CN201429904Y - Statistical system of traffic flow - Google Patents
Statistical system of traffic flow Download PDFInfo
- Publication number
- CN201429904Y CN201429904Y CN2009200336981U CN200920033698U CN201429904Y CN 201429904 Y CN201429904 Y CN 201429904Y CN 2009200336981 U CN2009200336981 U CN 2009200336981U CN 200920033698 U CN200920033698 U CN 200920033698U CN 201429904 Y CN201429904 Y CN 201429904Y
- Authority
- CN
- China
- Prior art keywords
- traffic flow
- vehicle
- background
- image
- image processor
- 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
- Traffic Control Systems (AREA)
Abstract
The utility model relates to the technical field of the traffic flow statistics, in particular to a statistical system of traffic flow. The traffic flow detection system in the prior art has the defects of complex structure, inconvenient operation and vulnerability to the outside environment. The statistical system of traffic flow comprises a video capture device, an image processor and a vehiclecounting device, wherein the image processor can process the imaged collected by the video capture device through background generation, background refreshing and vehicle edge detection, the input endof the image processor is connected with the video capture device, and the output end of the image process is connected with the vehicle counting device. The utility model has the advantages of simple structure, free control, stable operation and low vulnerability to the outside.
Description
One, technical field:
The utility model relates to magnitude of traffic flow statistical technique field, is specifically related to a kind of magnitude of traffic flow statistical system.
Two, background technology:
From nineteen sixties, along with the progress of city of the worldization with as the popularizing of the automobile of the daily main vehicles, problems such as congested in traffic aggravation, frequent accidents, traffic environment deterioration all are on the rise.Automobile quantity sharply increases, and causes existing road can not satisfy the needs of economic growth far away, and traffic progressively worsens.Under such overall background, intelligent transportation system has produced.Intelligent transportation system is the best measure of the solution urban traffic blocking problem of generally acknowledging in the world at present, based on video in real time, traffic flow analysis is one of gordian technique of intelligent transportation system accurately, also is the prerequisite that realizes that intelligent transportation is induced and controlled.
In intelligent transportation system, magnitude of traffic flow detection subsystem is very important ingredient in the whole intelligent transportation system, the vehicle that it mainly is responsible in the situation of urban traffic road is monitored and information acquisition, and the master data of subsequent treatment is provided for intelligent transportation system.In intelligent transportation system, intelligent road management and running when real-time traffic flow information feedback can take place for road congestion crest period or emergency traffic accident provide foundation, a passage is blocked up thereby reduce, and improves the road surface utilization factor, the existing transportation network of reasonable use.
The method that the at present domestic common magnitude of traffic flow detects has ultrasound examination, infrared detection, the detection of toroidal inductive circle and Computer Vision Detection etc.The ultrasound examination precision is not high, is subject to vehicle and blocks influence with the pedestrian, the distance short (generally being no more than 12m) of detection; Infrared detection is subjected to the influence of the thermal source of vehicle own, and is antimierophonic indifferent, and accuracy of detection is not high; Toroidal inductive circle accuracy of detection height, but require to be arranged in the civil structure of road surface, road pavement has damage, construction and installation inconvenience.
Three, utility model content:
The utility model provides a kind of magnitude of traffic flow statistical system, to overcome the traffic flow detection system complex structure that prior art exists, operates inconvenience, is subjected to the bigger shortcoming of ectocine.
A kind of magnitude of traffic flow statistical system comprises video collector, it is characterized in that: this system comprises that also the image that video collector is collected generates, refreshes the image processor and the car statistics device of background, vehicle edge detection processing through background; The input end of image processor is connected with video collector, and output terminal is connected with the car statistics device.
The utility model provides a kind of system that utilizes video image to detect the magnitude of traffic flow, simple in structure, control freely, working stability, be subjected to external influence less.Native system can be given image processor with the vehicle flowrate transmission of video images of video collector real-time recording, finishes wagon flow quantitative statistics in the surveyed area; Also can finish wagon flow quantitative statistics in the surveyed area by the vehicle flowrate video image that the displaying video collector records in advance.
Four, description of drawings:
Fig. 1 is a structured flowchart of the present utility model;
Fig. 2 is the utility model principle flow chart;
Five, embodiment:
Below in conjunction with accompanying drawing the utility model is described further:
Referring to accompanying drawing 1, a kind of magnitude of traffic flow statistical system comprises video collector, image processor and car statistics device; Described image processor generates, refreshes background, vehicle edge detection processing with the image that video collector collects through background; The input end of image processor is connected with video collector, and output terminal is connected with the car statistics device.
Principle of work of the present utility model is: at first utilize the video collector that is arranged on the highway top that the vehicle that travels on the highway is carried out video acquisition, obtain the video sequence of traffic scene; Then video image is imported image processor and carry out the Flame Image Process that background generated, refreshed background and vehicle edge detection, vehicle image on the video discerned vehicle image is extracted, by the car statistics device realization vehicle flowrate is estimated in the motion of vehicle at last.
Native system can be given image processor with the vehicle flowrate transmission of video images of video collector real-time recording, finishes wagon flow quantitative statistics in the surveyed area; Also can finish wagon flow quantitative statistics in the surveyed area by the vehicle flowrate video image that the displaying video collector records in advance.
Referring to accompanying drawing 2, the utility model comprises the steps: the implementation method of vehicle flowrate
1. by being arranged on the video sequence that highway top video collector is carried out video acquisition to the vehicle that travels on the highway and obtained traffic scene; Video sequence is made up of the multiframe time-series image under the dynamic scene, has comprised the vehicle image and relative static background image of motion in the video sequence;
2. video collector sends video image to image processor;
3. the virtual detection line segment is set: image processor in video image about a virtual detection line segment is set between two tracks, be used for determining surveyed area.The detection line direction is vertical substantially with the traffic lane line direction, width and about width between two tracks roughly the same.
4. owing to the invariant position of image acquisition device when gathering information of vehicle flowrate, taken at short notice video image information also remains unchanged substantially.Therefore, have only vehicle to move with respect to surrounding environment in the video at short notice, thus image processor will be in video image with vehicle identification and extract:
Want earlier generation background:,, can obtain one-dimensional signal about time shaft by on time shaft, extracting a certain locational pixel for an image sequence.Suppose certain ad-hoc location x, y}, t in the time period image sequence signal be { X
1, X
2..., X
t}={ I (x, y; I), 1≤i≤t}, time discrete function X
tSatisfy a certain specific Gaussian distribution, so just can obtain the statistical information of image sequence, can calculate the distribution average and the variance of all pixel processes thus, thus generation background.Under the less demanding situation of degree of accuracy, also can pass through
Generate the background of this ad-hoc location.The generation of initial background is a basis in the native system
Calculate.Because real background road surface color value generally all can change, if vehicle is detected with fixing background, in case the difference of real background and fixed background is excessive, what no matter adopt in the detection line zone so is background or vehicle, the difference of they and fixed background all will also just all be identified as vehicle above threshold values.Can reduce the recognition success rate of system like this.Image processor compares by the continual renovation of image sequence and to background color value, judges the current background and the hsv color value of initial background change whether surpass the setting threshold values; If changing value surpasses when setting threshold values, image processor adopts current background as a setting, otherwise then adopts initial background as a setting.Image processor dynamically refreshes the background in the detection line zone, improves system recognition rate.
At generation background and after constantly refreshing background, image processor can utilize edge detection method that vehicle image is discerned and extracted from the background that constantly refreshes: at first the brightness changing value according to present frame in the surveyed area and background frames obtains current brightness variation gray-scale map, changes the marginal information that gray-scale map calculates the vehicle wheel profile according to brightness then.Suppose that certain a bit is k on the current vehicle wheel profile, the consecutive point on the same outline line are k+1 (needing only on the same line all the time along unidirectional marginal information), and make Δ V=V
K+1-V
KIf, | Δ V|＞10 just make E
k=1, otherwise E
k=0, wherein, V
k, V
K+1Represent the lightness that k point and k+1 are ordered respectively, Δ V represents brightness changing value, E
kRepresent the marginal information of this point, E
k=1 this point of expression is confirmed to be vehicle edge, E
kThe non-vehicle edge of=0 expression.Count in edge on the calculating vehicle outline line then, during statistics, and the E that all are continuous
k=1 point is only calculated a marginal point.Image processor detects the marginal information on all vehicle wheel profiles in the surveyed area, and vehicle image is extracted from background image.
5. picture processor is transferred to the car statistics device with the marginal information of the vehicle wheel profile in the surveyed area, and the car statistics device is to the summation of the marginal information on all vehicle test lines in this surveyed area, and its value is E; The car statistics device is set a marginal information threshold values T in advance.Explanation has the vehicle passing detection zone when E＞T, car statistics device counting; Explanation does not have the vehicle passing detection zone when E＜T, and the car statistics device is not counted; Realized wagon flow quantitative statistics in the surveyed area.
Claims (1)
1. a magnitude of traffic flow statistical system comprises video collector, it is characterized in that: this system comprises that also the image that video collector is collected generates, refreshes the image processor and the car statistics device of background, vehicle edge detection processing through background; The input end of image processor is connected with video collector, and output terminal is connected with the car statistics device.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009200336981U CN201429904Y (en) | 2009-06-29 | 2009-06-29 | Statistical system of traffic flow |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009200336981U CN201429904Y (en) | 2009-06-29 | 2009-06-29 | Statistical system of traffic flow |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN201429904Y true CN201429904Y (en) | 2010-03-24 |
Family
ID=42033644
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2009200336981U Expired - Fee Related CN201429904Y (en) | 2009-06-29 | 2009-06-29 | Statistical system of traffic flow |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN201429904Y (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102542799A (en) * | 2011-12-19 | 2012-07-04 | 中山大学 | Line acquisition video vehicle detector based on pavement marker and detecting method thereof |
| CN102724484A (en) * | 2012-06-25 | 2012-10-10 | 中国科学院自动化研究所 | Bus stop people monitoring device and monitoring method thereof |
| CN104036639A (en) * | 2014-06-20 | 2014-09-10 | 上海理工大学 | Traffic flow statistics method |
| CN105447479A (en) * | 2015-12-29 | 2016-03-30 | 安徽海兴泰瑞智能科技有限公司 | Traffic state video monitoring method for high-speed bayonet road |
-
2009
- 2009-06-29 CN CN2009200336981U patent/CN201429904Y/en not_active Expired - Fee Related
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102542799A (en) * | 2011-12-19 | 2012-07-04 | 中山大学 | Line acquisition video vehicle detector based on pavement marker and detecting method thereof |
| CN102724484A (en) * | 2012-06-25 | 2012-10-10 | 中国科学院自动化研究所 | Bus stop people monitoring device and monitoring method thereof |
| CN104036639A (en) * | 2014-06-20 | 2014-09-10 | 上海理工大学 | Traffic flow statistics method |
| CN105447479A (en) * | 2015-12-29 | 2016-03-30 | 安徽海兴泰瑞智能科技有限公司 | Traffic state video monitoring method for high-speed bayonet road |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103646534B (en) | A kind of road real-time traffic accident risk control method | |
| US20230258791A1 (en) | Lane alignment detection method based on millimeter wave radar data | |
| CN101514993B (en) | Vehicle speed measurement device based on linear array CCD camera | |
| Hussein et al. | Automated pedestrian safety analysis at a signalized intersection in New York City: Automated data extraction for safety diagnosis and behavioral study | |
| CN104599502A (en) | Method for traffic flow statistics based on video monitoring | |
| CN105493502A (en) | Video monitoring method, video monitoring system, and computer program product | |
| CN103617412A (en) | Real-time lane line detection method | |
| CN103500322A (en) | Automatic lane line identification method based on low-altitude aerial images | |
| CN103150908B (en) | Average vehicle speed detecting method based on video | |
| CN103577809A (en) | Ground traffic sign real-time detection method based on intelligent driving | |
| CN110688922A (en) | Deep learning-based traffic jam detection system and detection method | |
| CN110992693A (en) | Deep learning-based traffic congestion degree multi-dimensional analysis method | |
| CN103129468A (en) | Vehicle-mounted roadblock recognition system and method based on laser imaging technique | |
| CN103198713A (en) | Traffic accident reduction vehicle regulation and control method based on traffic data and weather data | |
| CN103164958B (en) | Method and system for vehicle monitoring | |
| CN201429904Y (en) | Statistical system of traffic flow | |
| CN103313039B (en) | A kind of freeway tunnel entrance security prompt device on daytime and reminding method | |
| CN102426785A (en) | Traffic flow information perception method based on contour and local characteristic point and system thereof | |
| CN111027447B (en) | Road overflow real-time detection method based on deep learning | |
| CN103593981A (en) | Vehicle model identification method based on video | |
| CN114822029B (en) | Bridge deck traffic flow load space-time distribution reconstruction method, system and device | |
| CN104504363A (en) | Real-time identification method of sidewalk on the basis of time-space correlation | |
| CN106114505A (en) | A kind of front truck anti-collision warning method of vehicle DAS (Driver Assistant System) | |
| CN104331687B (en) | A kind of behavioral value method of not fastening the safety belt based on vehicular video analysis | |
| CN104658272A (en) | Street traffic volume statistics and sped measurement method based on binocular stereo vision |
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: 20100324 Termination date: 20100629 |