578B (en) | Method for detecting and identifying intrusion | |
| CN105354563B (en) | Face datection prior-warning device and implementation method are blocked in conjunction with depth and color image | |
| KR102195706B1 (en) | Method and Apparatus for Detecting Intruder | |
| CN106815958A (en) | Warning system, alarm analysis/display device, alarm analysis/display methods | |
| CN108376246A (en) | A kind of identification of plurality of human faces and tracking system and method | |
| KR101036947B1 (en) | The automatic guard system to prevent the crime and accident using computer video image analysis technology | |
| CN109389794A (en) | A kind of Intellectualized Video Monitoring method and system | |
| CN108230669B (en) | Road vehicle violation detection method and system based on big data and cloud analysis | |
| CN108460319B (en) | Abnormal face detection method and device | |
| CN110602453B (en) | Internet of things big data intelligent video security monitoring system | |
| KR20100119920A (en) | System for pursuit moving object using security crime prevention network and method thereof | |
| KR102142315B1 (en) | ATM security system based on image analyses and the method thereof | |
| KR102046591B1 (en) | Image Monitoring System and Method for Monitoring Image | |
| KR101113515B1 (en) | Video index system using surveillance camera and the method thereof | |
| KR20200086015A (en) | Situation linkage type image analysis device | |
| CN111241918B (en) | Vehicle tracking prevention method and system based on face recognition | |
| KR102081577B1 (en) | Intelligence Fire Detecting System Using CCTV | |
| KR100920937B1 (en) | Apparatus and method for detecting motion, and storing video within security system | |
| CN107784829A (en) | A kind of passenger flow sensory perceptual system based on security | |
| US20230088660A1 (en) | Identity-concealing motion detection and portraying device | |
| CN115100572A (en) | System and method for analyzing abnormal behaviors of campus | |
| CN101540891A (en) | Luggage delivery warehouse human body detecting system based on monitoring video |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB03 | Change of inventor or designer information | ||
| CB03 | Change of inventor or designer information |
Inventor after: Lv Yangbo Inventor after: Hu Rui Inventor after: Lang Hongfeng Inventor before: Liu Dong |
|
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20180213 Address after: Hangzhou City, Zhejiang province 310030 Wensan Road No. 630, room 218 Applicant after: Zhejiang 1one. Cn Technology Corp. Applicant after: Lv Yangbo Address before: 610041, No. 30, No. 1, building 1, 2, No. 177, Tianfu Road, Chengdu hi tech Zone, Sichuan Applicant before: CHENGFU MEILIAN WEIZHI TECHNOLOGY CO., LTD. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |