CN102637361A - Vehicle type distinguishing method based on video - Google Patents
Vehicle type distinguishing method based on video Download PDFInfo
- Publication number
- CN102637361A CN102637361A CN2012100960473A CN201210096047A CN102637361A CN 102637361 A CN102637361 A CN 102637361A CN 2012100960473 A CN2012100960473 A CN 2012100960473A CN 201210096047 A CN201210096047 A CN 201210096047A CN 102637361 A CN102637361 A CN 102637361A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- video
- virtual
- coils
- time
- 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.)
- Pending
Links
Images
Landscapes
- Traffic Control Systems (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a vehicle type distinguishing method based on a video. The length of a vehicle is solved by utilizing related technologies in visual detection and image processing, thus the vehicle type is distinguished. Compared with the prior art, the method disclosed by the invention can be used for distinguishing all the vehicle types in the range of the video without being limited by environments and can be used for reliably distinguishing the vehicle type in real time; and the method disclosed by the invention is easy to realize, higher in accuracy, wide in application prospect and applicable to real-time distinguishing on the vehicle type.
Description
Technical field
The invention belongs to the video detection technology field, be specifically related to a kind of vehicle method of discrimination based on video.
Background technology
Along with development of socialist market economic, people's living standard is greatly improved, and the quantity of motor vehicles also increases sharply thereupon.Traffic problems such as brought traffic congestion, traffic hazard to take place frequently thus, traffic environment worsens, toll mode is chaotic, traffic administration falls behind, thus a kind of on a large scale, comprehensive play a role in real time, the comprehensive traffic transportation management system just arises at the historic moment accurately and efficiently.Intelligent traffic system (Intelligent Transportation System is called for short ITS) produces under this condition just.Type of vehicle is differentiated, and abbreviates vehicle as and differentiates, and as an important branch among the ITS, has broad application prospects at aspects such as the vehicle that hits out against theft, standard traffic order, large parking lot management, highway automatic charging, magnitude of traffic flow statistics.At present, accurately differentiate vehicle and be still a newer problem, people constantly exploring simply, easily and fast recognition methods.
Be last up till now, having formed to utilize sensors such as infrared ray, toroid winding and radar is a series of vehicle identification and classification methods of means.These method principles are simple, and clear physics conception is clear and definite, implements to be easier to.But also exist hardware system complicated, the adaptive capacity to environment of system is relatively poor, so have defectives such as failure rate is higher, inconvenient maintenance, is difficult in actual use promote.
Summary of the invention
Defective or deficiency to prior art exists the objective of the invention is to, and a kind of vehicle method of discrimination based on video is provided, and this method can realize in real time, differentiate reliably all type of vehicle in the range of video.
In order to realize above-mentioned task, the present invention takes following technical solution:
A kind of vehicle method of discrimination based on video is characterized in that, implements according to the following step:
Step 1 in video sequence image, manually is provided with two virtual coils on perpendicular to the lane line direction, the video image with vehicle during through two virtual coils carries out binaryzation; Adopt a kind of known video camera geometric calibration method, obtain the mapping relations between the capable and actual range of image pixel, i.e. mapping table.Thereby can draw two actual ranges between the virtual coil.
Step 2 when the tailstock passes through two virtual coils, is noted two time frames respectively, thereby obtains vehicle through the used time of virtual coil.Distance between two coils that draw in the integrating step one again, the average velocity in the time of can obtaining vehicle through virtual coil.
Step 3 is provided with a fixation mark line between two virtual coils, write down the headstock and the tailstock time frame when this fixed position respectively, thus (this moment vehicle just in time pass by a vehicle commander apart from) used time when obtaining vehicle through this mark line.The speed of asking in the integrating step two can be obtained vehicle length again.
Step 4 utilizes vehicle length to differentiate its type.
Vehicle method of discrimination based on video of the present invention compared with prior art, can be discerned all type of vehicle in the range of video, does not receive environmental restraint, can carry out in real time type of vehicle, differentiate reliably.And be easy to realize, accuracy is higher, is well suited for the real time discriminating type of vehicle, have broad application prospects.
Description of drawings
Below in conjunction with accompanying drawing and specific embodiment the present invention is done further detailed description.
Fig. 1 is the dissimilar same highway section of vehicle process.
Fig. 2 is the virtual coil synoptic diagram in the video image.
Binaryzation design sketch when Fig. 3 is dolly and cart process fixation mark line.Wherein, the binaryzation design sketch when two width of cloth figure are dolly process fixation mark line among Fig. 3 (a), the binaryzation design sketch when two width of cloth figure are cart process fixation mark line among Fig. 3 (b).
Fig. 4 differentiates process flow diagram for vehicle.
Below in conjunction with accompanying drawing and embodiment the present invention is done further detailed description.
Embodiment
Present embodiment provides a kind of vehicle method of discrimination based on video, utilizes the correlation technique of Video Detection and Flame Image Process, obtains vehicle length, thereby type of vehicle is carried out in real time, differentiates reliably.Specifically follow these steps to carry out:
Step 1 in video sequence image, manually is provided with two virtual coils on perpendicular to the lane line direction, as shown in Figure 2.Video image with vehicle during through two virtual coils carries out binaryzation; Adopt a kind of known video camera geometric calibration method (one Chinese patent application: " the video camera geometric calibration method under a kind of linear model " (open (bulletin) number: CN102222332A); Obtain the mapping relations between the pixel column and actual range in the image; It is mapping table; According to this mapping table, can obtain two physical length M between the virtual coil.
Step 2 when the tailstock passes through first virtual coil, is write down present image frame number N
1When the tailstock passes through second virtual coil, write down present image frame number N
2Because playing the used time of 25 frame video images is 1 second, can know that vehicle passes through two used times of virtual coil and is: t
0=(N
2-N
1)/25 second.Therefore, the average velocity in the time of can obtaining vehicle through two virtual coils is: v=M/t
0
Step 3 is provided with a fixation mark line between two virtual coils, when this fixation mark line of headstock process, write down current images frame number N
3, and then when this fixation mark line of tailstock process, write down current images frame number N
4So vehicle is N through the used time frame number of this mark line
4-N
3, can draw thus, vehicle is through (being that vehicle is passed by a used time of vehicle length) t:t=(N of used time of this mark line
4-N
3)/25 second.Obtained the speed v of vehicle when two virtual coils again in the step 2, thereby can obtain the vehicle length L: be i.e. L=v * t.
Step 4 utilizes the vehicle length that draws that its type is differentiated.
As everyone knows, the length of dissimilar vehicles is different, and vehicle length is an important symbol of type of vehicle, and identification is got up also relatively more directly perceived, easily.Therefore, can utilize vehicle length that its type is differentiated exactly.
Vehicle of the present invention is differentiated algorithm, and the vehicle commander's size according to each car is divided into car greatly, in, little 3 types.Large car mainly comprises motorbus, lorry, and engineering truck; Compact car mainly comprises various two boxes, minivan, and some miniature picking-up vehicles; In-between car is meant except large-scale and other vehicles the compact car, mainly comprises some jubilee wagons and station wagon.Through the vehicle commander vehicle is carried out determinate judgement, its decision logic is shown in (1) formula:
Hypothesis in formula (1), compact car is 1, and in-between car is 2, and large car is 3.Visible through formula (1);, the vehicle commander directly can conclude that this car is a compact car when being less than or equal to 700 centimetres; And can judge directly that this car is a large car during more than or equal to 1400 centimetres as the vehicle commander, when the vehicle commander is within these two the determined scopes of value, then can be judged as in-between car.
Such judged result is real-time, and is accurately, reliable.Its decision logic relation is as shown in Figure 4.
It below is the specific embodiment that the inventor provides.
Embodiment:
In video sequence, on perpendicular to the lane line direction, two virtual coils are set, obtain mapping table according to a kind of labeling method, obtaining two actual ranges between the virtual coil is 1864.67 centimetres.
Embodiment 1: known in video sequence, and there is a dolly to pass through.When the tailstock passed through first virtual coil, noting the present image frame number was 20, and when the tailstock passed through second virtual coil, noting the present image frame number was 38, so vehicle is 0.72 second through the used time of virtual coil.So the speed of vehicle during through virtual coil is 2589.82 cels.When headstock during through the fixation mark line, current frame number is 24, and when the tailstock during through fixing calibration line, current frame number is 29, so vehicle is 0.20 second through the used time of mark line.The binary image of vehicle through the fixation mark line is shown in Fig. 3 (a).Thereby obtain vehicle length is 517.36 centimetres, can know that according to judging it is a dolly, with met in practice.
Embodiment 2: known in video sequence, and there is a cart to pass through.When the tailstock passed through first virtual coil, noting the present image frame number was 68, and when the tailstock passed through second virtual coil, noting the present image frame number was 93, so vehicle is 1 second through the used time of virtual coil.So the speed of vehicle during through virtual coil is 1864.67 cels.When headstock during through the fixation mark line, current frame number is 71, and when the tailstock during through fixing calibration line, current frame number is 92, so vehicle is 0.84 second through the used time of mark line.The binary image of vehicle through the fixation mark line is shown in Fig. 3 (b).Thereby obtain vehicle length is 1566.32 centimetres, can know that according to judging it is a cart, with met in practice.
Claims (1)
1. the vehicle method of discrimination based on video is characterized in that, implements according to the following step:
Step 1 in video sequence image, manually is provided with two virtual coils on perpendicular to the lane line direction, the video image with vehicle during through two virtual coils carries out binaryzation; Adopt known video camera geometric calibration method, obtain the mapping relations between the capable and actual range of image pixel, i.e. mapping table; Thereby can draw two actual ranges between the virtual coil;
Step 2 when the tailstock passes through two virtual coils, is noted two time frames respectively, thereby obtains vehicle through the used time of virtual coil; Distance between two coils that draw in the integrating step one again, the average velocity in the time of can obtaining vehicle through virtual coil;
Step 3; A fixation mark line is set between two virtual coils, writes down the headstock and the tailstock time frame when this fixed position respectively, thus the used time when obtaining vehicle through this mark line; The speed of asking in the integrating step two can be obtained vehicle length again;
Step 4 utilizes vehicle length to differentiate its type.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2012100960473A CN102637361A (en) | 2012-04-01 | 2012-04-01 | Vehicle type distinguishing method based on video |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2012100960473A CN102637361A (en) | 2012-04-01 | 2012-04-01 | Vehicle type distinguishing method based on video |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102637361A true CN102637361A (en) | 2012-08-15 |
Family
ID=46621741
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2012100960473A Pending CN102637361A (en) | 2012-04-01 | 2012-04-01 | Vehicle type distinguishing method based on video |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102637361A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103593981A (en) * | 2013-01-18 | 2014-02-19 | 西安通瑞新材料开发有限公司 | Vehicle model identification method based on video |
| CN105118141A (en) * | 2015-09-02 | 2015-12-02 | 昆山古鳌电子机械有限公司 | Banknote authenticity identification apparatus |
| CN105869109A (en) * | 2016-03-28 | 2016-08-17 | 长安大学 | Method for differentiating parking vehicles and fallen objects based on inverse projective planes of different heights |
| CN108346303A (en) * | 2018-04-09 | 2018-07-31 | 天津中兴智联科技有限公司 | A kind of implementation method that bus identifies and positions and realize system |
| CN108426902A (en) * | 2018-03-14 | 2018-08-21 | 中广核贝谷科技股份有限公司 | A kind of moving vehicle method for detecting position based on video |
| CN112629713A (en) * | 2020-10-22 | 2021-04-09 | 北京工业大学 | Method for detecting vehicle type corresponding to sensor data |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030189500A1 (en) * | 2002-04-04 | 2003-10-09 | Lg Industrial Systems Co., Ltd. | System for determining kind of vehicle and method therefor |
| CN101042802A (en) * | 2006-03-23 | 2007-09-26 | 安捷伦科技有限公司 | Traffic information sensor and method and system for traffic information detecting |
| CN101206229A (en) * | 2007-12-11 | 2008-06-25 | 长安大学 | Method for matching vehicle speed based on linear array CCD image |
| US20080205710A1 (en) * | 2005-09-27 | 2008-08-28 | Koninklijke Philips Electronics, N.V. | Motion Detection Device |
| CN101604448A (en) * | 2009-03-16 | 2009-12-16 | 北京中星微电子有限公司 | A kind of speed-measuring method of moving target and system |
| CN101615342A (en) * | 2008-06-27 | 2009-12-30 | 青岛海信电子产业控股股份有限公司 | A kind of vehicle checking method |
| CN102222332A (en) * | 2011-05-19 | 2011-10-19 | 长安大学 | Geometric calibration method of camera under linear model |
-
2012
- 2012-04-01 CN CN2012100960473A patent/CN102637361A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030189500A1 (en) * | 2002-04-04 | 2003-10-09 | Lg Industrial Systems Co., Ltd. | System for determining kind of vehicle and method therefor |
| US20080205710A1 (en) * | 2005-09-27 | 2008-08-28 | Koninklijke Philips Electronics, N.V. | Motion Detection Device |
| CN101042802A (en) * | 2006-03-23 | 2007-09-26 | 安捷伦科技有限公司 | Traffic information sensor and method and system for traffic information detecting |
| CN101206229A (en) * | 2007-12-11 | 2008-06-25 | 长安大学 | Method for matching vehicle speed based on linear array CCD image |
| CN101615342A (en) * | 2008-06-27 | 2009-12-30 | 青岛海信电子产业控股股份有限公司 | A kind of vehicle checking method |
| CN101604448A (en) * | 2009-03-16 | 2009-12-16 | 北京中星微电子有限公司 | A kind of speed-measuring method of moving target and system |
| CN102222332A (en) * | 2011-05-19 | 2011-10-19 | 长安大学 | Geometric calibration method of camera under linear model |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103593981A (en) * | 2013-01-18 | 2014-02-19 | 西安通瑞新材料开发有限公司 | Vehicle model identification method based on video |
| CN103593981B (en) * | 2013-01-18 | 2016-08-24 | 西安通瑞新材料开发有限公司 | A kind of model recognizing method based on video |
| CN105118141A (en) * | 2015-09-02 | 2015-12-02 | 昆山古鳌电子机械有限公司 | Banknote authenticity identification apparatus |
| CN105869109A (en) * | 2016-03-28 | 2016-08-17 | 长安大学 | Method for differentiating parking vehicles and fallen objects based on inverse projective planes of different heights |
| CN105869109B (en) * | 2016-03-28 | 2018-12-07 | 长安大学 | Parking based on different height inverse projection face and leave object differentiating method |
| CN108426902A (en) * | 2018-03-14 | 2018-08-21 | 中广核贝谷科技股份有限公司 | A kind of moving vehicle method for detecting position based on video |
| CN108426902B (en) * | 2018-03-14 | 2020-11-10 | 中广核贝谷科技有限公司 | Moving vehicle position detection method based on video |
| CN108346303A (en) * | 2018-04-09 | 2018-07-31 | 天津中兴智联科技有限公司 | A kind of implementation method that bus identifies and positions and realize system |
| CN108346303B (en) * | 2018-04-09 | 2021-06-11 | 天津中兴智联科技有限公司 | Method and system for realizing bus identification and positioning |
| CN112629713A (en) * | 2020-10-22 | 2021-04-09 | 北京工业大学 | Method for detecting vehicle type corresponding to sensor data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102637361A (en) | Vehicle type distinguishing method based on video | |
| US8041079B2 (en) | Apparatus and method for detecting obstacle through stereovision | |
| CN104021541A (en) | Vehicle-to-vehicle distance calculation apparatus and method | |
| CN103794056A (en) | Vehicle type accurate classification system and method based on real-time double-line video stream | |
| CN103150908B (en) | Average vehicle speed detecting method based on video | |
| CN105930787A (en) | Vehicle door opening early-warning method | |
| CA2885019A1 (en) | Robust windshield detection via landmark localization | |
| CN104029680A (en) | Lane departure warning system and method based on monocular camera | |
| CN103971097A (en) | Vehicle license plate recognition method and system based on multiscale stroke models | |
| CN104331703A (en) | Method of monitoring vehicle running state and automobile navigation equipment for implementing method | |
| CN110176151A (en) | A kind of method, apparatus, medium and the equipment of determining parking behavior | |
| CN103164958B (en) | Method and system for vehicle monitoring | |
| CN103021182B (en) | Method and device for monitoring motor vehicle in case of regulation violation for running red light | |
| CN103077608B (en) | Method for obtaining evidence of behavior of intentionally running yellow light of driver | |
| CN107464303A (en) | A kind of electric non-stop toll method and system based on vehicle electron identifying | |
| CN110232827A (en) | The recognition methods of free flow toll vehicle type, apparatus and system | |
| US11137256B2 (en) | Parking area map refinement using occupancy behavior anomaly detector | |
| KR20150029122A (en) | Apparatus and Method for Recognizing Signs in Road | |
| CN110930715A (en) | Method and system for identifying red light running of non-motor vehicle and violation processing platform | |
| CN108172008B (en) | Method, device and system for marking parking spaces | |
| CN106652462A (en) | Illegal parking management system based on Internet | |
| CN103337175A (en) | Vehicle type recognition system based on real-time video steam | |
| CN102779413A (en) | Motor vehicle characteristics data acquisition analyzer and analyzing method | |
| JP4961305B2 (en) | Vehicle monitoring device for toll road automatic toll booth | |
| CN106355664B (en) | Expressway tol lcollection management system, method and device based on vehicle electron identifying |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20120815 |