CN104575003B - A kind of vehicle speed detection method based on traffic surveillance videos - Google Patents
A kind of vehicle speed detection method based on traffic surveillance videos Download PDFInfo
- Publication number
- CN104575003B CN104575003B CN201310503592.4A CN201310503592A CN104575003B CN 104575003 B CN104575003 B CN 104575003B CN 201310503592 A CN201310503592 A CN 201310503592A CN 104575003 B CN104575003 B CN 104575003B
- Authority
- CN
- China
- Prior art keywords
- vehicle
- distance
- track
- pixel
- speed detection
- 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
Landscapes
- Traffic Control Systems (AREA)
- Image Processing (AREA)
Abstract
The present invention is a kind of vehicle speed detection method based on traffic surveillance videos, is comprised the following steps:1. segmentation is entered to track using track marking line method of identification, rejects unnecessary background;2. lane width information is obtained according to track picture;3. the actual range representated by each pixel is calculated;4. correction factor β is solved, to correct the actual track distance representated by each pixel;5. the light stream of vehicle movement is calculated, the movement locus of vehicle is obtained;6. vehicle actual motion distance is calculated;7. distance correction, correction factor is multiplied by by vehicle actual motion distance;8. speed is calculated.The present invention uses video image analysis technology, it is intended to reduce vehicle speed detection, the cost for the behavior record that exceeds the speed limit, while the present invention also acts as the scope effect of expansion car speed monitoring, all places for having camera can all carry out car speed monitoring.
Description
Technical field
Tested the speed field, more particularly to the car speed based on traffic surveillance videos image the present invention relates to intelligent traffic vehicle
Measuring method.
Background technology
With China's expanding economy, the popularity rate also rapidly lifting, traffic accident while vehicle increases therewith of automobile
Also come one after another.And it is one of arch-criminal that tragedy leads to drive over the speed limit.
The main method of current vehicle speed measuring has radar velocity measurement, laser velocimeter, ground sensing coil speed measuring etc..Three kinds of speed-measuring methods
The characteristics of all there is high-accuracy, but unlawful practice can not be recorded while measuring speed.Especially, ground induction coil
Needs test the speed in embedded underground coil, surface damage of satisfying the need is larger and maintenance cost is higher.
The content of the invention
The present invention is to solve problem above and produce, and is to provide a kind of car speed inspection based on traffic surveillance videos
Survey method, it is intended to reduce the cumbersome degree and cost of vehicle speed measuring.The vehicle speed measuring method based on traffic video of the invention, outside it
Hardware device only relies upon video camera, and software and hardware, which coordinates, can solve to test the speed and monitor two pieces of problems.
Innovative point of the present invention is:Propose actual range (pixel_distance)=track representated by each pixel
The number of pixels (pixel_number) of lane width in developed width (road_width)/video image.
Country has formulated a set of detailed standard for the width in track, the lane width of standardization be the invention provides
Great convenience.
Because the lane width in video image is continually changing with the shooting depth of camera, therefore representated by pixel
Distance is also relevant with the position where pixel.If pixel point coordinates is (x, y), then work as y=i(If each pixel is longitudinally wide
It is consistent with the distance representated by transverse width)When:
pixel_distance[i]=road_width/pixel_number[i]。
Vehicle speed detection method based on traffic surveillance videos, it is characterised in that comprise the following steps:
1. track picture is intercepted:One track is partitioned into from video image using track marking line method of identification.Delete many
Remaining background, obtains pure track picture.Artwork pixel used is consistent with video pixel during sectional drawing.Also Manual interception can be used
Method is obtained.
2. lane width information is obtained:The track picture 1. obtained according to step, calculates lane width, uses pixel number
(pixel_number) represent.
3. the actual range representated by per pixel is calculated:
Pixel_distance [i]=road_width/pixel_number [i], by one group of pixel_ of acquisition
Distance [i] value is stored in document, facilitates subsequent calls.
4. correction factor β is solved:Distance calibration is carried out to track, the gap of distance obtained by relatively more real distance and calculating,
As correction factor, to correct the actual track distance representated by each pixel.I.e. β=two mark between actual range/
Calculate distance between the mark of gained two.
5. registration of vehicle movement locus:Assuming that vehicle moves along a straight line, vehicle movement is obtained using feature point tracking mode
Light stream, the average for calculating all light streams is used for the approximate movement locus for replacing vehicle.
6. vehicle actual motion distance is calculated:If the change in direction is not present during the traveling of vehicle, according to deposit
Pixel_distance [i] and vehicle movement pixel number in document can calculate vehicle actual motion distance, and this is apart from number
It is equal to actual motion distance of the actual range sum in vehicle movement track representated by each pixel as vehicle in value.
I.e.:Distance solution formula replaces integral formula with sum formula;Formula ispixel_distance
[i]。
7. distance correction:Vehicle actual motion distance is multiplied by correction factor β;
8. speed is calculated:The actual motion speed of vehicle is calculated according to the move distance of vehicle in reality and run duration.
I.e.Wherein T is front and rear two frame video image frame differences/video frame rate.
Further, step 6. in, because there is the change in direction in the traveling of vehicle, therefore should be by the actual range and car of gained
Traveling deviation angle be combined, to obtain more accurate result.If the head of the step 5. vehicle movement track of middle record
End point pixel coordinate is respectively (x1, y1), (x2, y2), you can the deviation angle obtained in vehicle travel process isThen the final computing formula of vehicle actual motion distance is:
The present invention uses video image analysis technology, and the simple consumptive material of method is single, it is only necessary to a camera.By photographed data
Car speed monitoring can be carried out by passing back after Surveillance center, Detection results are excellent.It can be obtained after above-mentioned steps using the present invention
To the movement velocity of vehicle.The present invention can reduce the cost of vehicle speed detection and hypervelocity behavior record.In view of camera is general
And rate is higher, so large range of monitoring can be carried out to vehicle, if can alarm or submit car pipe office in the presence of hypervelocity behavior
It is punished, plays a part of supervising driving personnel, can effectively reduce the hypervelocity behavior of vehicle, ensure driving personnel and group
Many securities of the lives and property.
Brief description of the drawings
Fig. 1 is the flow chart of vehicle speed detection step of the present invention.
Embodiment
In order to further explain technical scheme, come to carry out in detail the present invention below by specific embodiment
Explain.
1. track picture is intercepted:The pure road picture in a track is intercepted from video image, as shown in Figure 1.
2. lane width information is obtained:If the coordinate of each pixel is (x, y) in carriageway image, then as y=i, obtain
Take lane width, represented with pixel number, i.e. pixel_number [i].
3. the track actual range representated by per pixel is calculated:If a width of 3.5 meters in fact of track, coordinate points are (x, i)
Some row pixels representated by actual track distance be:
pixel_distance[i]=3.5m/pixel_number[i].By the actual range pixel_ representated by each pixel
Distance [i] is stored in document, the vehicle speed measuring after being easy to.
4. correction factor is solved:Distance calibration is carried out to track, for example, sets a mark every 5m.If step 3. gained
The distance that calculates between two marks of data be z meters, then correction factor β=5/z.
5. registration of vehicle movement locus:The light stream of vehicle movement is obtained using feature point tracking mode, all light streams are calculated
Average be used for the approximate movement locus for replacing vehicle.
6. vehicle actual motion is calculated apart from D:If the first and last point coordinates of vehicle movement track is respectively (m1, n1), (m2,
), n2 obtain vehicle traveling deviation angle be:
The then final computing formula of vehicle actual motion distance
For:
7. distance correction:Vehicle actual motion distance is multiplied by correction factor and obtains vehicle movement distance in reality.
8. speed is calculated:If the frame difference between two width video images is k, the acquisition frame rate of CCTV camera is F, then speed
For:
Using can obtain the movement velocity of vehicle after above-mentioned steps, operation complete once 1.~4. after step, you can
5.~8. step is repeated, speed is measured infinitely.Such as need change track measurement speed, need to first re-operate step 1.~
4., the data to new road are initialized, so as to ensure the accuracy of vehicle speed measurement.
Claims (9)
1. a kind of vehicle speed detection method based on traffic surveillance videos, it is characterised in that comprise the following steps:
1. track is split;
2. lane width information is obtained;
3. the actual range representated by each pixel is calculated;
4. correction factor β is solved, to correct the actual track distance representated by each pixel;
5. vehicle movement track is obtained;
6. vehicle actual motion distance is calculated;
7. distance correction;
8. speed is calculated;
4. described step solves correction factor β, is first to carry out distance calibration to track, actual range is compared thereafter with calculating institute
Distance is obtained, ratio between the two is obtained, as correction factor, to correct the actual track distance representated by each pixel;
Distance i.e. obtained by the actual range/calculating in β=track.
2. the vehicle speed detection method according to claim 1 based on traffic surveillance videos, it is characterised in that in step
1. when track is split, unnecessary background is deleted using track marking line method of identification, track picture is obtained;Original image used during sectional drawing
Element is consistent with video pixel.
3. the vehicle speed detection method according to claim 1 based on traffic surveillance videos, it is characterised in that in step
When 2. entering curb-to-curb width information gathering, its lane width is represented with the pixel number of lane width in video image.
4. the vehicle speed detection method according to claim 1 based on traffic surveillance videos, it is characterised in that in step
When 3. calculating the actual range representated by per pixel, it is assumed that each longitudinally wide distance one with representated by transverse width of pixel
Cause.
5. the vehicle speed detection method according to claim 1 based on traffic surveillance videos, it is characterised in that its feature
It is, when 5. step obtains vehicle movement track, it is assumed that vehicle moves along a straight line.
6. the vehicle speed detection method according to claim 5 based on traffic surveillance videos, it is characterised in that in step
6. calculate vehicle actual motion apart from when, distance solution formula replaces integral formula with sum formula;Use vehicle movement track
In actual range sum representated by each pixel as vehicle actual motion distance.
7. the vehicle speed detection method according to claim 1 based on traffic surveillance videos, it is characterised in that in step
6. calculate vehicle actual motion apart from when, if there is the change in direction in the traveling of vehicle, need consideration vehicle deviation angle,
Actual motion distance to correct vehicle.
8. the vehicle speed detection method according to claim 1 based on traffic surveillance videos, it is characterised in that step is 7.
Distance correction, is that vehicle actual motion distance is multiplied by into correction factor β.
9. the vehicle speed detection method according to claim 1 based on traffic surveillance videos, it is characterised in that step is 8.
Speed is calculated, and is the actual motion speed that vehicle is calculated according to the move distance and run duration of vehicle in reality, i.e.,
Wherein D, which is that step is 7. middle, calculates obtained vehicle movement distance, and T is front and rear two frame video image frame differences/video frame rate.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310503592.4A CN104575003B (en) | 2013-10-23 | 2013-10-23 | A kind of vehicle speed detection method based on traffic surveillance videos |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310503592.4A CN104575003B (en) | 2013-10-23 | 2013-10-23 | A kind of vehicle speed detection method based on traffic surveillance videos |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104575003A CN104575003A (en) | 2015-04-29 |
CN104575003B true CN104575003B (en) | 2017-07-18 |
Family
ID=53090945
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201310503592.4A Expired - Fee Related CN104575003B (en) | 2013-10-23 | 2013-10-23 | A kind of vehicle speed detection method based on traffic surveillance videos |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104575003B (en) |
Families Citing this family (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105654060A (en) * | 2016-01-04 | 2016-06-08 | 中海网络科技股份有限公司 | Method for acquiring vehicle speed from road monitoring video |
CN106327880B (en) * | 2016-09-09 | 2019-01-25 | 成都通甲优博科技有限责任公司 | A kind of speed recognition methods and its system based on monitor video |
CN106254839A (en) * | 2016-09-30 | 2016-12-21 | 湖南中铁五新重工有限公司 | The anti-method and device of slinging of container truck |
KR102631964B1 (en) * | 2016-11-23 | 2024-01-31 | 엘지이노텍 주식회사 | Method, Apparatus, System, Program and Recording Medium for Analyzing Image using Vehicle Driving Information |
CN106991414A (en) * | 2017-05-17 | 2017-07-28 | 司法部司法鉴定科学技术研究所 | A kind of method that state of motion of vehicle is obtained based on video image |
CN107067752A (en) * | 2017-05-17 | 2017-08-18 | 北京联合大学 | Automobile speedestimate system and method based on unmanned plane image |
CN107315095B (en) * | 2017-06-19 | 2019-07-02 | 哈尔滨工业大学 | More vehicle automatic speed-measuring methods with illumination adaptability based on video processing |
CN109584305A (en) * | 2017-09-29 | 2019-04-05 | 宝沃汽车(中国)有限公司 | Panorama system scaling method, device and vehicle |
CN110503740B (en) * | 2018-05-18 | 2021-11-26 | 杭州海康威视数字技术股份有限公司 | Vehicle state determination method and device, computer equipment and system |
CN110809228B (en) * | 2018-07-18 | 2020-10-20 | 北京聚利科技有限公司 | Speed measurement method, device, equipment and computer readable storage medium |
CN109686088B (en) * | 2018-12-29 | 2021-07-30 | 重庆同枥信息技术有限公司 | Traffic video alarm method, equipment and system |
CN112309134B (en) * | 2019-07-29 | 2022-12-16 | 富士通株式会社 | Vehicle speed detection method and device |
CN112991769A (en) * | 2021-02-03 | 2021-06-18 | 中科视语(北京)科技有限公司 | Traffic volume investigation method and device based on video |
CN114333134B (en) * | 2022-03-10 | 2022-05-31 | 深圳灏鹏科技有限公司 | Cabin management method, device, equipment and storage medium |
Family Cites Families (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20060024696A (en) * | 2004-09-14 | 2006-03-17 | 엘지전자 주식회사 | Apparatus and method for car velocity measurement of mobile station |
CN1963884A (en) * | 2006-12-13 | 2007-05-16 | 王海燕 | Method and system of video frequency velometer |
CN101187671B (en) * | 2007-12-27 | 2010-06-02 | 北京中星微电子有限公司 | Method and device for determining automobile driving speed |
CN102622895B (en) * | 2012-03-23 | 2014-04-30 | 长安大学 | Video-based vehicle speed detecting method |
CN103150908B (en) * | 2013-02-05 | 2015-05-27 | 长安大学 | Average vehicle speed detecting method based on video |
-
2013
- 2013-10-23 CN CN201310503592.4A patent/CN104575003B/en not_active Expired - Fee Related
Also Published As
Publication number | Publication date |
---|---|
CN104575003A (en) | 2015-04-29 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104575003B (en) | A kind of vehicle speed detection method based on traffic surveillance videos | |
US7580547B2 (en) | Electronic traffic monitor | |
CN102622886B (en) | Video-based method for detecting violation lane-changing incident of vehicle | |
CN102467821B (en) | Road distance detection method based on video image and apparatus thereof | |
CN103456172B (en) | A kind of traffic parameter measuring method based on video | |
CN110307791B (en) | Vehicle length and speed calculation method based on three-dimensional vehicle boundary frame | |
CN104021676A (en) | Vehicle positioning and speed measuring method based on dynamic video feature of vehicle | |
CN102306284B (en) | Digital reconstruction method of traffic accident scene based on monitoring videos | |
AU2015352463B2 (en) | Double stereoscopic sensor | |
CN105321342A (en) | Intersection vehicle queuing length detection method based on aerial video | |
CN106803350A (en) | A kind of vehicle speed detection method and device based on camera shooting time difference | |
KR20160100788A (en) | Apparatus and method for measuring speed of moving object | |
CN110462682A (en) | Article detection device and vehicle | |
CN103150908A (en) | Average vehicle speed detecting method based on video | |
CN108416798A (en) | A kind of vehicle distances method of estimation based on light stream | |
CN110189363A (en) | A kind of low multi-view video speed-measuring method of the mobile target of airdrome scene | |
CN116503818A (en) | Multi-lane vehicle speed detection method and system | |
CN110415299B (en) | Vehicle position estimation method based on set guideboard under motion constraint | |
JP2011064639A (en) | Distance measuring device and distance measuring method | |
CN104006804A (en) | Method for detecting offset of contact net based on observation benchmark instability compensation | |
CN108122244B (en) | Video speed measuring method and device for video image | |
CN102622582B (en) | Road pedestrian event detection method based on video | |
Tan et al. | Vehicle speed measurement for accident scene investigation | |
CN109345576B (en) | Vehicle running speed identification method and system | |
CN113160299B (en) | Vehicle video speed measurement method based on Kalman filtering and computer readable storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20170718 Termination date: 20171023 |