CN201255897Y - Human flow monitoring device for bus - Google Patents
Human flow monitoring device for bus Download PDFInfo
- Publication number
- CN201255897Y CN201255897Y CNU2008200303752U CN200820030375U CN201255897Y CN 201255897 Y CN201255897 Y CN 201255897Y CN U2008200303752 U CNU2008200303752 U CN U2008200303752U CN 200820030375 U CN200820030375 U CN 200820030375U CN 201255897 Y CN201255897 Y CN 201255897Y
- Authority
- CN
- China
- Prior art keywords
- bus
- camera
- computing machine
- passenger
- monitoring device
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Images
Landscapes
- Closed-Circuit Television Systems (AREA)
Abstract
The utility model discloses a bus passenger flow detector, belonging to the overall distribution technical field of bus passenger flow, which comprises a computer connected with a front gate camera and a back gate camera, wherein the front gate camera is provided at the top of a bus front gate for picking up the video images of the passengers getting on, and the back gate camera is provided at the top of a bus back gate for picking up the video images of the passengers getting off, the computer obtains the video images of the front and the back gate cameras to process target identification and statistics.
Description
Technical field
The utility model relates to the pool scheduling field of the flow of the people of bus, relates in particular to a kind of passenger throughput of bus monitoring device.
Background technology
The statistics of existing passenger throughput of bus mainly contains three kinds of methods, the number that first driver or ticket seller get on or off the bus with the knob counters count; It two is to receive cash by checking, and calculates number of passengers by cash quantity; It three is to utilize the quantity that dissimilar holders get on the bus on the different websites of bus IC card system statistics, uses the cash quantity of being received to calculate the number of passengers of inserting coins again.In addition, adopt ultrasound wave in addition, infrared, and the flow of the people counter of photoelectricity etc.
Method one is statistical number of person more accurately, yet this manual method workload is big, and the cost height is difficult to lasting use; Method two just can be calculated one day non-total patronage of holding the public transport monthly ticket of a regular bus, can not get each platform number of getting on or off the bus and the number of holding the public transport monthly ticket; Method three can calculate that one day regular bus holds the quantity of IC-card holder on total patronage of public transport 1C card and each website, can not get the passengers quantity of inserting coins and getting on the bus on each website, also can not get the number that each website is got on or off the bus.Adopt infrared or radio frequency induction technology, can't distinguish the quantity of passenger when entering simultaneously side by side.Therefore error ratio is bigger, can't satisfy the demand of analyst's flow data on the bus; Adopt those traditional methods to calculate the number that has now on the car simultaneously, can not when number overloads, report to the police, simultaneously can not be to the inlet of bus, outlet is monitored.
Summary of the invention
The purpose of this utility model is at the prior art problem, and a kind of passenger throughput of bus monitoring device is provided, and this device can provide the statistical data analysis of day part flow of the people, for bus scheduling provides reference information.
In order to address the above problem, the utility model is achieved by the following technical solutions: a kind of passenger throughput of bus monitoring device, it is characterized in that, and comprise computing machine, and be connected Qianmen camera and back door camera with computing machine; Described Qianmen camera is arranged on the top at bus Qianmen, absorbs the passenger's that gets on the bus video image; Described back door camera is arranged on the top at bus back door, absorbs the passenger's that gets off video image; Described computing machine obtains the video image of forward and backward door camera, carries out target identification and statistics.
Further characteristics of the present utility model are:
(1) also comprise be connected with computing machine display, described display subregion shows the video image of forward and backward door camera.
(2) also comprise the alarm that is connected with computing machine, when bus passenger number transfinites, give the alarm.
The utility model adopts camera collection passenger getting on/off image information, after by computing machine image information being discerned, counts, and the people flow rate statistical of bus is accurate, and does not need to drop into manpower; Be equipped with display simultaneously, make things convenient for the monitoring of driver and conductor the gateway; Alarm is set, when number surpasses limit value, reports to the police, avoid occurring the potential safety hazard of overcrowding overload.
Description of drawings
Below in conjunction with description of drawings and embodiment the utility model is described in further details.
Fig. 1 hardware of the present utility model connects block diagram;
The synoptic diagram of Fig. 2 the utility model camera tracing area;
The image recognition process flow diagram of Fig. 3 computing machine.
Embodiment
With reference to Fig. 1, the utility model comprises computing machine, and the display that is connected with computing machine, alarm, Qianmen camera and back door camera; Described Qianmen camera and back door camera are separately positioned on the top of the front and back door of bus.The Qianmen camera is used for gathering passenger's the video image of getting on the bus, and the back door camera is used for gathering passenger's the image of getting off, and computing machine carries out identification with the video image of camera collection, adds up then, stores.The a certain moment, when the passenger's that gets on the bus quantity and the passenger's that gets off quantity difference greater than the carrying of bus in limited time, the computer control alarm gives the alarm, and carries out safety instruction to the driver and conductor.Simultaneously, with the output of display, during bus running, can provide the image information of front and back door to the driver and conductor as computing machine.
With reference to Fig. 2, camera 1 is installed in position directly over bus door 2 tops, and the shooting hole slopes inwardly. Space line segment 11,12,13,14 among Fig. 2 is formed tracing area, and this tracing area is the actual monitoring zone of camera, and computing machine is to discerning processing through this regional moving image, the number of statistics process.Bus goes out (going into) mouthful floor 3 corresponding tracing areas and is provided with counting line 4, and counting line 4 is that computing machine is used for judging whether moving object crosses tracing area fully.When having only passenger image to cross counting line 4, just real gets on the bus, just it is counted, when the benefit of utilizing the counting line technology mainly is the number of statistics moving object, computing machine only need be followed the tracks of processing roughly near the object the counting line, and not needing complicated trace routine to come all-the-way tracking in the tracing area is carried out in moving object, this has greatly reduced the complexity of algorithm.
With reference to Fig. 3, the image processing procedure of this practical sexual type is elaborated.At first camera is caught the car door passenger's of place that comes in and goes out image, is transferred to computing machine and stores; Secondly computing machine carries out motion segmentation to the image that obtains, and is partitioned into the moving target piece; Computing machine carries out Target Recognition to the moving target piece then; The moving mass figure is carried out cluster analysis, analyze the center of mass point of wherein each human body; When crossing counting line, counts moving target piece barycenter processing.
(1) Video Capture
Camera is installed in position directly over the bus door top, and the shooting hole slopes inwardly, and catches the video image at the car door place of coming in and going out.
(2) motion segmentation
The main effect of motion segmentation part is the moving mass that is partitioned in the video image, makes things convenient for the needs of latter feature extraction, pattern-recognition.In moving Object Segmentation, based on moving Object Segmentation method based on the frame difference.This paper mainly adopts based on the motion segmentation algorithm that shifts frame difference and morphological operator.
This algorithm is a kind of higher order statistical characteristic of transfer frame difference DFD and object video automatic division method of mathematical morphology operators of utilizing, and basic idea is to isolate the motion object from static image background.
Whole algorithm mainly is divided into three steps, specifically describes as follows:
(a) detection of interframe variation
The detection that interframe changes is adopted a detection to the certificate variation to be modeled as and detect non-Gaussian signal at random in Gaussian noise, wherein adopts the Fourth-order moment detecting device.
(b) extraction of pre-segmentation template
Suppose that one section video sequence contains N two field picture (f
0～f
N-1), calculate the difference d=f of i frame and i+m two field picture
i-f
I+m(0≤i, i+m≤N-1), then d is carried out Fourth-order moment and detect can obtain a binary map M
i(template).If M
i(x, y)=0, then pixel belongs to static background; If M
i(x, y)=1, then pixel belongs to the motion object.Wherein, m is a constant of determining according to the object motion speed.Calculate i=0, m+1,2m+1 ..., a series of templates during N-m-1 are carried out the template that logical OR is operated to the end to these templates.
(c) to the corrosion of template
In order to obtain the accurate expression of object shapes, the template that we obtain previous step is carried out the morphological erosion operation.Corrosion is inwardly carried out along beginning from the ragged edge of the template that previous step is obtained, up to the edge of object.
(3) Target Recognition
Target Recognition mainly is divided into two parts, the feature selecting of moving mass and to the mode identification procedure of proper vector.
(a) feature selecting
The purpose of feature extraction part mainly is to determine that can fully be described the feature of the figure of required identification, so that utilize these features that figure is discerned.
Choosing of proper vector, whether consider the requirement of system: both having required to identify is human body, require to identify the number in the human body again, with reference to the scheme of being taked in present existing human body identification and the number system, present embodiment makes up the constitutive characteristic vector with shape and area parameters.Utilize shape to discern human region, utilize area to come the judgement number.In concrete operation, area parameters can obtain by the area that calculates the moving region that is obtained.And,,, adopt Fourier descriptors with reference to present existing algorithm in order to adapt to the influence of translation, rotation for form parameter.
The computation process of Fourier descriptors:
At first, the motion module image that is obtained is carried out morphologic opening operation, with the trick of eliminating some shades in the extraction process and human body influence to graphics shape;
Secondly, extract the edge chain code of motion module, what present embodiment adopted is that 8 neighborhood methods are found the solution, and chain code is carried out normalized, to make things convenient for subsequent treatment.In actual treatment, be N=128 with all chain code length normalization method, employing be equidistant extraction, consider the pattern of extraction point slope simultaneously; Central point with chain code is normalized to true origin simultaneously, eliminates the Influence of Displacement in the follow-up Fourier descriptors, only keeps pure shape information.
Then, calculate the Fourier descriptors of the chain code that obtains, the method for utilizing G.H.Granlund to propose is simultaneously carried out normalization to Fourier descriptors; After Fourier descriptors is carried out normalization operation, the Fourier descriptors that is obtained will be irrelevant with the selection of translation, size, rotation and the profile starting point of image, only keep the shape information of image.
At last, the low frequency part of the Fourier descriptors that intercepting is obtained, 32 parameters of low coefficient.
(b) extraction of proper vector
Behind the constitutive characteristic of having selected proper vector, concrete proper vector leaching process probably is divided into following several steps:
At first, the binary map that comprises moving target of input is carried out the morphology pre-service, eliminate the fritter target in the figure and adjust the shape of target, make that it is more level and smooth, continuously; Eliminate the influence of hand, foot section simultaneously, make the image of whole movement human more similar;
Then, to each continuum in the figure after handling, the boundary coordinate sequence of adding up them respectively, the normalization Fourier descriptors of computation bound, the area of zoning makes up the proper vector that obtains final each moving mass of description simultaneously.
(4) the moving mass barycenter obtains
At the attribute of having determined moving mass (being the people, the several people) afterwards, need to follow the tracks of this moving mass, to provide actual volume of the flow of passengers information.For the ease of following the tracks of, after determining attribute, utilized the method for a cluster analysis, the moving mass figure is carried out cluster analysis, provide the wherein central point of each human body.In the process of the tracking of back and demographics, it is just passable only to need to follow the tracks of each central point, and does not need each moving mass is carried out complicated tracking Control.
(5) counting
At the bus Qianmen,, corresponding counting variable is added 1 operation when having captured barycenter when crossing counting line; At the bus back door,, corresponding counting variable is subtracted 1 operation when having captured barycenter when crossing counting line.
The course of work of passenger throughput of bus supervising device; The count initialized variable, initialization number output variable, initialization time counter; At the Qianmen, when the people will get on the bus, camera obtained the video sequence that needs the monitored area in real time, and computing machine utilizes image process method to handle to video sequence, is partitioned into moving mass wherein; Moving mass is described, obtains the proper vector of movement human, utilize the method for pattern-recognition to discern then, provide the character (information such as several people are arranged) of moving mass; The moving mass of judging the people is carried out the clustering block operation, analyze the center of mass point of wherein each human body; Near counting line, center of mass point is followed the tracks of operation, when having captured barycenter when crossing counting line, corresponding counting variable is added 1 operation; At the back door, the video sequence of camera collection monitored area carries out Flame Image Process to it equally, when having captured barycenter when crossing counting line, corresponding counting variable is subtracted 1 operation.So just can write down the situation of each number of getting on or off the bus constantly, can deduct number on the real-time record car of the number of getting off according to the number of getting on the bus simultaneously, when suppose that number on the car is too much, above limit value, can alarm.
In addition, the video information of camera is stored in the computing machine, computing machine connects display, can carry out monitoring to the place, gateway of bus.
Claims (3)
1, a kind of passenger throughput of bus monitoring device is characterized in that, comprises computing machine, and is connected Qianmen camera and back door camera with computing machine; Described Qianmen camera is arranged on the top at bus Qianmen, absorbs the passenger's that gets on the bus video image; Described back door camera is arranged on the top at bus back door, absorbs the passenger's that gets off video image; Described computing machine obtains the video image of forward and backward door camera, carries out target identification and statistics.
2, a kind of passenger throughput of bus monitoring device according to claim 1 is characterized in that, also comprise be connected with computing machine display, described display subregion shows the video image of forward and backward door camera.
3, a kind of passenger throughput of bus monitoring device according to claim 1 is characterized in that, also comprises the alarm that is connected with computing machine, when bus passenger number transfinites, gives the alarm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNU2008200303752U CN201255897Y (en) | 2008-09-23 | 2008-09-23 | Human flow monitoring device for bus |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNU2008200303752U CN201255897Y (en) | 2008-09-23 | 2008-09-23 | Human flow monitoring device for bus |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN201255897Y true CN201255897Y (en) | 2009-06-10 |
Family
ID=40739223
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNU2008200303752U Expired - Fee Related CN201255897Y (en) | 2008-09-23 | 2008-09-23 | Human flow monitoring device for bus |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN201255897Y (en) |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101877058A (en) * | 2010-02-10 | 2010-11-03 | 杭州海康威视软件有限公司 | People flow rate statistical method and system |
| CN102514489A (en) * | 2011-12-16 | 2012-06-27 | 张翔宇 | Automotive display controller for avoiding traffic accidents due to overload of passengers |
| CN103021059A (en) * | 2012-12-12 | 2013-04-03 | 天津大学 | Video-monitoring-based public transport passenger flow counting method |
| CN103440763A (en) * | 2013-08-09 | 2013-12-11 | 百灵时代传媒集团有限公司 | Auxiliary method and device for taking buses |
| CN104021605A (en) * | 2014-04-16 | 2014-09-03 | 湖州朗讯信息科技有限公司 | Real-time statistics system and method for public transport passenger flow |
| CN104112309A (en) * | 2014-08-01 | 2014-10-22 | 西安电子科技大学 | Device and method for automatically recording passenger flow of bus by adopting video monitor |
| CN105512720A (en) * | 2015-12-15 | 2016-04-20 | 广州通达汽车电气股份有限公司 | Public transport vehicle passenger flow statistical method and system |
| CN105847646A (en) * | 2016-05-11 | 2016-08-10 | 陕西法士特齿轮有限责任公司 | Commercial vehicle intelligent monitoring system |
| CN106650593A (en) * | 2016-09-30 | 2017-05-10 | 王玲 | Passenger flow statistical method and device |
| CN107832731A (en) * | 2017-11-24 | 2018-03-23 | 无锡职业技术学院 | Motor passenger vehicle number wireless supervisory control system based on GSM |
| CN108944447A (en) * | 2018-06-27 | 2018-12-07 | 安徽佐泽智能科技有限公司 | Intelligent management system for preventing overload of passenger car |
| CN110717352A (en) * | 2018-07-11 | 2020-01-21 | 杭州海康威视系统技术有限公司 | Platform passenger flow volume statistical method, server and image acquisition equipment |
| CN110848897A (en) * | 2020-01-16 | 2020-02-28 | 恒大智慧科技有限公司 | Intelligent air conditioner adjusting method and computer readable storage medium |
| CN111319578A (en) * | 2018-12-17 | 2020-06-23 | 现代自动车株式会社 | Vehicle and control method thereof |
| US10699572B2 (en) | 2018-04-20 | 2020-06-30 | Carrier Corporation | Passenger counting for a transportation system |
| CN111932411A (en) * | 2020-09-30 | 2020-11-13 | 深圳市城市交通规划设计研究中心股份有限公司 | Method and device for determining urban land function and terminal equipment |
-
2008
- 2008-09-23 CN CNU2008200303752U patent/CN201255897Y/en not_active Expired - Fee Related
Cited By (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101877058B (en) * | 2010-02-10 | 2012-07-25 | 杭州海康威视软件有限公司 | People flow rate statistical method and system |
| CN101877058A (en) * | 2010-02-10 | 2010-11-03 | 杭州海康威视软件有限公司 | People flow rate statistical method and system |
| CN102514489A (en) * | 2011-12-16 | 2012-06-27 | 张翔宇 | Automotive display controller for avoiding traffic accidents due to overload of passengers |
| CN103021059A (en) * | 2012-12-12 | 2013-04-03 | 天津大学 | Video-monitoring-based public transport passenger flow counting method |
| CN103440763A (en) * | 2013-08-09 | 2013-12-11 | 百灵时代传媒集团有限公司 | Auxiliary method and device for taking buses |
| CN103440763B (en) * | 2013-08-09 | 2016-07-13 | 百灵时代传媒集团有限公司 | A kind of bus takes householder method and device |
| CN104021605A (en) * | 2014-04-16 | 2014-09-03 | 湖州朗讯信息科技有限公司 | Real-time statistics system and method for public transport passenger flow |
| CN104112309A (en) * | 2014-08-01 | 2014-10-22 | 西安电子科技大学 | Device and method for automatically recording passenger flow of bus by adopting video monitor |
| CN105512720B (en) * | 2015-12-15 | 2018-05-08 | 广州通达汽车电气股份有限公司 | A kind of public transit vehicle passenger flow statistics method and system |
| CN105512720A (en) * | 2015-12-15 | 2016-04-20 | 广州通达汽车电气股份有限公司 | Public transport vehicle passenger flow statistical method and system |
| CN105847646A (en) * | 2016-05-11 | 2016-08-10 | 陕西法士特齿轮有限责任公司 | Commercial vehicle intelligent monitoring system |
| CN106650593A (en) * | 2016-09-30 | 2017-05-10 | 王玲 | Passenger flow statistical method and device |
| CN107832731A (en) * | 2017-11-24 | 2018-03-23 | 无锡职业技术学院 | Motor passenger vehicle number wireless supervisory control system based on GSM |
| CN107832731B (en) * | 2017-11-24 | 2019-12-24 | 无锡职业技术学院 | GSM-based wireless passenger car passenger number monitoring system |
| US10699572B2 (en) | 2018-04-20 | 2020-06-30 | Carrier Corporation | Passenger counting for a transportation system |
| CN108944447A (en) * | 2018-06-27 | 2018-12-07 | 安徽佐泽智能科技有限公司 | Intelligent management system for preventing overload of passenger car |
| CN110717352A (en) * | 2018-07-11 | 2020-01-21 | 杭州海康威视系统技术有限公司 | Platform passenger flow volume statistical method, server and image acquisition equipment |
| CN110717352B (en) * | 2018-07-11 | 2022-05-31 | 杭州海康威视系统技术有限公司 | Platform passenger flow volume statistical method, server and image acquisition equipment |
| CN111319578A (en) * | 2018-12-17 | 2020-06-23 | 现代自动车株式会社 | Vehicle and control method thereof |
| CN111319578B (en) * | 2018-12-17 | 2023-10-24 | 现代自动车株式会社 | Vehicle and control method thereof |
| CN110848897A (en) * | 2020-01-16 | 2020-02-28 | 恒大智慧科技有限公司 | Intelligent air conditioner adjusting method and computer readable storage medium |
| CN111932411A (en) * | 2020-09-30 | 2020-11-13 | 深圳市城市交通规划设计研究中心股份有限公司 | Method and device for determining urban land function and terminal equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN201255897Y (en) | Human flow monitoring device for bus | |
| CN102542289B (en) | Pedestrian volume statistical method based on plurality of Gaussian counting models | |
| CN108986064B (en) | People flow statistical method, equipment and system | |
| CN103218816B (en) | A kind of crowd density estimation method and people flow rate statistical method based on video analysis | |
| US12002225B2 (en) | System and method for transforming video data into directional object count | |
| Wang et al. | Review on vehicle detection based on video for traffic surveillance | |
| CN104637058B (en) | A kind of volume of the flow of passengers identify statistical methods based on image information | |
| CN110298300B (en) | Method for detecting vehicle illegal line pressing | |
| CN103310444B (en) | A kind of method of the monitoring people counting based on overhead camera head | |
| CN104933710A (en) | Intelligent analysis method of store people stream track on the basis of surveillance video | |
| CN101882217B (en) | Target classification method of video image and device | |
| US9449511B2 (en) | System and method for automated sequencing of vehicle under low speed conditions from video | |
| CN103324955A (en) | Pedestrian detection method based on video processing | |
| CN106845620B (en) | A kind of passenger flow counting method based on quene state analysis | |
| CN112800860A (en) | Event camera and visual camera cooperative high-speed scattered object detection method and system | |
| CN107845264A (en) | A kind of volume of traffic acquisition system and method based on video monitoring | |
| CN107563310A (en) | A kind of lane change detection method violating the regulations | |
| CN102855508B (en) | Opening type campus anti-following system | |
| CN102254428A (en) | Traffic jam detection method based on video processing | |
| US20230154192A1 (en) | Person flow analysis method and system, electronic device, and readable storage medium | |
| CN112257520A (en) | People flow statistical method, device and system | |
| Pan et al. | Traffic surveillance system for vehicle flow detection | |
| CN107590431B (en) | Quantity counting method and device based on image recognition | |
| Park et al. | Vision-based surveillance system for monitoring traffic conditions | |
| CN116311166A (en) | Traffic obstacle recognition method and device and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20090610 Termination date: 20091023 |