CN102280028B - Vehicle monitoring method based on dynamic background analysis and monitoring tape scanning - Google Patents
Vehicle monitoring method based on dynamic background analysis and monitoring tape scanning Download PDFInfo
- Publication number
- CN102280028B CN102280028B CN 201110118812 CN201110118812A CN102280028B CN 102280028 B CN102280028 B CN 102280028B CN 201110118812 CN201110118812 CN 201110118812 CN 201110118812 A CN201110118812 A CN 201110118812A CN 102280028 B CN102280028 B CN 102280028B
- Authority
- CN
- China
- Prior art keywords
- vehicle
- monitoring
- image
- space
- picture
- 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)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention provides a vehicle monitoring method based on dynamic background analysis and monitoring tape scanning, in which a video file photographed for a one-way lane is analyzed, so as to count the number of passing vehicles, and additionally count information such as the colors of the vehicles, the speeds of the vehicles and the time for passing, and the information is stored into a database for referring to any time. According to the invention, neighboring pictures are processed by difference analysis, and then binarized, thus realizing dynamic background processing and overcoming the influence of background change; therefore the method is more effective. When the vehicles are monitored, a monitoring tape is selected, which may reflect information of all vehicles on the monitoring tape, thereby the processing is simpler, and the method is more efficient; and the vehicles are counted by scanning gaps between vehicles, thus obtaining the number of the vehicles.
Description
Technical field
The present invention is the intelligent vehicle monitoring system based on dynamic background extracts and the monitoring band scans, be used for measuring or test process control, in real-time traffic, can be applied to traffic flow monitoring, control, thereby can analyze the video under recording simultaneously information of vehicles is collected.
Technical background
Vehicle is monitored the two kinds of methods that generally comprise, the one, by being installed on ground, adds up hardware facility, and another kind of method is capture video or image, carries out Treatment Analysis by computing machine then.The former obtainable information is less, and only is applicable to some toll site usually; The latter includes a large amount of available informations, and usable range is wide, can camera be installed at road and take, and handles by computer then.
Based on the general background that all is based on a static state of video or image processing, then other image comparison backgrounds are analyzed, determine the number of vehicle by some characteristic parameters of coupling vehicle.But in the real process, background is because difference all can take place in the variation of weather or sunlight, and the feature of different vehicle has very big difference simultaneously, is easy to cause erroneous judgement.This method to reach the function of Dynamic Extraction background, is monitored scanning to the picture edge by adopting the difference between the contrast adjacent image simultaneously, can be easy to judge the number of vehicle, simple and efficient.
Summary of the invention
The present invention reads one section traffic monitoring video, and monitoring dynamically shows monitoring result by the vehicle of monitoring point, and the number of registration of vehicle, simultaneously can subsidiary information of vehicles (comprising speed, color, the time of passing through).The side that passage can simulated automotive passes by passage or takes two-way lane.May have a plurality of targets and appear at simultaneously in the picture, the vehicle size size may be different.The source of video: the video camera level is installed in passage one side, and its installation site distance to target is moderate; Passage background frame monitored when driftlessness passes through is static.
The technical scheme that the above-mentioned technical matters that the present invention will solve adopts is as follows:
A kind of vehicle monitoring method based on dynamic background analysis and the scanning of monitoring band, step comprises:
1) reads one section traffic monitoring video: at first open the target video source, target is play, regularly obtain the image in the video then, preserve image;
2) in real time image is handled: extract the picture of the described image of record, the difference of carrying out between adjacent picture is handled, and obtains the only black and white binaryzation picture of remaining vehicle shape;
3) described binaryzation picture is further handled, obtained number of vehicles, step is as follows:
Set a monitoring band at picture earlier, the monitoring band is an elongated rectangle, and the length of monitoring band is identical with the width of picture;
To the scanning of monitoring band, if scan a large amount of points, then expression has vehicle to enter the monitoring band again, if counting of scanning is almost nil, then expression does not have vehicle to enter the monitoring band;
For there being vehicle to enter monitoring band situation, monitoring starting point and the terminal point that the strip length direction obtains described a large amount of points, namely all vehicles are at starting point and the terminal point of monitoring strip length direction;
Between described starting point and terminal point, continuous images in the monitoring band is scanned, obtain the space quantity between all vehicle starting points and the terminal point; When vehicle leaves, read the number in space, the space number adds 1 and is vehicle number;
Whether the space is that the determination methods in the space between the vehicle is as follows:
After the binaryzation, it is 255 that the local image rgb value of car is arranged;
Scan between starting point and terminal point, scan a rectangular area at every turn, this rectangle region field width is the horizontal direction along image, and height is the vertical direction along image; If the point in this rectangular area is then thought the space between two cars less than threshold values; So then scanning, the terminal point until this space just can record starting point and the terminal point in this space this moment;
The terminal point in space may be another car, so from this car, according to said method, continues the next space of scanning, up to terminal point.
Comprise step 4) judgement vehicle color and car speed after the described step 3), simultaneously these information deposited into database,
Judging vehicle color, is the image that step 1) is obtained, and the rgb value of vehicle location is determined in the analysis image;
Judge vehicle speed measurement, in step 2) in the binaryzation picture that obtains, at first scan the headstock of vehicle to the length of the tailstock, then the time of being with by monitoring divided by vehicle, multiply by scale-up factor again.
Described monitoring band is located at edge of image.
Described step 2) in, for a pictures,
2.1) at first a last pictures is imported in the internal memory, as background, then photo current to importing in the internal memory, present image is scanned, the subtracting background picture obtains having only the picture of vehicle;
2.2) carry out the gray processing processing then, carry out binaryzation and obtain black and white picture;
Step 2.2) in, if because the variation of light, the difference of front and back two pictures is very big sometimes, then the influence of cancellation light.
In the influence of cancellation light in the existing technology, generally be regularly to carry out the judgement again of a background, such as to one section Video processing, if the background of image will not be known at which place will change at, that can carry out background extracting one time to image every T second, to replace the background of last time, also can control at T in second even make a mistake so.But this method efficient is not high, and big because if T gets, mistake may be many so, if but T is less, and then background process can seem redundant, consuming time frequently.
In the technical program, select a monitoring band (general monitoring band can be chosen in edge of image), the monitoring band is an elongated rectangle, the length of monitoring band and the width of image, and width can be selected voluntarily, and width is less.Monitoring is brought into line scanning, scan a large amount of points and represent that vehicle enters, when counting when almost nil, the expression vehicle leaves or do not have car to enter this moment.Starting point and terminal point when can obtain vehicle and enter after the scanning simultaneously in the monitoring band, i.e. vehicle starting point and terminal point in the horizontal direction.When entering not leave, vehicle is, between starting point and terminal point, continuous images scanned, thereby obtain more desired parameter informations that as total several spaces between the starting point, which space is the space between the vehicle, the starting point in space and terminal point; Whether have bigger car to enter, whether too near that two cars leans on is arranged, whether starting point and terminal point have had change etc.When vehicle leaves, read the number in space, the space number adds 1 and is vehicle number.Such as two spaces are arranged, three cars are arranged then.
Can carry out judgement and the car speed of vehicle color in the time of calculating vehicle simultaneously, and vehicle by the time judgement, simultaneously these information are deposited into database, so that later calling.The judgement of vehicle color is by the rgb value of vehicle location in the analysis image.The measurement of the speed of a motor vehicle at first scans vehicle headstock in picture and divided by the time of vehicle by the monitoring band, multiply by scale-up factor again and get final product then to the length of the tailstock.
With the different technical characterictic of immediate prior art be:
The processing of background, adopt adjacent image to be analyzed, to reach dynamic background process, (and technology in the past is the first few hundred frame of at first analysis image, extracts background, later image is handled all according to this background, this method length that expends time in, and after the change of background of image (change such as light, camera position changes), later image all the sum of errors mistake will occur);
Adopt monitoring band calculating vehicle, the monitoring band is one section narrow zone, but when vehicle by the time can scan, the space that obtains according to scanning and the size of vehicle compare analysis, the easier statistics of vehicle then, handle also more efficiently, add up more accurate, accuracy is higher; Perfect system not only can the calculating vehicle number, and the time that vehicle is passed through, and information such as color and speed can obtain, and use database store car information simultaneously, calls after convenient.
Description of drawings
Fig. 1 is the image that carries out after the binary conversion treatment,
Fig. 2 brings line scanning for adopting into to monitoring, obtains start line a and the finishing line b of car;
Fig. 3 .1 be have vehicle by the time situation, Fig. 3 .2 is the situation of vehicle when leaving;
Among Fig. 3 .1 and 3.2, regional c is monitoring band, wherein, and (from figure, not seeing Fig. 3 .1 and 3.2 variant)
Fig. 4 provided a kind of many cars by the time situation.A space is arranged as can be seen, so two cars is arranged between start line a and the finish line b.
Embodiment
Below in conjunction with accompanying drawing how the present invention is finished monitoring vehicle automatically and stores and be described further:
Open file earlier, read video, after selecting to play, system will analyze automatically.
(time regularly gets according to actual conditions at first regularly to obtain the image of video window oral region, general 100ms once), and store, extract picture, carry out some pre-service, comprise filtering, gray processing is then handled (present image deducts last width of cloth background image) to carrying out difference between adjacent picture, carries out binaryzation then, obtain having only the black and white binaryzation picture of vehicle imaging, as Fig. 1.
With reference to figure 2～Fig. 4, the binaryzation picture of handling, have only vehicle is handled, select a monitoring band, monitoring is brought into line scanning, scan a large amount of points and represent that vehicle enters, when counting when almost nil, the expression vehicle leaves.Starting point and terminal point when obtaining vehicle simultaneously and enter in the monitoring band, i.e. the starting point of horizontal direction, terminal point.When vehicle enter and its when not leaving, between starting point and terminal point, continuous images is scanned, thereby obtains more desired parameter informations, as total several spaces between the starting point, which space is the space between the vehicle, the starting point in space and terminal point; Whether have bigger car to enter, whether too near that two cars leans on is arranged, whether starting point and terminal point have had change etc.Need consider array modes all when vehicle enters this moment.After taking all factors into consideration, when vehicle leaves, just can get rid of some possible interference according to above information, obtain number of vehicles according to the space number.If when occurring the interference of pedestrian or motorcycle in the image, can get rid of by width and speed.
Practice significance of the present invention is:
By to the processing of the video image vehicle in monitoring objective zone in real time, provide theoretical foundation for rationally regulating the magnitude of traffic flow.
In order to better meet actual needs, the present invention is also to color and the speed of vehicle, with and time of passing through all measure, store by the access database.At any time the people obtains historical information for checking database, helps the transport solution problem to comprise some hypervelocity behaviors, illegal phenomenon such as high speed on the motorcycle.
Claims (4)
1. vehicle monitoring method based on dynamic background analysis and monitoring band scanning is characterized in that step comprises:
1) reads one section traffic monitoring video: at first open the target video source, target is play, regularly obtain the image in the video then, preserve image;
2) in real time image is handled: extract the picture of the described image of record, the difference of carrying out between adjacent picture is handled, and obtains the only black and white binaryzation picture of remaining vehicle shape;
3) described binaryzation picture is further handled, obtained number of vehicles, step is as follows:
Set a monitoring band at picture earlier, the monitoring band is an elongated rectangle, and the length of monitoring band is identical with the width of picture;
To the scanning of monitoring band, if scan a large amount of points, then expression has vehicle to enter the monitoring band again, if counting of scanning is almost nil, then expression does not have vehicle to enter the monitoring band;
For there being vehicle to enter monitoring band situation, monitoring starting point and the terminal point that the strip length direction obtains described a large amount of points, namely all vehicles are at starting point and the terminal point of monitoring strip length direction;
Between described starting point and terminal point, continuous images in the monitoring band is scanned, obtain the space quantity between all vehicle starting points and the terminal point; When vehicle leaves, read the number in space, the space number adds 1 and is number of vehicles;
Whether the space is that the determination methods in the space between the vehicle is as follows:
After the binaryzation, it is 255 that the local image rgb value of car is arranged;
Scan between starting point and terminal point, scan a rectangular area at every turn, this rectangle region field width is the horizontal direction along image, and height is the vertical direction along image; If the point in this rectangular area is then thought the space between two cars less than threshold values; So then scanning, the terminal point until this space just can record starting point and the terminal point in this space this moment;
The terminal point in space may be another car, so from this car, according to said method, continues the next space of scanning, up to terminal point.
2. the vehicle monitoring method based on dynamic background analysis and the scanning of monitoring band according to claim 1 is characterized in that comprising step 4) judgement vehicle color and car speed after the described step 3), simultaneously these information is deposited into database,
Judging vehicle color, is the image that step 1) is obtained, and the rgb value of vehicle location is determined in the analysis image;
Judge vehicle speed measurement, in step 2) in the binaryzation picture that obtains, at first scan the headstock of vehicle to the length of the tailstock, then the time of being with by monitoring divided by vehicle, multiply by scale-up factor again.
3. the vehicle monitoring method based on dynamic background analysis and the scanning of monitoring band according to claim 1 and 2 is characterized in that described monitoring band is located at edge of image.
4. the vehicle monitoring method based on dynamic background analysis and the scanning of monitoring band according to claim 1 and 2 is characterized in that described step 2) in, for a pictures,
2.1) at first a last pictures is imported in the internal memory, as background, then photo current to importing in the internal memory, present image is scanned, the subtracting background picture obtains having only the picture of vehicle;
2.2) carry out the gray processing processing then, carry out binaryzation and obtain black and white picture;
Step 2.2) in, if because the variation of light, the difference of front and back two pictures is very big sometimes, then the influence of cancellation light.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201110118812 CN102280028B (en) | 2011-05-09 | 2011-05-09 | Vehicle monitoring method based on dynamic background analysis and monitoring tape scanning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201110118812 CN102280028B (en) | 2011-05-09 | 2011-05-09 | Vehicle monitoring method based on dynamic background analysis and monitoring tape scanning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102280028A CN102280028A (en) | 2011-12-14 |
| CN102280028B true CN102280028B (en) | 2013-08-28 |
Family
ID=45105451
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201110118812 Expired - Fee Related CN102280028B (en) | 2011-05-09 | 2011-05-09 | Vehicle monitoring method based on dynamic background analysis and monitoring tape scanning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102280028B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102637360B (en) * | 2012-04-01 | 2014-07-16 | 长安大学 | Video-based road parking event detection method |
| CN108648181A (en) * | 2018-04-25 | 2018-10-12 | 佛山科学技术学院 | A kind of automobile quantity statistics method and its system based on particle filter algorithm |
| CN109189832A (en) * | 2018-08-21 | 2019-01-11 | 北京深瞐科技有限公司 | City vehicle statistical method and its device, computer-readable medium |
| CN112433261A (en) * | 2020-11-25 | 2021-03-02 | 猫岐智能科技（上海）有限公司 | Scanning imaging monitoring system and method and electric vehicle elevator entering scanning imaging monitoring system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101398932A (en) | 2008-10-16 | 2009-04-01 | 青岛海信电子产业控股股份有限公司 | Video vehicle detection method and device |
| DE102009024634A1 (en) | 2009-06-12 | 2010-03-11 | Daimler Ag | Method for determining correspondence of pixels in two stereoscopically captured images in driver assistance system of motor vehicle, involves performing correspondence determination only along scanning curve and in visualization region |
| CN101968924A (en) | 2010-10-19 | 2011-02-09 | 北方工业大学 | High-resolution image-based real-time traffic flow detection method |
-
2011
- 2011-05-09 CN CN 201110118812 patent/CN102280028B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101398932A (en) | 2008-10-16 | 2009-04-01 | 青岛海信电子产业控股股份有限公司 | Video vehicle detection method and device |
| DE102009024634A1 (en) | 2009-06-12 | 2010-03-11 | Daimler Ag | Method for determining correspondence of pixels in two stereoscopically captured images in driver assistance system of motor vehicle, involves performing correspondence determination only along scanning curve and in visualization region |
| CN101968924A (en) | 2010-10-19 | 2011-02-09 | 北方工业大学 | High-resolution image-based real-time traffic flow detection method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102280028A (en) | 2011-12-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106373426B (en) | Parking stall based on computer vision and violation road occupation for parking monitoring method | |
| CN102722704B (en) | Method and system for recognizing vehicle license plate by integrating video dynamic tracking | |
| CN108765976B (en) | Roadside parallel parking information management system and method | |
| CN104246821B (en) | Three-dimensional body detection device and three-dimensional body detection method | |
| CN109446917B (en) | Vanishing point detection method based on cascading Hough transform | |
| CN111626277B (en) | Vehicle tracking method and device based on over-station inter-modulation index analysis | |
| CN100454355C (en) | Video method for collecting information of vehicle flowrate on road in real time | |
| CN113781537B (en) | Rail elastic strip fastener defect identification method and device and computer equipment | |
| CN104183133A (en) | Method for acquiring and transmitting road traffic flow dynamic information | |
| CN102280028B (en) | Vehicle monitoring method based on dynamic background analysis and monitoring tape scanning | |
| CN109035446B (en) | Automatic charging system for parking lot | |
| CN110880205B (en) | Parking charging method and device | |
| CN115331191B (en) | Vehicle type recognition method, device, system and storage medium | |
| Medina et al. | Automated visual inspection of road surface cracks | |
| CN103778790B (en) | A kind of vehicle flowrate square wave statistic law based on video sequence | |
| CN114049306A (en) | Traffic anomaly detection system design based on image camera and high-performance display card | |
| CN114359883A (en) | Electric automobile management system | |
| CN111833463B (en) | Parking lot electronic toll collection method and system and storage medium | |
| CN104573698A (en) | Image texture main direction recognition method oriented to imaging measurement | |
| JPH1031797A (en) | Method for specifying vehicle in parking lot, system therefor, parking charge calculation method and system therefor | |
| CN114005185B (en) | Intelligent park management system based on Internet of things | |
| Radopoulou et al. | Improving patch defect detection using vision tracking on video data | |
| CN114627653B (en) | 5G intelligent barrier gate management system based on binocular recognition | |
| CN109063675A (en) | Vehicle density calculation method, system, terminal and computer readable storage medium | |
| CN102622888B (en) | Self-learning identification method and device for licence plate based on video detection |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20130828 Termination date: 20160509 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |