CN101790083A - Monitoring method of lane space occupancy based on videos - Google Patents
Monitoring method of lane space occupancy based on videos Download PDFInfo
- Publication number
- CN101790083A CN101790083A CN201010105689A CN201010105689A CN101790083A CN 101790083 A CN101790083 A CN 101790083A CN 201010105689 A CN201010105689 A CN 201010105689A CN 201010105689 A CN201010105689 A CN 201010105689A CN 101790083 A CN101790083 A CN 101790083A
- Authority
- CN
- China
- Prior art keywords
- length
- trapezoidal
- lane
- sigma
- centerdot
- 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.)
- Granted
Links
Images
Landscapes
- Traffic Control Systems (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a monitoring method of lane space occupancy based on videos, comprising the following concrete steps: step 1: arranging an effective monitoring field of view, and acquiring basic data; step 2: carrying out nonlinear correction of a lane: calibrating the lane, computing the gradients of the upper bottom and the lower bottom of a calibrated trapezoid, computing the weight of each row in the calibrated trapezoid, carrying out curve fitting on the corrected weight, iterating and optimizing weight corrected parameters, and computing the actual length of the corrected calibrated trapezoid region lane, the fitting actual length obtained after curve fitting and the height of a triangle capable of being expressed by the inclination angles of the upper bottom and the lower bottom of the calibrated trapezoid region; and step 3: computing the space occupancy of the lane to obtain the actual value of the lane space occupancy. The method of the invention can effectively reduce the computing error of the lane occupancy since vehicles are big in a near place and are small in a far place in traffic video monitoring.
Description
Technical field
The invention belongs to technical field of video monitoring, relate to a kind of method for supervising of the lane space occupancy based on video.
Background technology
The traffic flow information statistics is the key link in the intelligent transportation system, and lane occupancy ratio is in the traffic flow information, weighs an important indicator of traffic, and lane occupancy ratio is divided into time occupancy and space occupancy.Time occupancy is meant in the unit interval that vehicle is by the cumulative time of a certain section and the ratio of unit minute.Space occupancy is meant the ratio that at a time records all vehicles occupy on the known road road surface length and road section length.
The flow information that time occupancy is added up goes wrong under some opposite extreme situations usually, for example, when time occupancy is low, may be because vehicle is very few, also may be because the vehicle that causes of traffic congestion can't fast running state.And for an intelligent traffic information processing system, preventing and in time handling extreme case is the core place, and therefore, the calculating of space occupancy has just had very important effect.Yet under video monitoring mode, adopt space occupancy to calculate lane occupancy ratio, a problem also can appear, because car moves the track of same physical length, and the target vehicle of same physical length in monitoring the visual field, the capital is because three dimensions arrives the projection of two dimensional surface, and different length different problem in position appears, promptly closely then long apart from camera lens, far away then short situation.In addition, the different vehicle size of travelling on road is also normally differentiated, therefore, directly does ratio calculation according to the length and the road section length of the target of extracting in the monitoring image simply, can produce very big deviation.
Summary of the invention
The purpose of this invention is to provide a kind of method for supervising of the lane space occupancy based on video, can effectively reduce in the traffic video monitoring, the near big and far smaller error of bringing of calculating to lane occupancy ratio of vehicle.
The technical solution adopted in the present invention is, a kind of method for supervising of the lane space occupancy based on video, and this method is specifically implemented according to following steps:
Step 1: effective supervision visual field is set, obtains basic data
Delimiting the zone, track is effectively to monitor the visual field, this supervision visual field is one trapezoidal, and the length of this visual field is trapezoidal height, and it is capable of p1 distance in the ranks to be p0 on picture, if the image of monitoring image size is M * N, to each the row definition effective range ident value K on the monitoring image
j, j=1,2 ..., M, for:
Ident value K then
j, j=1,2 ..., M, given effective video monitors the visual field;
Step 2: the gamma correction in track
2.1) demarcation in track
According to the road markings line of national regulation the empty ratio of line of lines is set, find out in view of the above be positioned in the picture monitor the middle part, visual field can complete imaging institute wired, empty, according to sequence arrangement from big to small, be made as
And to measure the minimum trapezoidal upper base width that comprises the line sky from picture be a, is positioned at the tr of image
0OK and the width of going to the bottom be b, be positioned at the tr of image
1OK, the upper base a and the b that goes to the bottom constitute trapezoidal be defined as demarcate trapezoidal;
2.2) calculate trapezoidal upper base of this demarcation and the gradient tan θ that goes to the bottom:
Wherein, h can get according to following formula for demarcating trapezoidal height:
Wherein, l
i, i=1,2 ..., n
LBe the length of i bar line on the monitoring image image on the dotted line;
b
i, i=1,2 ..., n
BBe i empty length on the monitoring image image on the dotted line;
2.3) calculate to demarcate trapezoidal in the weight w of every row
k, k=tr wherein
0, tr
0+ 1 ..., tr
1:
2.4) to weight w
i, i=tr
0, tr
0+ 1 ..., tr
1Carry out segmentation and proofread and correct, make adjacent line, empty than the empty ratio of the line that satisfies aforementioned setting, that is:
S is the sky of Lane Mark, the length ratio of line;
2.5) to the weights after proofreading and correct
K=tr
0, tr
0+ 1 ..., tr
1Carry out curve fitting, obtain new weights
K=tr
0, tr
0+ 1 ..., tr
1, that is:
Proportionality coefficient ρ can be calculated by following formula:
Then the correction weights that obtain by curve fit are:
Wherein, k=tr
0, tr
0+ 1 ..., tr
1
2.6) iteration optimization weights corrected parameter w
* k
The physical length L of the demarcation trapezoid area road behind the calculation correction, the match physical length L that obtains after the curve fit
*, and demarcate the leg-of-mutton height H that the angle of slanted floor up and down of trapezoid area can be represented
Calculate L and L
*Deviation e:e=|L
*-L| if deviation e approaches 0, then need not to revise w
* k, k=tr
0, tr
0+ 1 ..., tr
1, otherwise, carry out following judgement:
If L>H then makes a=a-Δ a, Δ a>0, otherwise make a=a-Δ a, step 2.2 is returned in Δ a<0) in, when obtaining minimum deviation e, iteration finishes; Thus, weight w of trapezoidal interior all row have been obtained to demarcate
* k, k=tr
0, tr
0+ 1 ..., tr
1
Step 3: the space occupancy R that calculates the track
3.1) calculate the weight w of the trapezoidal upper section of demarcation in the effective monitoring visual field
* k, k=p
0, p
0+ 1 ..., tr
0-1, and the weight w of demarcating trapezoidal below part
* kK=tr
1+ 1, tr
1+ 2 ..., p
1:
K=p
0, p
0+ 1 ..., tr
0-1, and k=tr
1+ 1, tr
1+ 2 ..., p
1
3.2) calculate the total length in track, the length L in single track
Sr *For
If the effective supervision track in the supervision visual field is N, then the total length in track is:
L
r *=N·L
sr *; (17)
3.3) calculate the road occupying length L of vehicle target
C *
If detect current time M vehicle target arranged, lay respectively at the c in the picture image
K0Walk to c
K1OK, k=1,2 ..., M calculates the road occupying length L of each vehicle target
Ck *For:
Total road occupying length L of vehicle then
C *For:
The beneficial effect of the inventive method is, by setting up the relation of the Nonlinear Mapping between the track length and actual physical size on the monitoring image, these mapping relations are represented with the mode of weights, after detected target length is carried out gamma correction in the video pictures, can obtain the result of calculation of correct lane space occupancy, weights represent that mode has also guaranteed the requirement of real-time simultaneously.
Description of drawings
Fig. 1 is that the used track of the inventive method monitors the visual field schematic diagram.
Embodiment
The present invention is described in detail below in conjunction with the drawings and specific embodiments.
The method for supervising of the lane space occupancy based on video of the present invention is implemented according to following steps,
Step 1: effective supervision visual field is set, obtains basic data
As Fig. 1, be that the track monitors the visual field schematic diagram, in order to reduce the influence of track surrounding environment, the zone of delimiting wherein, track be effective supervision visual field, is one trapezoidal.The length of this visual field is trapezoidal height, sees on picture, is p
0Row is to p
1Distance in the ranks.If the image of monitoring image size is M * N, to each the row definition effective range ident value K on the monitoring image
j, j=1,2 ..., M is:
Ident value K then
j, j=1,2 ..., M, given effective video monitors the visual field.
Step 2: the gamma correction in track
2.1) demarcation in track
According to Fig. 1, according to setting up standard of the road lines of present China, highway, Class I highway and city expressway lines are 6 * 9, and other roads are 2 * 4.Embodiment is example with the city expressway, and the line of lines is empty to be generated than the mode with " line/sky=2/3 ", in view of the above, find out be positioned in the picture monitor the middle part, visual field can complete imaging wired, empty, according to sequence arrangement from big to small, be made as
And to measure the minimum trapezoidal upper base width that comprises the line sky from picture be a, and being positioned at the capable and width of going to the bottom of the tr0 of image is b, and the tr1 that is positioned at image is capable, the upper base a and the b that goes to the bottom constitute trapezoidal be defined as demarcate trapezoidal.
2.2) calculate trapezoidal upper base of this demarcation and the gradient tan θ that goes to the bottom, promptly
Wherein, h can get according to following formula for demarcating trapezoidal height:
Wherein, l
i, i=1,2 ..., n
LBe the length of i bar line on the monitoring image image on the dotted line;
b
i, i=1,2 ..., n
BBe i empty length on the monitoring image image on the dotted line.
2.3) calculate to demarcate trapezoidal in the weight w of every row
k, k=tr wherein
0, tr
0+ 1 ..., tr
1:
2.4) to weight w
i, i=tr
0, tr
0+ 1 ..., tr
1Carry out segmentation and proofread and correct, purpose is in order to make adjacent line, empty than satisfying 1/s, that is:
Wherein, s is the sky of Lane Mark, the length ratio of line, and s is the inverse of the empty ratio of line, according to the road markings line setting of national regulation, and for city expressway s=3/2=1.5, other roads s=4/2=2.
2.5) big because monitor generally trapezoidal than the demarcation that the is provided with zone of the scope of visual field, for this reason, to the weights after proofreading and correct
K=tr
0, tr
0+ 1 ..., tr
1Carry out curve fitting, to obtain new weights
K=tr
0, tr
0+ 1 ..., tr
1, that is:
Wherein, proportionality coefficient ρ can be calculated by following formula:
The correction weights that obtain by curve fit are:
Wherein, k=tr
0, tr
0+ 1 ..., tr
1
2.6) iteration optimization weights corrected parameter w
* k, k=tr wherein
0, tr
0+ 1 ..., tr
1
The physical length L of the demarcation trapezoid area road behind the calculation correction, the match physical length L that obtains after the curve fit
*, and demarcate the leg-of-mutton height H that the angle of slanted floor up and down of trapezoid area can be represented
Calculate L and L
*Deviation e:e=|L
*-L| if deviation e approaches 0, then need not to revise w
* k, k=tr
0, tr
0+ 1 ..., tr
1, otherwise, carry out following judgement: if L>H then makes a=a-Δ a, Δ a>0, otherwise make a=a-Δ a, Δ a<0 turns back to step 2.2) in, when obtaining minimum deviation e, iteration finishes, and thus, has obtained to demarcate weight w of trapezoidal interior all row
* k, k=tr
0, tr
0+ 1 ..., tr
1
Step 3: the space occupancy R that calculates the track
3.1) calculate the trapezoidal part outward of demarcation in the effective monitoring visual field, the weight w of promptly demarcating trapezoidal upper section
* k, k=p
0, p
0+ 1 ..., tr
0-1, and the weight w of demarcating trapezoidal below part
* kK=tr
1+ 1, tr
1+ 2 ..., p
1:
K=p
0, p
0+ 1 ..., tr
0-1, and k=tr
1+ 1, tr
1+ 2 ..., p
1(15)
If the effective supervision track in the supervision visual field is L
*Individual, then the total length in track is:
L
r *=N·L
sr * (17)
3.3) calculate the road occupying length L of all vehicle targets
C *
If detect current time M vehicle target arranged, lay respectively at the c in the picture image
K0Walk to c
K1OK, k=1,2 ..., M calculates the road occupying length L of each vehicle target
Ck *For:
Total road occupying length L of vehicle then
C *For:
Claims (2)
1. method for supervising based on the lane space occupancy of video, it is characterized in that: this method is specifically implemented according to following steps:
Step 1: effective supervision visual field is set, obtains basic data
Delimiting zone, track be effective supervision visual field, and this supervision visual field is one trapezoidal, and the length of this visual field is trapezoidal height, is p on picture
0Row is to p
1Distance in the ranks, the image size of establishing monitoring image is M * N, to each the row definition effective range ident value K on the monitoring image
j, j=1,2 ..., M, for:
Ident value K then
j, j=1,2 ..., M, given effective video monitors the visual field;
Step 2: the gamma correction in track
2.1) demarcation in track
According to the road markings line of national regulation the empty ratio of line of lines is set, find out in view of the above be positioned in the picture monitor the middle part, visual field can complete imaging institute wired, empty, according to sequence arrangement from big to small, be made as
And to measure the minimum trapezoidal upper base width that comprises the line sky from picture be a, is positioned at the tr of image
0OK and the width of going to the bottom be b, be positioned at the tr of image
1OK, the upper base a and the b that goes to the bottom constitute trapezoidal be defined as demarcate trapezoidal;
2.2) calculate trapezoidal upper base of this demarcation and the gradient tan θ that goes to the bottom:
Wherein, l
i, i=1,2 ..., n
LBe the length of i bar line on the monitoring image image on the dotted line;
b
i, i=1,2 ..., n
BBe i empty length on the monitoring image image on the dotted line;
2.3) calculate to demarcate trapezoidal in the weight w of every row
k, k=tr wherein
0, tr
0+ 1 ..., tr
1:
2.4) to weight w
i, i=tr
0, tr
0+ 1 ..., tr
1Carry out segmentation and proofread and correct, make adjacent line, empty than the empty ratio of the line that satisfies aforementioned setting, that is:
S is the sky of Lane Mark, the length ratio of line;
2.5) to the weights after proofreading and correct
K=tr
0, tr
0+ 1 ..., tr
1Carry out curve fitting, obtain new weights
K=tr
0, tr
0+ 1 ..., tr
1, that is:
Proportionality coefficient ρ can be calculated by following formula:
Then the correction weights that obtain by curve fit are:
Wherein, k=tr
0, tr
0+ 1 ..., tr
1
2.6) iteration optimization weights corrected parameter w
* k
The physical length L of the demarcation trapezoid area road behind the calculation correction, the match physical length L that obtains after the curve fit
*, and demarcate the leg-of-mutton height H that the angle of slanted floor up and down of trapezoid area can be represented
Calculate L and L
*Deviation e:e=|L
*-L| if deviation e approaches 0, then need not to revise w
* k, k=tr
0, tr
0+ 1 ..., tr
1, otherwise, carry out following judgement:
If L>H then makes a=a-Δ a, Δ a>0, otherwise make a=a-Δ a, step 2.2 is returned in Δ a<0) in, when obtaining minimum deviation e, iteration finishes; Thus, weight w of trapezoidal interior all row have been obtained to demarcate
* k, k=tr
0, tr
0+ 1 ..., tr
1
Step 3: the space occupancy R that calculates the track
3.1) calculate the weight w of the trapezoidal upper section of demarcation in the effective monitoring visual field
* k, k=p
0, p
0+ 1 ..., tr
0-1, and the weight w of demarcating trapezoidal below part
* kK=tr
1+ 1, tr
1+ 2 ..., p
1:
K=p
0, p
0+ 1 ..., tr
0-1, and k=tr
1+ 1, tr
1+ 2 ..., p
1
3.2) calculate the total length in track, the length L in single track
Sr *For
If the effective supervision track in the supervision visual field is N, then the total length in track is:
L
r *=N·L
sr *; (17)
3.3) calculate the road occupying length L of vehicle target
C *
If detect current time M vehicle target arranged, lay respectively at the c in the picture image
K0Walk to c
K1OK, k=1,2 ..., M calculates the road occupying length L of each vehicle target
Ck *For:
3.4) lane space occupancy R is:
Promptly obtain the actual value of lane space occupancy.
2. preparation method according to claim 1 is characterized in that: the sky of described Lane Mark, the length of line is than s, be provided with according to the road markings line of national regulation, and for city expressway s=3/2=1.5, other roads s=4/2=2.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2010101056896A CN101790083B (en) | 2010-02-04 | 2010-02-04 | Monitoring method of lane space occupancy based on videos |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2010101056896A CN101790083B (en) | 2010-02-04 | 2010-02-04 | Monitoring method of lane space occupancy based on videos |
Publications (2)
Publication Number | Publication Date |
---|---|
CN101790083A true CN101790083A (en) | 2010-07-28 |
CN101790083B CN101790083B (en) | 2012-01-11 |
Family
ID=42533106
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN2010101056896A Expired - Fee Related CN101790083B (en) | 2010-02-04 | 2010-02-04 | Monitoring method of lane space occupancy based on videos |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN101790083B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102867407A (en) * | 2012-09-13 | 2013-01-09 | 东南大学 | Multistep prediction method for effective parking space occupation rate of parking lot |
CN106464843A (en) * | 2014-09-05 | 2017-02-22 | 堺显示器制品株式会社 | Image generation apparatus, image generation method, and computer program |
CN109102548A (en) * | 2018-08-23 | 2018-12-28 | 武汉中观自动化科技有限公司 | It is a kind of for identifying the method and system of following range |
CN113255469A (en) * | 2021-05-06 | 2021-08-13 | 南京大学 | Method and device for measuring road occupancy of traffic monitoring scene |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN2775768Y (en) * | 2005-02-21 | 2006-04-26 | 西安理工大学 | Dynamic traffic signal controller of image identifying traffic flow |
JP4685561B2 (en) * | 2005-09-12 | 2011-05-18 | 株式会社日立国際電気 | Display method of camera system and camera system |
CN100492434C (en) * | 2006-11-30 | 2009-05-27 | 上海交通大学 | Traffic flow state analysis required detection vehicle sampling quantity obtaining method |
-
2010
- 2010-02-04 CN CN2010101056896A patent/CN101790083B/en not_active Expired - Fee Related
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102867407A (en) * | 2012-09-13 | 2013-01-09 | 东南大学 | Multistep prediction method for effective parking space occupation rate of parking lot |
CN102867407B (en) * | 2012-09-13 | 2014-07-09 | 东南大学 | Multistep prediction method for effective parking space occupation rate of parking lot |
CN106464843A (en) * | 2014-09-05 | 2017-02-22 | 堺显示器制品株式会社 | Image generation apparatus, image generation method, and computer program |
CN106464843B (en) * | 2014-09-05 | 2019-05-14 | 堺显示器制品株式会社 | Video generation device, image generating method and computer readable storage medium |
CN109102548A (en) * | 2018-08-23 | 2018-12-28 | 武汉中观自动化科技有限公司 | It is a kind of for identifying the method and system of following range |
CN113255469A (en) * | 2021-05-06 | 2021-08-13 | 南京大学 | Method and device for measuring road occupancy of traffic monitoring scene |
Also Published As
Publication number | Publication date |
---|---|
CN101790083B (en) | 2012-01-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103927754B (en) | A kind of scaling method of vehicle-mounted vidicon | |
JP4631750B2 (en) | Image processing system | |
CN101727671B (en) | Single camera calibration method based on road surface collinear three points and parallel line thereof | |
US6188778B1 (en) | Traffic congestion measuring method and apparatus and image processing method and apparatus | |
CN103325255B (en) | The method of region transportation situation detection is carried out based on photogrammetric technology | |
KR20130004895A (en) | Method for estimating the roll angle in a travelling vehicle | |
WO2021098211A1 (en) | Road condition information monitoring method and device | |
US8964035B2 (en) | Method for calibrating the image of a camera | |
CN109871776B (en) | All-weather lane line deviation early warning method | |
CN105136153B (en) | A kind of lane line exact position harvester and acquisition method | |
CN101790083B (en) | Monitoring method of lane space occupancy based on videos | |
US20200164884A1 (en) | Road damage calculation system, road damage calculation method, and non-transitory recording medium storing road damage control program | |
CN104575003A (en) | Method for detecting vehicle speed based on road monitoring videos | |
CN106019264A (en) | Binocular vision based UAV (Unmanned Aerial Vehicle) danger vehicle distance identifying system and method | |
KR20160100788A (en) | Apparatus and method for measuring speed of moving object | |
CN106327880B (en) | A kind of speed recognition methods and its system based on monitor video | |
CN106257563A (en) | The management in parking lot | |
CN112382085A (en) | System and method suitable for intelligent vehicle traffic scene understanding and beyond visual range perception | |
CN106128126A (en) | Plane cognition technology is utilized to reduce the method and system meeting red light number of times rush hour | |
CN109186624A (en) | A kind of unmanned vehicle traveling right of way planing method based on friendship specification beam | |
CN106448191B (en) | A kind of plane perception crossing traffic signal control method and system that can take into account pedestrian | |
CN104764465B (en) | A kind of measuring method of remote sensing satellite ground resolution | |
CN106600987A (en) | Intersection traffic signal control method and system having multi-dimensional detection function | |
CN110415299B (en) | Vehicle position estimation method based on set guideboard under motion constraint | |
CN207380901U (en) | A kind of mobile laser type vehicle detecting system |
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 | ||
C17 | Cessation of patent right | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20120111 Termination date: 20140204 |