CN105096591A - Intelligent road monitoring system and method - Google Patents
Intelligent road monitoring system and method Download PDFInfo
- Publication number
- CN105096591A CN105096591A CN201410203668.6A CN201410203668A CN105096591A CN 105096591 A CN105096591 A CN 105096591A CN 201410203668 A CN201410203668 A CN 201410203668A CN 105096591 A CN105096591 A CN 105096591A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- binocular
- image
- intelligent road
- morphological feature
- 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
Landscapes
- Image Processing (AREA)
- Image Analysis (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses an intelligent road monitoring system and method. According to the invention, binocular photographing is performed for a vehicle, feature extraction and matching are performed for binocular images obtained by the binocular photographing, and three-dimensional information of the vehicle is calculated according to the matching relationship of image features; and thus, morphological feature parameters can be calculated according to the three-dimensional information of the vehicle and monitoring judgment can be made depending on the morphological feature parameters, that is, the intelligent road monitoring based on binocular stereo vision is achieved. Compared with the intelligent road monitoring depending on a sensor device in the prior art, the intelligent road monitoring based on the binocular stereo vision can be free from constraints of the sensor device, and therefore flexibly expanded; and besides, compared with the intelligent road monitoring depending on single image processing in the prior art, the intelligent road monitoring based on the binocular stereo vision requires no manual calibration of a camera and can easily ensure high accuracy.
Description
Technical field
The present invention relates to monitoring technique, particularly a kind of intelligent road supervisory system and a kind of intelligent road method for supervising.
Background technology
Intelligent road monitoring has vital meaning for traffic dispersion and safety.Such as:
(1), when deliver superelevation, over-wide container vehicle by conventional sense passage or toll collection lanes time, not only can impact the passage rate of passage, even cause and block up, but also likely can cause the destruction to facility in passage;
(2) if not public transit vehicle takies public transportation lane in violation of rules and regulations within the restriction period, the smooth and easy of public transport is affected;
(3), for the town road and the bridge that there is Vehicle length or height limitation, length or the vehicle highly exceeding restriction are prone to accidents and cause damage to road and bridge.
As above visible, in order to realize the monitoring of effective intelligent road, must need to distinguish the morphological feature of vehicle, and the three-dimensional information of vehicle is obviously very important for the differentiation of morphological feature.
In the monitoring of prior art intelligent road, what mainly rely on the acquisition of the three-dimensional information of vehicle is the such as sensing equipment such as radar, infrared sensor, or relies on the process to single image completely.
But when adopting sensing equipment to realize the acquisition of three-dimensional information, the size on each coordinate direction in three-dimensional information needs to use dissimilar sensing equipment respectively.Such as, the width of vehicle and highly usually need to adopt radar range finding, it is long that the length of vehicle then can adopt infrared sensor to survey.Thus, the realization with regard to result in intelligent road monitoring is limited to the type of sensing equipment, thus causes intelligent road monitoring to be difficult to flexible expansion.
And when rely on completely the acquisition of three-dimensional information is realized to the process of single image time, need to utilize the geological information in practical field of view to carry out scene to camera and manually demarcate.Thus, not only to adds additional the artificial staking-out work in the scene of camera and the accuracy of intelligent road monitoring places one's entire reliance upon to the artificial stated accuracy of camera, thus the enforcement difficulty causing intelligent road monitor is large and be difficult to guarantee pin-point accuracy.
Summary of the invention
In view of this, the invention provides a kind of intelligent road supervisory system and a kind of intelligent road method for supervising.
A kind of intelligent road supervisory system provided by the invention, comprising:
Parallel binocular shooting unit, for taking the binocular image arriving the vehicle capturing position;
Image Feature Matching unit, for extracting image characteristic point respectively and mating the image characteristic point extracted in binocular image from the binocular image of vehicle;
Binocular vision computing unit, for calculating the three-dimensional information of vehicle according to the matching relationship of the image characteristic point in binocular image;
Intelligent logical control module, for calculating the morphological feature parameter of vehicle according to the three-dimensional information of vehicle and carrying out monitoring judgement according to the morphological feature parameter calculated.
Alternatively, parallel binocular shooting unit is configured to shooting direction towards the head of vehicle or afterbody.
Alternatively, comprising further and capture location sensing unit, arriving for sensing the vehicle that position is captured in monitoring; And when capturing location sensing unit senses to when having vehicle to arrive monitoring candid photograph position, the present frame of parallel binocular shooting unit is triggered and is chosen for the binocular image of this vehicle.
Alternatively, comprise information reporting delivery unit further, for the result of monitoring judgement and/or the morphological feature parameter of vehicle are sent to local network and/or strange land network.
Alternatively, comprise monitoring intelligent applying unit further, for triggering corresponding intelligent use according to from the court verdict of information reporting delivery unit and/or the morphological feature parameter of vehicle.
Preferably, morphological feature parameter comprise in the length of vehicle one of at least, and intelligent logical control module is realized monitoring judgement by the threshold value judging the morphological feature parameter of vehicle and whether meet setting and is provided the judgement state whether transfinited according to this.
A kind of intelligent road method for supervising provided by the invention, comprising:
The binocular image reaching the vehicle capturing position is got in shooting;
From the binocular image of vehicle, extract image characteristic point respectively and the image characteristic point extracted in binocular image is mated;
Matching relationship according to the image characteristic point in binocular image calculates the three-dimensional information of vehicle;
Calculate the morphological feature parameter of vehicle according to the three-dimensional information of vehicle and carry out monitoring judgement according to the morphological feature parameter calculated.
Alternatively, the shooting direction of binocular image is towards the head of vehicle or afterbody.
Alternatively, comprise further: sensing arrives the vehicle that position is captured in monitoring, when having sensed vehicle arrival monitoring and having captured position, the present frame of parallel binocular shooting has been triggered and has been chosen for the binocular image of this vehicle.
Alternatively, comprise further: the result of monitoring judgement and/or the morphological feature parameter of vehicle are sent to local network and/or strange land network.
Alternatively, comprise further: the morphological feature parameter according to the court verdict transmitted and/or vehicle triggers corresponding intelligent use.
Preferably, morphological feature parameter comprise in the length of vehicle one of at least, and monitoring judgement is whether the morphological feature parameter by judging vehicle meets that the threshold value that sets realizes and the judgement state whether transfinited.
As above visible, the present invention carries out binocular shooting to vehicle, the binocular image obtained is taken to binocular and carries out feature extraction and matching and the three-dimensional information calculating vehicle according to the matching relationship of characteristics of image, thus the morphological feature parameter of vehicle can be calculated according to the three-dimensional information of vehicle and carry out monitoring judgement according to morphological feature parameter, that is, the Intelligent road achieved based on binocular stereo vision is monitored.Compared to the intelligent road monitoring depending on sensing equipment in prior art, the monitoring of the Intelligent road based on binocular stereo vision that the present invention realizes can depart from the restriction of sensing equipment, thus can realize flexible expansion; And compared to the intelligent road monitoring depending on single image process in prior art, the monitoring of the Intelligent road based on binocular stereo vision that the present invention realizes is without the need to manually demarcating camera and easily guaranteeing pin-point accuracy.
Further, the monitoring of the Intelligent road based on binocular stereo vision that the present invention realizes can also support that the long-range of monitor message reports and remote application.
Accompanying drawing explanation
Fig. 1 is the exemplary architecture schematic diagram of the intelligent road supervisory system in the embodiment of the present invention;
Fig. 2 is the image-forming principle schematic diagram of parallel double item stereo vision;
Fig. 3 is a kind of schematic diagram of image coordinate system;
Fig. 4 is the schematic diagram of a kind of imaging geometry model of video camera;
Fig. 5 is a kind of relation schematic diagram between camera coordinate system and path coordinate system;
Fig. 6 is the exemplary flow schematic diagram of the intelligent road method for supervising in the embodiment of the present invention.
Embodiment
For making object of the present invention, technical scheme and advantage clearly understand, to develop simultaneously embodiment referring to accompanying drawing, the present invention is described in more detail.
Refer to Fig. 1, the intelligent road supervisory system in the present embodiment comprises: parallel binocular shooting unit 10, Image Feature Matching unit 20, binocular vision computing unit 30, intelligent logical control module 40, information reporting delivery unit 50, monitoring intelligent applying unit 60.
Parallel binocular shooting unit 10 is for taking the binocular image arriving the vehicle capturing position S.
In the present embodiment, parallel binocular shooting unit 10 comprises optical axis and is parallel to each other and a pair video camera that there is public viewing area.Parallel double object stereo vision imaging principle is shown can see Fig. 2.As shown in Figure 2, each in a pair video camera has a corresponding camera coordinate system (being expressed as x-y-z and x '-y '-z '), and the camera coordinates of one of them video camera is basis coordinates system; Camera coordinate system initial point O and O ' of each video camera is all to should the camera lens photocentre place of video camera; Further, to the wire length B between the projection centre C of video camera and C ', this represents that this is to the baseline distance between video camera; And the moving target M (i.e. vehicle) detected is positioned at the public viewing area of two cameras.
Further preferably, the shooting accessories such as such as flashlamp can also be comprised in each parallel binocular shooting unit 10.
It should be noted that, the parallel binocular shooting of two shown in Fig. 1 unit 10 is all realize binocular shooting along the travel direction of vehicle; And, in shown in Figure 1 two parallel binocular shooting unit 10, one of them is erected at one end of the candid photograph position S of road and is configured to the head of shooting direction towards vehicle, and another is then erected at the other end of the candid photograph position S of road and is configured at the afterbody of opposite side road shooting direction towards vehicle.But the above-mentioned arrangement mode of the parallel binocular of two shown in Fig. 1 shooting unit 10 is only a kind of optimal way of layout architecture and nonessential.
In actual applications, realize the mode of binocular shooting according to the travel direction along vehicle, can only the either end of road set up a parallel binocular shooting unit 10 and be configured to shooting direction towards vehicle at this road driving time head or afterbody.Further, for two-way road, in each travel direction, all can the only parallel binocular shooting unit 10 of erection one be configured to road shooting direction towards the head of vehicle in this travel direction or afterbody, or, also in each travel direction, two parallel binocular shooting unit 10 can be set up according to mode as shown in Figure 1.Certainly, in the travel direction with vehicle be arbitrary inclination any direction on, can realize equally binocular shooting, the present embodiment will not enumerate.
In addition, the intelligent road supervisory system in the present embodiment can also comprise captures location sensing unit, and this candid photograph location sensing unit arrives for sensing the vehicle that position S is captured in monitoring; Correspondingly, when capturing location sensing unit senses to when having vehicle to arrive monitoring candid photograph position S, the present frame that parallel binocular shooting unit 10 photographs is triggered and is chosen for the binocular image of this vehicle.
Image Feature Matching unit 20 for extracting image characteristic point respectively and mating the image characteristic point extracted in binocular image from the binocular image of vehicle.
Wherein, image characteristic point as herein described and invariable rotary more stable compared to pixel grey scale feature, color character, provincial characteristics, textural characteristics, contour feature, edge feature, Corner Feature etc. can gray inversion be overcome.Such as, SIFT (Scale-invariantfeaturetransform, scale invariant feature is changed) unique point, or, computing velocity is higher than SURF (speededuprobustfeatures the accelerates scale invariant feature) unique point of SIFT feature point.
In practical application, the preferred SURF unique point of image characteristic point, correspondingly, when selecting SURF unique point, the leaching process of image characteristic point mainly comprises integral image calculating, quick extra large gloomy matrix detection (wherein specifically comprise extra large gloomy matrix computations, metric space builds and point of interest detects and accurately locates) and point of interest description (comprising the generation of direction distribution and 64 descriptors).
Binocular vision computing unit 30 is for calculating the three-dimensional information of vehicle according to the matching relationship of the image characteristic point in binocular image.
Specifically, binocular vision calculates and comprises following process: respectively for the image coordinate dimension of the every platform video camera in parallel binocular shooting unit 10, respectively for the path coordinate conversion of two video cameras in the projective transformation linearization of the every platform video camera in parallel binocular shooting unit 10 and simultaneous parallel binocular shooting unit 10.Below, respectively these processes are described in detail.
For image coordinate dimension, refer to Fig. 3.As shown in Figure 3, usually all determine to have to think O in image
0the u-o-v rectangular coordinate system of initial point, the position coordinates (u, v) of each pixel in u-o-v rectangular coordinate system represents the column of this pixel in the pel array of image and row respectively.That is, position coordinates (u, v) is coordinate in units of pixel column column position and does not have physics dimension.Therefore, need the image coordinate system setting up physics dimension, this image coordinate system is with the arbitrary set point O in image
1for initial point, represent long measure (such as in units of millimeter) x-axis and y-axis respectively with represent that column position is parallel with v axle with the u axle of line position.
Wherein, for each video camera in parallel binocular shooting unit 10, in the image coordinate system x-o-y of its physics dimension, initial point O
1be defined in the optical axis of this video camera and the point of intersection of the plane of delineation, under normal circumstances, this focus is generally positioned at the center of image, but due to the making reason of video camera, sometimes also can there is a little deviation.
Based on above-mentioned coordinate system definition mode, if the initial point O of image coordinate system x-o-y
1position coordinates corresponding in u-o-v is (u
0, v
0) and set the length coordinate of each pixel in x-axis and y-axis direction and be respectively dx and dy, then in image, the coordinate of any one pixel under two coordinate systems has following relation:
Further, above-mentioned relation can be expressed as by the form of homogeneous coordinates and matrix:
Correspondingly, reverse-power can be expressed as:
For projective transformation linearization, refer to Fig. 4.A kind of imaging geometry model of video camera is shown, wherein O in Fig. 4
epoint is called the photocentre of video camera, X
eaxle and Y
eaxle is parallel with y-axis with the x-axis in the plane of delineation, Z
eaxle is the optical axis of video camera and vertical with the plane of delineation, as the Z of optical axis
ethe intersection point of axle and the plane of delineation is the initial point O of the coordinate system x-o-y in the plane of delineation
1.By the photocentre point O of video camera
ewith X
eaxle, Y
eaxle, Z
ethe rectangular coordinate system of axle composition is called camera coordinate system, O
ewith O
1between distance be the focal distance f of video camera.
When with video camera shooting vehicle, there is perspective projection relation between the actual coordinate of vehicle and the image photographed, can be obtained by similar triangles:
Wherein ,-z
ethe expression degree of depth,
expression scale down,
represent magnification ratio.
Due to perspective projection can only ensure all spatial point on the one_to_one corresponding between picture point in image and projection line, a projection line can same picture point in correspondence image, therefore, in fact projection process can lose the depth information of spatial point, for this reason, the depth information of the perspective projection relation ability measuring vehicle by simultaneous two video cameras is just needed.
Specifically, in order to obtain the depth information of vehicle, need projective transformation linearization.In the present embodiment, in order to by projective transformation linearization, introduced homogeneous coordinate system, correspondingly, volume coordinate is the picpointed coordinate V that the spatial point projection of V is formed
pcan be expressed as:
Wherein, V=[wx
e, wy
e, wz
e, w]
t, w is the volume coordinate V=[x of normalized factor, normalizing
ey
ez
e]
t; P is projection matrix, and s is zoom factor and removes after normalization.
Thus, the Cartesian coordinates of picture point can be obtained:
Correspondingly, its Inverse projection is as follows:
By the Cartesian coordinates of picture point, can calculate the position coordinates of spatial point under camera coordinate system, this position coordinates can be expressed as:
After being eliminated by w in above formula, the position coordinates of spatial point under camera coordinate system can be expressed as:
Known based on above formula, the projection relation after homogeneous coordinate transformation is consistent with the conclusion that similar triangle relation obtains.Wherein, (x, y) is the known location coordinate in the plane of delineation, and z
ethen can regard parameter as, therefore, the above-mentioned expression formula utilizing parallel binocular to take two video cameras in unit corresponding carries out simultaneous solution, can obtain spatial point as the three-dimensional location coordinates (x in a camera coordinate system of basis coordinates system
e, y
e, z
e).After this, just the three-dimensional location coordinates three-dimensional location coordinates in camera coordinate system be converted in path coordinate system is needed.
For the conversion of path coordinate system, refer to Fig. 5.Relation between camera coordinate system and path coordinate system has been shown in Fig. 5, and relation can describe with rotation matrix R and translation vector t, and correspondingly, the camera coordinate system shown in Fig. 5 and the relation between path coordinate system just can be expressed as:
Wherein, R is the orthogonal matrices of 3 × 3, and t is D translation vector, M
1it is the matrix of 4 × 4.In practical application, after parallel binocular shooting unit completes erection, first automatically can detect the lane line of road, to calculate the three-dimensional coordinate of actual road surface under camera coordinate system of present road, thus the rotation matrix R automatically calculated between camera coordinate system and path coordinate system and translation vector t using the actual pavement-height of road as Z
wthe starting point in direction.
By above formula, the three-dimensional coordinate of spatial point in path coordinate system can be obtained, and spatial point described here, the namely location point of each several part of vehicle.Therefore, comprising the information of the three-dimensional coordinate of vehicle each position point in path coordinate system, is exactly the three-dimensional information of previously described vehicle.
Intelligent logical control module 40 is for calculating the morphological feature parameter of vehicle according to the three-dimensional information of vehicle, such as, in the length of vehicle one of at least, and carry out monitoring judgement according to the morphological feature parameter calculated, such as judge whether the morphological feature parameter of vehicle meets the threshold value of setting and provide the judgement state whether transfinited according to this.Wherein, mentioned herein and threshold value can be obtained by artificial calibration measurements or obtain by empirical value or by the survey calculation based on image procossing.
In practical application, comprise the length of vehicle for the morphological feature parameter of vehicle simultaneously, when calculating the morphological feature parameter of vehicle according to the three-dimensional information of vehicle, first can carry out cluster analysis to the unique point of all doubtful vehicle ' s contours, and according to the shadow interference point of the Given information filter vehicle such as the actual pavement-height of road, then to all unique points belonging to vehicle ' s contour at three coordinate direction X
w, Y
w, Z
wabove to sort respectively.
Correspondingly, the length of vehicle can be determined by the unique point of the front end face of vehicle ' s contour and rear end face, calculates the unique point after sequence at Y
wdifference on direction can obtain; In like manner, the width of vehicle can by the unique point after sorting at X
wthe mathematic interpolation in direction obtains; And for the height of vehicle, consider invisible at the bottom of car and there is the situation of isolated point error, get Z
wthe maximal value in direction 10% is averaging, and thinks that this mean value is the height of vehicle.
And, by the length of each vehicle to be monitored, width and highly can form a morphological feature parameter group corresponding to this vehicle.
Information reporting delivery unit 50 is for being sent to local network and/or strange land network by the result (the judgement state such as whether transfinited) of monitoring judgement and/or the morphological feature parameter of vehicle.
Wherein, for the transmission to local network, the server 70 in local network can be sent to by switching equipment 60; And for the transmission to strange land network, then can realize transmitting by fiber optical transceiver 80 and the destination of transmitting can be the monitor supervision platform of strange land network by switching equipment 60.And concrete load mode, can select arbitrarily according to index requests such as user type, system scale, area coverage, signal transmission distance, information capacities.
Monitoring intelligent applying unit 60 can be arranged in the server of local network, also can be positioned at the monitor supervision platform of strange land network, and for triggering corresponding intelligent use in the morphological feature parameter of the server of local network or the monitor supervision platform foundation court verdict of strange land network and/or vehicle.Such as, to confirming that the vehicle transfinited carries out mark and captures; Again such as, in conjunction with the lane situation of the present road collected, reminded in time before vehicle enters specified link driver, can from allowing the special lane of overrun vehicle traffic.
And, by the intelligent use at local network or strange land network, can also retrieve by assisting vehicle, provide great facility for traffic control department and public security department trace accident vehicles peccancy.Such as, if judgement is transfinited, then court verdict can be informed the driver of overrun vehicle by monitor supervision platform, and can provide voice and/or word suggestion further, changes its course as suggestion changes driver; And/or, court verdict can be informed the relevant departments staff such as traffic control, and voice and/or word suggestion can be provided further, as suggestion relevant departments staff closes ring road etc.
In the present embodiment, binocular vision computing unit 30, intelligent logical control module 40, information reporting delivery unit 50 and monitoring intelligent applying unit 60 can exist with the form of the separate hardware such as such as phy chip respectively, or, also the form of software can exist and be carried in the processors such as such as CPU.In addition, information reporting delivery unit 50 and monitoring intelligent applying unit 60 are optional functional unit.
It is more than the detailed description to the intelligent road supervisory system in the present embodiment.Except this intelligent road supervisory system, the present embodiment additionally provides a kind of intelligent road method for supervising.
Refer to Fig. 6, this intelligent road method for supervising comprises:
Step 600, shooting arrives the binocular image of the vehicle capturing position.Wherein, this step preferably adopts the travel direction along vehicle to realize the mode of binocular shooting, and now, the shooting direction of binocular image can towards the head of vehicle or afterbody; Further, this step can sense and arrive the vehicle that position is captured in monitoring, and when having sensed vehicle arrival monitoring and having captured position, the present frame taken by parallel binocular triggers the binocular image being chosen for this vehicle
Step 601, extracts image characteristic point respectively and mates the image characteristic point extracted in binocular image from the binocular image of vehicle.Wherein, this step is extracted and the unique point of mating, and can be SIFT feature point, or computing velocity is higher than the SURF unique point of SIFT feature point.
Step 602, the matching relationship according to the image characteristic point in binocular image calculates the three-dimensional information of vehicle.Wherein, the binocular vision that this step performs calculates and comprises following process: as previously described respectively for the image coordinate dimension of every platform video camera, as previously described respectively for the projective transformation linearization of every platform video camera and the path coordinate conversion of simultaneous two video cameras as previously described.
Step 603, the morphological feature parameter of vehicle is calculated according to the three-dimensional information of vehicle, the length etc. of such as vehicle, and carry out monitoring judgement according to the morphological feature parameter calculated, such as judge whether the morphological feature parameter of vehicle meets the threshold value of setting and provide the judgement state whether transfinited according to this.Wherein, mentioned in this step threshold value can be obtained by artificial calibration measurements or obtain by empirical value or by the survey calculation based on image procossing.
In practical application, this step is when calculating the morphological feature parameter of vehicle according to the three-dimensional information of vehicle, first can carry out cluster analysis to the unique point of all doubtful vehicle ' s contours, and according to the shadow interference point of the Given information filter vehicle such as the actual pavement-height of road, then all unique points belonging to vehicle ' s contour are sorted respectively on three coordinate directions, and according to previously described mode, according to the position difference of the unique point that different coordinate direction sorts or ratio average to determine the length of vehicle.And, by the length of each vehicle to be monitored, width and highly can form a morphological feature parameter group corresponding to this vehicle.
Step 604, is sent to local network and/or strange land network by the result (the judgement state such as whether transfinited) of monitoring judgement and/or the morphological feature parameter of vehicle.Wherein, for the transmission to local network, the server in local network can be sent to by switching equipment; And for the transmission to strange land network, then can realize transmitting by fiber optical transceiver and the destination of transmitting can be the monitor supervision platform of strange land network by switching equipment.And concrete load mode, can select arbitrarily according to index requests such as user type, system scale, area coverage, signal transmission distance, information capacities.
Step 605, the morphological feature parameter according to court verdict and/or vehicle triggers corresponding intelligent use.Such as, to confirming that the vehicle transfinited carries out mark and captures; Again such as, in conjunction with the lane situation of the present road collected, reminded in time before vehicle enters specified link driver, can from allowing the special lane of overrun vehicle traffic.
In practical application, this step can perform at local network or strange land network.And, by the intelligent use at local network or strange land network, can also retrieve by assisting vehicle, provide great facility for traffic control department and public security department trace accident vehicles peccancy.Such as, if judgement is transfinited, then court verdict can be informed the driver of overrun vehicle by monitor supervision platform, and can provide voice and/or word suggestion further, changes its course as suggestion changes driver; And/or, court verdict can be informed the relevant departments staff such as traffic control, and voice and/or word suggestion can be provided further, as suggestion relevant departments staff closes ring road etc.
It should be noted that, above-mentioned steps 604 ~ 605 is optional step.
The foregoing is only preferred embodiment of the present invention, not in order to limit the present invention, within the spirit and principles in the present invention all, any amendment made, equivalent replacement, improvement etc., all should be included within the scope of protection of the invention.
Claims (12)
1. an intelligent road supervisory system, is characterized in that, comprising:
Parallel binocular shooting unit, for taking the binocular image arriving the vehicle capturing position;
Image Feature Matching unit, for extracting image characteristic point respectively and mating the image characteristic point extracted in binocular image from the binocular image of vehicle;
Binocular vision computing unit, for calculating the three-dimensional information of vehicle according to the matching relationship of the image characteristic point in binocular image;
Intelligent logical control module, for calculating the morphological feature parameter of vehicle according to the three-dimensional information of vehicle and carrying out monitoring judgement according to the morphological feature parameter calculated.
2. intelligent road supervisory system according to claim 1, is characterized in that, parallel binocular shooting unit is configured to shooting direction towards the head of vehicle or afterbody.
3. intelligent road supervisory system according to claim 1, is characterized in that, comprises further and captures location sensing unit, arrives for sensing the vehicle that position is captured in monitoring; And when capturing location sensing unit senses to when having vehicle to arrive monitoring candid photograph position, the present frame of parallel binocular shooting unit is triggered and is chosen for the binocular image of this vehicle.
4. intelligent road supervisory system according to claim 1, is characterized in that, comprises information reporting delivery unit further, for the result of monitoring judgement and/or the morphological feature parameter of vehicle are sent to local network and/or strange land network.
5. intelligent road supervisory system according to claim 4, is characterized in that, comprises monitoring intelligent applying unit further, for triggering corresponding intelligent use according to from the court verdict of information reporting delivery unit and/or the morphological feature parameter of vehicle.
6. intelligent road supervisory system according to any one of claim 1 to 5, it is characterized in that, morphological feature parameter comprise in the length of vehicle one of at least, and intelligent logical control module is by judging whether the morphological feature parameter of vehicle meets the threshold value set and realize monitoring judgement and provide the judgement state whether transfinited according to this.
7. an intelligent road method for supervising, is characterized in that, comprising:
Shooting arrives the binocular image of the vehicle capturing position;
From the binocular image of vehicle, extract image characteristic point respectively and the image characteristic point extracted in binocular image is mated;
Matching relationship according to the image characteristic point in binocular image calculates the three-dimensional information of vehicle;
Calculate the morphological feature parameter of vehicle according to the three-dimensional information of vehicle and carry out monitoring judgement according to the morphological feature parameter calculated.
8. intelligent road method for supervising according to claim 7, is characterized in that, the shooting direction of binocular image is towards the head of vehicle or afterbody.
9. intelligent road method for supervising according to claim 7, it is characterized in that, comprise further: sensing arrives the vehicle that position is captured in monitoring, when having sensed vehicle arrival monitoring and having captured position, the present frame of parallel binocular shooting has been triggered and has been chosen for the binocular image of this vehicle.
10. intelligent road method for supervising according to claim 9, is characterized in that, comprise further: the result of monitoring judgement and/or the morphological feature parameter of vehicle are sent to local network and/or strange land network.
11. intelligent road method for supervising according to claim 10, is characterized in that, comprise further: the morphological feature parameter according to the court verdict transmitted and/or vehicle triggers corresponding intelligent use.
12. intelligent road method for supervising according to any one of claim 7 to 11, it is characterized in that, morphological feature parameter comprise in the length of vehicle one of at least, and monitoring judgement is that the whether satisfied threshold value set of morphological feature parameter by judging vehicle realizes and the judgement state whether transfinited.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410203668.6A CN105096591B (en) | 2014-05-14 | Intelligent road monitoring system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410203668.6A CN105096591B (en) | 2014-05-14 | Intelligent road monitoring system and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105096591A true CN105096591A (en) | 2015-11-25 |
| CN105096591B CN105096591B (en) | 2018-06-01 |
Family
ID=
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106781601A (en) * | 2016-11-24 | 2017-05-31 | 上海交通大学 | Vehicle position tracing system and method in cabin based on 3D vision splicing |
| CN107452035A (en) * | 2016-06-01 | 2017-12-08 | 纬创资通股份有限公司 | Method and apparatus for analyzing lane line image and computer readable medium thereof |
| CN108022429A (en) * | 2016-11-04 | 2018-05-11 | 浙江大华技术股份有限公司 | A kind of method and device of vehicle detection |
| CN109523570A (en) * | 2017-09-20 | 2019-03-26 | 杭州海康威视数字技术股份有限公司 | Beginning parameter transform model method and device |
| CN109564682A (en) * | 2016-08-12 | 2019-04-02 | 松下知识产权经营株式会社 | Road surface estimation device, controller of vehicle, road surface estimation method and program |
| CN110853367A (en) * | 2019-12-02 | 2020-02-28 | 江苏纬信工程咨询有限公司 | Intelligent risk prompt indicating system and method for road driving safety |
| US10861177B2 (en) | 2015-11-11 | 2020-12-08 | Zhejiang Dahua Technology Co., Ltd. | Methods and systems for binocular stereo vision |
| CN112489448A (en) * | 2019-09-11 | 2021-03-12 | 浙江宇视科技有限公司 | Snapshot output filtering method and device |
| CN114329074A (en) * | 2022-03-07 | 2022-04-12 | 深圳市中交阳光科技有限公司 | Method and system for detecting traffic energy efficiency of ramp road section |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101140701A (en) * | 2007-10-18 | 2008-03-12 | 张荣楠 | Automatic control overrun vehicle traffic system |
| CN201345166Y (en) * | 2008-12-26 | 2009-11-11 | 北京工业大学 | Video-based passage vehicle detector |
| CN102044155A (en) * | 2009-10-20 | 2011-05-04 | 西安费斯达自动化工程有限公司 | Vehicle configuration dimension image detection and analysis device based on field programmable gate array (FPGA) |
| CN102230798A (en) * | 2011-04-12 | 2011-11-02 | 清华大学 | Portable quick staff-free investigation system of traffic accident scene based on binocular vision |
| CN202454078U (en) * | 2012-02-27 | 2012-09-26 | 上海理工大学 | Road vehicle detecting device |
| CN103164958A (en) * | 2011-12-15 | 2013-06-19 | 无锡中星微电子有限公司 | Method and system for vehicle monitoring |
| CN203365452U (en) * | 2013-05-02 | 2013-12-25 | 中国计量学院 | Vehicle speed detection system based on binocular stereoscopic vision |
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101140701A (en) * | 2007-10-18 | 2008-03-12 | 张荣楠 | Automatic control overrun vehicle traffic system |
| CN201345166Y (en) * | 2008-12-26 | 2009-11-11 | 北京工业大学 | Video-based passage vehicle detector |
| CN102044155A (en) * | 2009-10-20 | 2011-05-04 | 西安费斯达自动化工程有限公司 | Vehicle configuration dimension image detection and analysis device based on field programmable gate array (FPGA) |
| CN102230798A (en) * | 2011-04-12 | 2011-11-02 | 清华大学 | Portable quick staff-free investigation system of traffic accident scene based on binocular vision |
| CN103164958A (en) * | 2011-12-15 | 2013-06-19 | 无锡中星微电子有限公司 | Method and system for vehicle monitoring |
| CN202454078U (en) * | 2012-02-27 | 2012-09-26 | 上海理工大学 | Road vehicle detecting device |
| CN203365452U (en) * | 2013-05-02 | 2013-12-25 | 中国计量学院 | Vehicle speed detection system based on binocular stereoscopic vision |
Non-Patent Citations (1)
| Title |
|---|
| 苏建等: "汽车整车尺寸机器视觉测量系统的研究", 《公路交通科技》 * |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10861177B2 (en) | 2015-11-11 | 2020-12-08 | Zhejiang Dahua Technology Co., Ltd. | Methods and systems for binocular stereo vision |
| CN107452035A (en) * | 2016-06-01 | 2017-12-08 | 纬创资通股份有限公司 | Method and apparatus for analyzing lane line image and computer readable medium thereof |
| CN107452035B (en) * | 2016-06-01 | 2020-01-07 | 纬创资通股份有限公司 | Method and apparatus for analyzing lane line image and computer readable medium thereof |
| CN109564682A (en) * | 2016-08-12 | 2019-04-02 | 松下知识产权经营株式会社 | Road surface estimation device, controller of vehicle, road surface estimation method and program |
| CN108022429A (en) * | 2016-11-04 | 2018-05-11 | 浙江大华技术股份有限公司 | A kind of method and device of vehicle detection |
| CN106781601A (en) * | 2016-11-24 | 2017-05-31 | 上海交通大学 | Vehicle position tracing system and method in cabin based on 3D vision splicing |
| CN109523570A (en) * | 2017-09-20 | 2019-03-26 | 杭州海康威视数字技术股份有限公司 | Beginning parameter transform model method and device |
| CN109523570B (en) * | 2017-09-20 | 2021-01-22 | 杭州海康威视数字技术股份有限公司 | Motion parameter calculation method and device |
| CN112489448A (en) * | 2019-09-11 | 2021-03-12 | 浙江宇视科技有限公司 | Snapshot output filtering method and device |
| CN110853367A (en) * | 2019-12-02 | 2020-02-28 | 江苏纬信工程咨询有限公司 | Intelligent risk prompt indicating system and method for road driving safety |
| CN114329074A (en) * | 2022-03-07 | 2022-04-12 | 深圳市中交阳光科技有限公司 | Method and system for detecting traffic energy efficiency of ramp road section |
| CN114329074B (en) * | 2022-03-07 | 2022-05-10 | 深圳市中交阳光科技有限公司 | Traffic energy efficiency detection method and system for ramp road section |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2021004548A1 (en) | Vehicle speed intelligent measurement method based on binocular stereo vision system | |
| WO2021004312A1 (en) | Intelligent vehicle trajectory measurement method based on binocular stereo vision system | |
| JP3895238B2 (en) | Obstacle detection apparatus and method | |
| US20050270286A1 (en) | Method and apparatus for classifying an object | |
| KR20160062880A (en) | road traffic information management system for g using camera and radar | |
| US20180293450A1 (en) | Object detection apparatus | |
| CN111045000A (en) | Monitoring system and method | |
| EP0631683A1 (en) | An object monitoring system | |
| CN108107462A (en) | The traffic sign bar gesture monitoring device and method that RTK is combined with high speed camera | |
| KR20200064873A (en) | Method for detecting a speed employing difference of distance between an object and a monitoring camera | |
| EP3224820B1 (en) | Double stereoscopic sensor | |
| CN106019264A (en) | Binocular vision based UAV (Unmanned Aerial Vehicle) danger vehicle distance identifying system and method | |
| CN109828267A (en) | The Intelligent Mobile Robot detection of obstacles and distance measuring method of Case-based Reasoning segmentation and depth camera | |
| CN112382085A (en) | System and method suitable for intelligent vehicle traffic scene understanding and beyond visual range perception | |
| EP3364336B1 (en) | A method and apparatus for estimating a range of a moving object | |
| Lu et al. | Automated visual surveying of vehicle heights to help measure the risk of overheight collisions using deep learning and view geometry | |
| Ke et al. | Roadway surveillance video camera calibration using standard shipping container | |
| CN112146620A (en) | Target object ranging method and device | |
| Ho et al. | Localization on freeways using the horizon line signature | |
| Cook et al. | An automated system for persistent real-time truck parking detection and information dissemination | |
| CN109711352A (en) | Vehicle front road environment based on geometry convolutional neural networks has an X-rayed cognitive method | |
| CN116958195A (en) | Object tracking integration method and integration device | |
| KR102418344B1 (en) | Traffic information analysis apparatus and method | |
| CN105096591A (en) | Intelligent road monitoring system and method | |
| CN105096591B (en) | Intelligent road monitoring system and method |
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 | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20190613 Address after: 201203 2nd Floor, No. 77-78, 887 Lane, Zuchong Road, Zhangjiang High-tech Park, Pudong New Area, Shanghai Patentee after: Gaodewei Intelligent Traffic System Co., Ltd., Shanghai Address before: 310053 Haikang Science and Technology Park, 555 Qianmo Road, Binjiang District, Hangzhou City, Zhejiang Province Patentee before: Hangzhou Hikvision Digital Technology Co., Ltd. |