CN106373394A - Vehicle detection method and system based on video and radar - Google Patents
Vehicle detection method and system based on video and radar Download PDFInfo
- Publication number
- CN106373394A CN106373394A CN201610816421.0A CN201610816421A CN106373394A CN 106373394 A CN106373394 A CN 106373394A CN 201610816421 A CN201610816421 A CN 201610816421A CN 106373394 A CN106373394 A CN 106373394A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- video
- radar
- detected
- video image
- 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
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/052—Detecting movement of traffic to be counted or controlled with provision for determining speed or overspeed
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/065—Traffic control systems for road vehicles by counting the vehicles in a section of the road or in a parking area, i.e. comparing incoming count with outgoing count
Abstract
The invention relates to a vehicle detection method and a system based on video and radar, wherein the method comprises the following steps: collecting radar data of vehicles on a road section to be detected; collecting a video image of a road section to be detected; matching a vehicle graph on the video image with radar data corresponding to a vehicle on a road section to be detected based on the radar data and the video image, and calculating a vehicle detection index; and superposing the calculation result of the vehicle detection index on the video image for displaying. According to the invention, by fusing the video and the radar detection, the visualization of the radar data can be realized by overlapping the data on the video image, and the problem that the video detection is greatly influenced by the environment can be solved by using the radar data, so that the detection data is more comprehensive, and the detection accuracy is greatly improved.
Description
Technical field
The present invention relates to technical field of vehicle detection, more particularly, to a kind of vehicle checking method based on video and radar and
System.
Background technology
More and more intensive with urban population, vehicle gets more and more, and alleviating urban traffic blocking becomes each urban development
Urgent problem.Real-time optimization municipal traffic control signal, the travel time reducing people is that to reach this purpose effective
Means, and round-the-clock, real-time, comprehensively and accurately information of vehicles detection means be the inevitable approach of this means.
Wagon detector in the market mainly has the coil checker of contact, geomagnetism detecting device and contactless
Video detector and radar detector.For contact measurement device, maximum defect is to need destruction road surface to be pacified
Dress, detection data are more single.So noncontacting proximity sensor is main flow wagon detector at present and in the future.
The effect of visualization of video-based vehicle detection is good, can with monitor in real time road vehicle, in real time viewing Detection results with
And vehicle running orbit.Additionally, also comparing comprehensively based on the Testing index that video data obtains, including vehicle flowrate, occupation rate, row
The indexs such as team leader's degree, headstock distance, and do not affected by vehicle target quantity.Video-based vehicle detection can also detect motion
With static all vehicles.But, video-based vehicle detection does not adapt to various weather environments, such as dense fog, heavy rain, heavy snow
Omit under weather Deng disliking, verification and measurement ratio is greatly reduced it is impossible to normal use.And video-based vehicle detection nor accurately detection vehicle
Instantaneous velocity.
Radar vehicle detector is not then affected by weather environment and light, can be truly realized round-the-clock vehicle inspection
Survey, and the indexs such as vehicle flowrate, instantaneous radial direction speed, vehicle radial position information can also be detected.But radar vehicle detects
Device has no idea to visualize, and can't see the image information of vehicle.When vehicle congestion, that is, destination number is excessive
When, verification and measurement ratio will decline, and can not detect the too low vehicle of static or speed.
Content of the invention
The technical problem to be solved in the present invention is, for existing single video-based vehicle detection or radar vehicle detection
Device there is a problem of each not enough.A kind of vehicle checking method based on video and radar and system are provided.
In order to solve above-mentioned technical problem, the invention provides a kind of vehicle checking method based on video and radar, bag
Include following steps:
Gather the radar data of section to be detected vehicle；
Gather the video image in section to be detected；
Based on described radar data and video image, by the vehicle graphic on video image and vehicle pair on section to be detected
The radar data answered matches, and calculates vehicle detection index；
The result of calculation of vehicle detection index is superimposed upon on video image and is shown.
According in the vehicle checking method based on video and radar of the present invention, described by the car on video image
The step that figure is matched with the corresponding radar data of vehicle on section to be detected includes: to be detected based on video image identification
Whether there is vehicle graphic in region, be then to obtain vehicle graphic vehicle graphic coordinate on the video images, described vehicle figure
Shape coordinate includes abscissa rx' and vertical coordinate ry', when according to vertical coordinate ry' judge that this vehicle graphic is in described area to be detected
When in the start line delimited in advance in domain, according to the abscissa r of vehicle graphicx' and the region to be detected of measured in advance on each
Lateral separation xa of individual lane line1' to xan' judge this vehicle graphic lane information, if xai’≤|rx’|<xai+1', then sentence
This vehicle graphic disconnected belongs to i-th track, 1≤i < n；Wherein n is the quantity in track on region to be detected；Carried based on radar data
The actual position coordinate of pick-up, described actual position coordinate includes abscissa rxAnd vertical coordinate ry, when according to vertical coordinate rySentence
When this vehicle disconnected is in the start line delimit in advance in section to be detected, according to the abscissa r of vehiclexAnd measured in advance
Lateral separation xa of each lane line on section to be detected1To xanJudge the lane information of this vehicle, if xaj≤|rx|<
xaj+1, then judge that this vehicle belongs to j-th track, 1≤j < n；Wherein n is the quantity in track on section to be detected；As i=j,
The vehicle graphic that will identify that is matched with the radar data of this vehicle.
According in the vehicle checking method based on video and radar of the present invention, described vehicle detection index includes
Environment visibility, vehicle holding time, vehicle location, lane information, car speed, queue length, flow and/or road
Situation.
According in the vehicle checking method based on video and radar of the present invention, described calculating vehicle detection index
Step include following vehicle holding time calculation procedure: based on video image computing environment visibility；Judge environment visibility
Whether higher than default visibility, it is to be whether there is based on video images detection vehicle, and record the holding time；Otherwise it is based on thunder
Reach the Data Detection vehicle holding time.
According in the vehicle checking method based on video and radar of the present invention, described based on video images detection
Vehicle whether there is inclusion: using the lbp operator extraction vehicle textural characteristics of 9 × 9 neighborhoods, after sample image is trained
To the strong classifier with lbp feature, judge to whether there is vehicle in video image using the strong classifier that this carries lbp feature
Whether figure, had the primary testing result of car；When described primary testing result detects and has car, using haar feature extraction
Vehicle textural characteristics, obtain the strong classifier with haar feature after sample image is trained, utilization should be special with haar
The strong classifier levied judges to whether there is vehicle graphic in video image, whether is had the secondary detection result of car.
According in the vehicle checking method based on video and radar of the present invention, described based on video images detection
Judged herein in connection with radar data when vehicle whether there is, comprised the following steps: using the lbp operator extraction car of 9 × 9 neighborhoods
Textural characteristics, obtain the strong classifier with lbp feature after sample image is trained, carry lbp feature using this
Strong classifier judges to whether there is vehicle graphic in video image, whether is had the primary testing result of car；In described primary
When testing result detection has car, using haar feature extraction vehicle textural characteristics, carried after sample image is trained
The strong classifier of haar feature, judges to whether there is vehicle figure in video image using the strong classifier that this carries haar feature
Whether shape, had the secondary detection result of car and confidence level；Judge whether it higher than default value, is to sentence according to confidence level
Disconnected vehicle exists；Otherwise judge that vehicle whether there is based on radar data, be then to judge that vehicle exists, otherwise judge that vehicle is not deposited
?.
According in the vehicle checking method based on video and radar of the present invention, described calculating vehicle detection index
Step include following car speed calculation procedure: extract the automobile's instant velocity of vehicle based on radar data, judge that automobile's instant velocity is
No is then as car speed using described automobile's instant velocity higher than preset vehicle speed, is otherwise based on video image and calculates average speed and makees
For car speed.
According in the vehicle checking method based on video and radar of the present invention, described based on video image calculating
Average speed is the average speed calculating vehicle by below equation:
Wherein, d2 and d1 is respectively the corresponding reality of vehicles identifications horizontal line on n-th frame and the 1st frame video image in video data
The fore-and-aft distance of border road, t is the picture frame interval time of video image, vaUnit be kilometer per hour.
According in the vehicle checking method based on video and radar of the present invention, described calculating vehicle detection index
Step include following queue length calculation procedure: judge the automobile's instant velocity of afterbody vehicle on section to be detected based on radar data
During less than preset vehicle speed, start based on video data and utilize the method for telescopic window to calculate queue length, and judge in radar data
The moment speed of the head vehicle on described section to be detected is more than withdrawal telescopic window when zero, cancels the queueing condition of vehicle.
Present invention also offers a kind of vehicle detecting system based on video and radar, comprising: radar installations, for gathering
The radar data of section to be detected vehicle；Video acquisition device, for gathering the video image in section to be detected；Fusion detection fills
Put, based on described radar data and video image, the vehicle graphic on video image is corresponding with vehicle on section to be detected
Radar data matches, and calculates vehicle detection index；Display device, for being superimposed upon the result of calculation of vehicle detection index
Shown on video image.
Implement the vehicle checking method based on video and radar and the system of the present invention, have the advantages that this
Bright by blending video and detections of radar, both can realize radar data by superposition of data on the video images can
Depending on changing, can solve the problems, such as that Video Detection is affected by environment big using radar data again, make detection data more comprehensively, detection
Accuracy rate also greatly improves.
Brief description
Fig. 1 is the flow chart according to the preferred embodiment of the present invention based on video and the vehicle checking method of radar；
Fig. 2 is the coordinate schematic diagram according to section to be detected on real road of the present invention；
Fig. 3 is the coordinate schematic diagram of the radar data according to present invention collection；
Fig. 4 is the video image in the section to be detected according to present invention collection；
Fig. 5 is the display situation map of Some vehicles Testing index in Fig. 4；
Fig. 6 is the vehicle holding time calculation procedure stream according to the present invention based on video and the vehicle checking method of radar
Cheng Tu；
Fig. 7 is the flow chart that whether there is step first embodiment based on video images detection vehicle according to the present invention；
Fig. 8 a and Fig. 8 b is respectively the example of positive sample image and negative sample image；
Fig. 9 is the schematic diagram of the lbp operator extraction vehicle textural characteristics according to the present invention Ji Yu 9 × 9 neighborhoods；
Figure 10 is the flow chart that whether there is step second embodiment based on video images detection vehicle according to the present invention；
Figure 11 is the car speed calculation procedure flow process according to the present invention based on video and the vehicle checking method of radar
Figure；
Figure 12 a and 12b be respectively according to the present invention choose the 1st frame and n-th frame video image；
Figure 13 is the module frame chart according to the present invention based on video and the first embodiment of the vehicle detecting system of radar；
Figure 14 is the module frame of the middle fusion detection device of the vehicle detecting system according to the present invention based on video and radar
Figure；
Figure 15 is the module frame chart according to the present invention based on video and the second embodiment of the vehicle detecting system of radar.
Specific embodiment
Purpose, technical scheme and advantage for making the embodiment of the present invention are clearer, below in conjunction with the embodiment of the present invention
In accompanying drawing, the technical scheme in the embodiment of the present invention is clearly and completely described it is clear that described embodiment is
A part of embodiment of the present invention, rather than whole embodiments.Based on the embodiment in the present invention, ordinary skill people
The every other embodiment that member is obtained on the premise of not making creative work, broadly falls into the scope of protection of the invention.
Refer to Fig. 1, be the flow process according to the preferred embodiment of the present invention based on video and the vehicle checking method of radar
Figure.As shown in figure 1, this embodiment provide comprised the following steps based on the vehicle checking method of video and radar:
The radar data of vehicle on s101, collection section to be detected.
The present invention installs radar installations in advance on real road, and delimits the square for l for the length along along vehicle heading
Shape region is section to be detected.As shown in Figure 2, start line l1 and terminated line l2, structure between the two delimited on track in the same direction
Become the section to be detected that length is l.Radar installations is arranged on the road edge in section to be detected, and the installation with this radar installations
Position, as zero o, gathers the speed of vehicle and positional information in section to be detected.Refer to Fig. 3, be according to the present invention
The coordinate schematic diagram of the radar data of collection.In Fig. 3, zero o is radar installations infield, and this coordinate is that two dimension is sat
Mark, that is, radar installations and vehicle, in the projection of ground level, are represented using polar mode.In figure x, y represents track width respectively
Degree direction and vehicle heading, vrRepresent the automobile's instant velocity of vehicle, rrRepresent vehicle location.vX,rxRepresent vehicle along track
The speed of width and positional information, vY,ryRepresent vehicle along the speed of vehicle heading and positional information, radar fills
Put and will detect speed and the actual vehicle position of this both direction.And by below equation calculate vehicle automobile's instant velocity and
Vehicle location:
S102, the video image in collection section to be detected.
This step can be realized by the video acquisition device being pre-installed on real road.Refer to Fig. 4, according to
The video image in the section to be detected of present invention collection.The present invention does not limit to the execution sequence of above-mentioned steps s101 and s102
System is it is also possible to first gather video image, then gathers radar data, or both synchronously execute.
S103, it is based on radar data and video image, by vehicle on the vehicle graphic on video image and section to be detected
Corresponding radar data matches, and calculates vehicle detection index.
Please refer to Fig. 2 and Fig. 4, this step detects vehicle graphic, such as c1 ' and c2 ' on the video images, and with
On section to be detected, the radar data of vehicle matches, such as by the automobile's instant velocity of vehicle graphic c1 ' and vehicle c1 and actual car
Position matches, and equally vehicle graphic c2 ' is matched with vehicle c2.Subsequently, based on above-mentioned radar data and video image
Calculate the vehicle detection index of each vehicle, including but not limited to environment visibility, vehicle holding time, vehicle location, car
The indexs such as road information, car speed, queue length, flow and/or road situation.
S104, the result of calculation of vehicle detection index is superimposed upon on video image is shown.
As shown in figure 4, the present invention is by the car speed of vehicle and actual vehicle position display vehicle figure on the video images
Shape region.Specifically, by the car speed of vehicle c1 and actual vehicle position display vehicle graphic c1 ' rectangle frame
In, " p:20,13 " and " v:35 " in such as Fig. 4.Preferably, the result of calculation of Some vehicles Testing index can also be superimposed upon
The upper left corner of video image is shown.This Some vehicles Testing index be particularly shown situation as shown in figure 5, the upper left corner show
The vehicle detection index shown can include but is not limited to environment visibility, and the flow in each track, road situation, car
The index such as holding time and queue length.Wherein road situation can be expressed as congestion, slow and unimpeded three kinds of states.
These vehicle detection indexs are the testing result in the measurement period setting, and can often 50ms update once.
The present invention, by video image and radar data match, can select and calculate effective vehicle detection and refer to
Mark data, and then watched in real time on the video images.
In a preferred embodiment of the invention, vehicle data is mated by the affiliated track of vehicle.Due to for the moment
Carve same track and only exist a car, therefore the present invention is based respectively on radar data and video images detection goes out the affiliated track of vehicle
Number, if number of track-lines is identical, judges the vehicle graphic that this radar data correspondence detects.
In actual mechanical process, the present invention need in advance on the section to be detected of real road delimit start line l1 and
Terminated line l2, sets up real road coordinate system with the intersection point o of terminated line l2 and road edge for initial point.And measure each lane line
Lateral separation xa at start line l11To xan, i.e. each lane line la on real road1To lanIn real road coordinate system
On abscissa numerical value.Subsequently, gather the video image in this section to be detected, and start line l1 on the video images ' draw and whole
Only delimit region to be detected between line l2 '.The intersection point o ' of the terminated line l2 ' with video image and road edge, as initial point, sets up
Video image coordinate system.Measure each lane line on region to be detected in start line l1 simultaneously ' lateral separation xa at place1' extremely
xan', i.e. each lane line la on video image1' to lan' abscissa in video image coordinate system.
In the vehicle checking method based on video and radar, above-mentioned steps s103 by the vehicle on video image with corresponding
The step that matches of radar data specifically include:
First, identify on the video images in region to be detected and whether there is vehicle graphic, be then to obtain vehicle graphic to exist
Vehicle graphic coordinate (r on video imagex', ry’).This vehicle graphic coordinate includes abscissa rx' and vertical coordinate ry’.With
Start line l1 on video image ' on the basis of, vehicle enter start line l1 ' position when start detect the affiliated track of vehicle.Root
Process according to the affiliated lane information of video images detection vehicle graphic is as follows:
When according to vertical coordinate ry' judge that this vehicle graphic is in start line l1 delimit in advance in described region to be detected '
When upper, i.e. ryDuring '=l, according to the abscissa r of vehicle graphicx' and the region to be detected of measured in advance on each lane line
Lateral separation xa1' to xan' judge this vehicle graphic lane information, if xai’≤|rx’|<xai+1', then judge this vehicle figure
Shape belongs to i-th track, 1≤i < n；Wherein n is the quantity in track on region to be detected.As judged vehicle graphic c2 ' in Fig. 4
Abscissa meets xa2’≤rx’≤xa3', therefore, it is determined that vehicle graphic c2 ' is located at the 2nd track.
Extract actual position coordinate (r in real road coordinate system for the vehicle subsequently, based on radar datax, ry), this is real
Border position coordinateses include abscissa rxAnd vertical coordinate ry.On the basis of start line l1 on real road, enter initial in vehicle
Start during line l1 position to detect the affiliated track of vehicle.Detect that according to radar data the process of the affiliated lane information of vehicle is as follows:
When according to vertical coordinate ryWhen judging that this vehicle is in start line l1 delimit in advance in section to be detected, i.e. ry=l
When, according to the abscissa r of vehiclexAnd on the section to be detected of measured in advance each lane line lateral separation la1To lanSentence
The lane information of this vehicle disconnected, if laj≤|rx|<laj+1, then judge that this vehicle belongs to j-th track, 1≤j < n；Wherein n is
The quantity in track on section to be detected.As judged in Fig. 2, the abscissa of vehicle graphic c2 meets xa2≤rx≤xa3, therefore, it is determined that
Vehicle c2 is located at the 2nd track.
Finally, judge whether the number of track-lines based on video images detection is identical with the number of track-lines detecting based on radar data,
As i=j, vehicle graphic and the radar data of this vehicle that will identify that match；Otherwise mismatch.For example, aforementioned vehicle
Figure c2 ' and vehicle c2 is respectively positioned on the 2nd track, therefore can be by the automobile's instant velocity of this vehicle graphic c2 ' and vehicle c2 and actual car
Position matches.
The present invention also detects to the vehicle holding time.Refer to Fig. 6, be based on video and radar according to the present invention
Vehicle checking method vehicle holding time calculation procedure flow chart.As shown in fig. 6, in a preferred embodiment of the present invention
In, aforementioned calculating vehicle detection index specifically includes following vehicle holding time calculation procedure:
S201, flow process start；
Whether s202, detection environment visibility, more than default visibility, are to go to step s208-s212 and start based on video
Image detection vehicle whether there is, and the registration of vehicle holding time, otherwise goes to step s203-s207 and is based on radar data detection car
Whether there is, and the registration of vehicle holding time.This default visibility is preferably 40%-60%, for example, judge environment visibility
Whether it is more than 50%.
S203, detected whether based on radar data vehicle enter section to be detected, be to go to step s204, otherwise turn step
Rapid s213 terminates flow process；Whether the actual vehicle position judgment vehicle that detections of radar can be passed through in this step is located at road to be detected
In section.
S204, registration of vehicle entry time t1；
S205, detect whether that vehicle leaves detection zone based on radar data, be to go to step s206, otherwise go to step
S205 continues detection；
S206, registration of vehicle time departure t2；
S207, calculating vehicle holding time are t2-t1；
S208, whether had based on video images detection vehicle enter region to be detected, be to go to step s209, otherwise turn step
Rapid s213 terminates flow process；
S209, registration of vehicle entry time t3；
S210, whether there is vehicle to leave region to be detected based on video images detection, be to go to step s211, otherwise turn step
Rapid s210 continues detection；
S211, registration of vehicle time departure t4；
S212, calculating vehicle holding time are t4-t3；
S213, flow process terminate.
It is described based on the detailed process that video images detection vehicle whether there is in the present invention below.The present invention adopts
With two grades of vehicle feature recognition algorithms of Machine self-learning, vehicle graphic is detected and identified.Due to road, vehicle with
And the complexity of weather environment, currently a lot of video frequency vehicle recognizer, including background subtraction, Multi Frame Difference point-score, light stream
Method, local detection, feature detection etc. all can be subject to certain restrictions.Present invention employs the local vehicle characteristic method of two-stage, that is,
Carry out the primary detection of extensive style using follow-on lbp feature, then using haar feature, two are carried out to primary testing result
Secondary identification, the result with recognition confidence for the output.
Refer to Fig. 7, be that step first embodiment be whether there is based on video images detection vehicle according to the present invention
Flow chart.As shown in fig. 7, being included based on the step that video images detection vehicle whether there is in above-mentioned steps s208-s212:
S301, flow process start；
S302, using 9 × 9 neighborhoods lbp operator extraction vehicle textural characteristics, carried after sample image is trained
There is the strong classifier of lbp feature, judge to whether there is vehicle figure in video image using the strong classifier that this carries lbp feature
Whether shape, had the primary testing result of car.Go to step s303 when the detection of primary testing result has car to analyze further, no
Then go to step s305 and judge that vehicle does not exist.
The present invention collection video image resolution be 1080p, and detect vehicle graphic minimum pixel require be
100 about, therefore the present invention is extended to the lbp operator of traditional 3 × 3 neighborhoods, using the lbp operator extraction of 9 × 9 neighborhoods
Vehicle textural characteristics.Therefore, carry out the concrete mistake of the primary detection of extensive style in this step s302 using follow-on lbp feature
Journey is as follows:
First, the present invention, by the negative sample image of the positive sample image of Fig. 8 a example and Fig. 8 b example, is normalized, rule
Model is to the size of 64*64.
Extract image texture subsequently, based on positive and negative sample image.The image of the lbp operator extraction based on traditional 3 × 3 neighborhoods
Texture t can be defined using basic operation unit: whole window is by a center window gcWith 8 adjacent 3 × 3 pixels
Subwindow g0,......,g7Composition, is shown below:
T～(g0-gc,g1-gc,….g7-gc)；
9 × 9 pixel basic operation units after extension are then divided into 93 × 3 pixel subwindows by the present invention, as Fig. 9 institute
Show.Ask for the average gray of 9 subwindows respectively, and replace original 3 × 3 subwindows with this meansigma methods, extract image stricture of vagina
Reason t '.Image texture t ' is represented by:
T～(g0’-gc’,g1’-gc’,….g7’-gc’)；
By this simple operation, compression of images can be on the one hand the 1/9 of original image, greatly reduce the meter of subsequent algorithm
Calculation amount, and also comply with the vehicle pixel request of reality.On the other hand 9 × 9 neighborhood extending lbp operators can be transformed to basic 3
× 3lbp operation operator.
Subsequently with center pixel gcFor reference pixel, local binarization process is carried out to rest of pixels in window.Process side
Method is represented by:
Image texture characteristic after binarization operation is represented by:
ts～(s(g0’-gc’),s(g1’-gc’),….s(g7’-gc’))
Now, basic operation unit can be represented using 8 binary numbers, using following formula, pixel is weighted asking
And process, can obtain this basic operation unit lbp eigenvalue is:
This lbp eigenvalue describes edge, texture, characteristic point and the plane domain of basic operation unit regional area.Will
Aforesaid operations method expands to whole sample image region, can get binary sequence description and the lbp characteristic pattern of sample image.
Based on above modified model lbp characterization method, using discrete adaboost machine learning algorithm by a large amount of
The training of the negative sample image of the positive sample image of vehicle and some specific environments, selects most general character and discriminating energy
The lbp feature of power, and based on this characteristic component grader, build corresponding Weak Classifier.Then, level is carried out to weak typing
Connection, one strong classifier carrying lbp feature of composition.Judge the to be detected of collection using the strong classifier that this carries lbp feature
Whether there is vehicle graphic in the video image in section, and whether had the primary testing result of car.
S303, use haar feature extraction vehicle textural characteristics, obtain special with haar after sample image is trained
The strong classifier levied, judges to whether there is vehicle graphic in video image using the strong classifier that this carries haar feature, obtains
Whether there is the secondary detection result of car.Go to step s304 when the detection of primary testing result is with the presence of car and judge vehicle, otherwise turn
Step s305 judges that vehicle does not exist.
The detailed process in this step s303, primary testing result being recognized for using haar feature is as follows:
Using vehicle's contour and the textural characteristics of the positive and negative sample image of haar feature extraction, by real-time adaboost machine
Learning algorithm is trained to the haar feature obtaining, and builds two grades of strong classifiers carrying confidence level.Using 4 kinds during being somebody's turn to do
Type, totally 6 kinds of haar feature composition characteristic templates, vehicle target is carried out with edge and textural characteristics description, including 2 kinds of edge spies
Levy, 2 kinds of linear characters, a kind to corner characteristics and a kind of central feature.Haar characterizing definition is chequered with black and white rectangle segmentation, permissible
Carry out quantitative expression using the gray scale difference in adjacent rectangle region in image, both using white rectangle pixel in image and deduct black
Rectangular pixels and.In order to preferably extract characteristics of image, the present invention aligns sample image and has carried out gray processing and histogram equalization
Change is processed.
Whether there is vehicle graphic in the video image judging collection using the strong classifier obtaining with haar feature,
Whether there is the secondary detection result of car, and using this secondary detection result as final testing result.
S304, judge vehicle exist.
S305, judge that vehicle does not exist.
S306, flow process terminate.
Refer to Figure 10, be that step second embodiment be whether there is based on video images detection vehicle according to the present invention
Flow chart.As shown in Figure 10, this second embodiment is essentially identical with first embodiment, the difference is that only, based on video
Image detection vehicle is judged herein in connection with radar data while whether there is, and specifically includes following steps:
S401, flow process start；
S402, using 9 × 9 neighborhoods lbp operator extraction vehicle textural characteristics, carried after sample image is trained
There is the strong classifier of lbp feature, judge to whether there is vehicle figure in video image using the strong classifier that this carries lbp feature
Whether shape, had the primary testing result of car.Go to step s403 when the detection of primary testing result has car to analyze further, no
Then go to step s407 and judge that vehicle does not exist.
S403, use haar feature extraction vehicle textural characteristics, obtain special with haar after sample image is trained
The strong classifier levied, judges to whether there is vehicle graphic in video image using the strong classifier that this carries haar feature, obtains
Whether there is the secondary detection result of car and confidence level.Go to step s404 when the detection of secondary detection result has car and carry out next step
Judge, otherwise go to step s407 and judge that vehicle does not exist.
The haar feature that the present invention uses, has the advantages that recognition accuracy height, robustness are good, special compared to training lbp
The binaryzation discrete type adaboost levying, it can be trained in the range of whole gray values, so its obtain strong point
Appliances have the predictive value size of testing result, i.e. confidence level.Therefore make use of haar feature detection defeated in this second embodiment
The confidence level parameter going out, is easy to subsequently the verity obtaining result is analyzed and compare.
S404, judged whether according to confidence level higher than default value, be, go to step s406 judge vehicle exist, otherwise turn
Step s405 carries out next step judgement.For example, if default value is 50%.When the confidence level of step s403 output is more than 50%,
Then can confirm that as being real vehicle target；If less than 50% it is likely that being error detection it is possible to not be real car
Target, this when is corresponding with the target of the detections of radar in this track again.
S405, judge that vehicle whether there is based on radar data, be, go to step s406 and judge that vehicle exists, otherwise turn step
Rapid s407 judges that vehicle does not exist.If that is, radar data also detects that the target of this vehicle, can confirm that to be yes
Real target, otherwise it is assumed that be false vehicle target.Therefore, the present invention, after carrying out secondary detection to video image, is detecting
To vehicle graphic and confidence level relatively low in the case of, can be determined whether by radar data, improve vehicle detection standard
Really property.
S406, judge vehicle exist.
S407, judge that vehicle does not exist.
S408, flow process terminate.
Also car speed is detected in the present invention.Refer to Figure 11, be based on video and thunder according to the present invention
The car speed calculation procedure flow chart of the vehicle checking method reaching.As shown in figure 11, in a preferred embodiment of the present invention
In, aforementioned calculating vehicle detection index specifically includes following car speed calculation procedure:
S501, flow process start；
S502, the automobile's instant velocity v based on radar data extraction vehicler, judge automobile's instant velocity vrWhether it is higher than preset vehicle speed,
It is to go to step s503, otherwise go to step s504；
S503, the automobile's instant velocity v that detections of radar is arrivedrAs car speed v；
S504, based on video image calculate average speed vaAs car speed v.
S505, flow process terminate.
So that preset vehicle speed is as 5km/h as a example, work as vrWhen≤5, then car speed v is the average speed v of Video Detectiona；Otherwise
Car speed v value is the automobile's instant velocity v of detections of radarr.Because radar velocity measurement uses Doppler frequency shift principle, measurement
It is the relative velocity between object and radar, if car speed is than relatively low or transfixion, measuring speed accuracy is significantly
Reduce, so radar velocity measurement will be combined by the present invention with video frequency speed-measuring, to improve tachometric survey accuracy.
Preferably, in the vehicle checking method based on video and radar for the present invention, average car is calculated based on video image
Fast vaDetailed process as follows.The average speed v of vehicle of the present inventionaDisplacement that can be according to the positional information of vehicle on image
To calculate.First, detect vehicle graphic on the 1st frame and the video image of n-th frame.And vehicle mark is drawn on this vehicle graphic
Know horizontal line, in such as Figure 12 a and Figure 12 b, depict the 1st frame vehicles identifications horizontal line s2 and n-th frame vehicles identifications horizontal line s1 respectively.
Calculate fore-and-aft distance d2 and d1 along vehicle heading on this corresponding real road of vehicles identifications horizontal line.
Average speed by below equation calculating vehicle:
Wherein, t is the picture frame interval time of video image, vaUnit be kilometer per hour.Preferably, image interframe
Choose 40ms every time t.
The velocity amplitude that the present invention is measured by said method, is the actual average speed value of vehicle, and requires no complexity
Image calibration and coordinate transform.
The present invention also queue length to vehicle detects.In the present invention be using telescopic window method to video image
Detected, to calculate vehicle queue length.The method of telescopic window is method based on morphologic edge extracting to judge car
Presence, its cardinal principle is: virtual region top several rows grey scale pixel value sum is more than a certain threshold value, and telescopic window
Interior whole grey scale pixel value and with previous moment window in all pixels detest angle value and difference less than a certain threshold value when, then telescopic window is stretched
A long pixel unit.Conversely, telescopic window shortens a pixel unit.
If the information of vehicles that rim detection obtains is not exclusively, this brings very big being stranded just to the setting of telescopic window threshold value
Difficulty, has vehicle queue but the phenomenon that cannot extend of telescopic window, causes few detection in practical application.If when vehicle is no longer queued up
When starting to run at a low speed, in the case that is, only wagon flow is without queue, telescopic window can not rapidly be retracted, in such a scenario
Still detect dequeue, cause error detection.Therefore the present invention can accurately detect the feature of moving target using radar data, first
First parking queuing is done with a pre- judgement.
In a preferred embodiment of the invention, to specifically include following vehicle queue long for aforementioned calculating vehicle detection index
Degree calculation procedure: whether preset vehicle speed is less than based on the automobile's instant velocity that radar data judges afterbody vehicle on section to be detected, is
Then starting based on video data utilizes the method for telescopic window to calculate queue length, does not otherwise operate.This preset vehicle speed is preferably
5km/h.I.e. when vehicle by smooth to queue up when, the vehicle later afterbody in region to be detected, car speed can be by fast
Slack-off, the automobile's instant velocity that radar installations detects also can slowly be reduced to less than 5 kilometers/hour, now proceeds by video row
The detection of team's algorithm, using the marginal information of vehicle, gradient difference and colour-difference grade to obtain the accurate information of vehicle queue.
Judge subsequently, based on radar data whether the moment speed of the head vehicle on section to be detected is more than zero, be then
Withdraw telescopic window, cancel the queueing condition of vehicle, otherwise do not operate.I.e. when vehicle is begun to change into by queuing and do not queue up, treating
When the head vehicle of detection zone starts mobile, radar installations can accurately detect, and now withdraws telescopic window, cancels
The queueing condition of vehicle.
Refer to Figure 13, be the module according to the present invention based on video and the first embodiment of the vehicle detecting system of radar
Block diagram.As shown in figure 13, what this embodiment provided is included based on the vehicle detecting system of video and radar: radar installations 10, regards
Frequency harvester 20, fusion detection device 30 and display device 40.
Wherein radar installations 10 is used for gathering the radar data of section to be detected vehicle.The radar installations 10 of the present invention is installed
Road edge in section to be detected.
Video acquisition device 20 is used for gathering the video image in section to be detected.This video acquisition device 20 can be using peace
The photographic head being contained on section to be detected is realized.
Fusion detection device 30 is connected with radar installations 10 and video acquisition device 20, for based on radar data and video
Image, the vehicle graphic on video image is matched with the corresponding radar data of vehicle on section to be detected, and calculates vehicle
Testing index.
Display device 40 is connected with fusion detection device 30, for the result of calculation of vehicle detection index is superimposed upon video
Shown on image, be easy to watch in real time Detection results.Preferably, this display device 40 can using by Ethernet with melt
Close semaphore and the signal lighties realization that detection means 30 connects.
The vehicle detection index that in the present invention, fusion detection device 30 calculates includes but is not limited to environment visibility, vehicle accounts for
There are time, vehicle location, lane information, car speed, queue length, flow and/or road situation.
Refer to Figure 14, be the middle fusion detection device 30 of the vehicle detecting system based on video and radar according to the present invention
Module frame chart.As shown in figure 14, fusion detection device 30 one or more unit including but not limited to following: vehicle exists
Matching unit 31, vehicle holding time computing unit 32, car speed computing unit 33 and queue length computing unit 34.
Wherein there is matching unit 31 and be used for the vehicle graphic on video image and vehicle pair on section to be detected in vehicle
The radar data answered matches.Specifically, this vehicle has matching unit 31 based in video image identification region to be detected is
No have vehicle graphic, is then to obtain vehicle graphic vehicle graphic coordinate on the video images, described vehicle graphic coordinate bag
Include abscissa rx' and vertical coordinate ry', when according to vertical coordinate ry' judge that this vehicle graphic is in described region to be detected in advance
When in the start line delimited, according to the abscissa r of vehicle graphicx' and the region to be detected of measured in advance on each lane line
Lateral separation xa1' to xan' judge this vehicle graphic lane information, if xai’≤|rx’|<xai+1', then judge this vehicle
Figure belongs to i-th track, 1≤i < n；Wherein n is the quantity in track on region to be detected.Extract subsequently, based on radar data
The actual position coordinate of vehicle, described actual position coordinate includes abscissa rxAnd vertical coordinate ry, when according to vertical coordinate ryJudge
When this vehicle is in the start line delimit in advance in section to be detected, according to the abscissa r of vehiclexAnd the treating of measured in advance
Lateral separation xa of each lane line on detection section1To xanJudge the lane information of this vehicle, if xaj≤|rx|<xaj+1,
Then judge that this vehicle belongs to j-th track, 1≤j < n；Wherein n is the quantity in track on section to be detected.As i=j, will know
The vehicle graphic not gone out is matched with the radar data of this vehicle.
Vehicle holding time computing unit 32 is used for calculating the vehicle holding time.First, based on video image computing environment
Visibility.Judge that environment visibility, whether higher than default visibility, is to whether there is based on video images detection vehicle again, and
The record holding time；Otherwise it is based on radar data and detect the vehicle holding time.
Preferably, whether vehicle holding time computing unit 32 is deposited based on video images detection vehicle by following steps
: 1) use the lbp operator extraction vehicle textural characteristics of 9 × 9 neighborhoods, obtain special with lbp after sample image is trained
The strong classifier levied, judges to whether there is vehicle graphic in video image using the strong classifier that this carries lbp feature, is
The no primary testing result having car.2) when described primary testing result detects and has car, using haar feature extraction vehicle texture
Feature, obtains the strong classifier with haar feature after sample image is trained, carry the strong of haar feature using this and divide
Class device judges to whether there is vehicle graphic in video image, whether is had the secondary detection result of car.
It is highly preferred that whether vehicle holding time computing unit 32 is deposited based on video images detection vehicle by following steps
: 1) use the lbp operator extraction vehicle textural characteristics of 9 × 9 neighborhoods, obtain special with lbp after sample image is trained
The strong classifier levied, judges to whether there is vehicle graphic in video image using the strong classifier that this carries lbp feature, is
The no primary testing result having car.2) when the detection of primary testing result has car, using haar feature extraction vehicle textural characteristics,
Obtain the strong classifier with haar feature after sample image is trained, sentenced using the strong classifier that this carries haar feature
Whether there is vehicle graphic in disconnected video image, whether had the secondary detection result of car and confidence level.3) according to confidence
Degree judges whether higher than default value, is then to judge that vehicle exists；Otherwise judge that vehicle whether there is based on radar data, be then
Judge that vehicle exists, otherwise judge that vehicle does not exist.
Car speed computing unit 33 is used for calculating car speed.First, the instantaneous car of vehicle is extracted based on radar data
Speed, judges whether automobile's instant velocity is higher than preset vehicle speed, is then as car speed using described automobile's instant velocity, is otherwise based on video figure
As calculating average speed as car speed.
Preferably, this car speed computing unit 33 is by the average speed of below equation calculating vehicle:
Wherein, d2 and d1 is respectively the corresponding reality of vehicles identifications horizontal line on n-th frame and the 1st frame video image in video data
The fore-and-aft distance of border road, t is the picture frame interval time of video image, vaUnit be kilometer per hour.
Queue length computing unit 34 is used for calculating queue length.First, judged on section to be detected based on radar data
When the automobile's instant velocity of afterbody vehicle is less than preset vehicle speed, starts based on video data and utilize the method for telescopic window to calculate length of queuing up
Degree, and withdraw telescopic window when the moment speed judging the head vehicle on described section to be detected in radar data is more than zero, take
Disappear the queueing condition of vehicle.
Refer to Figure 15, be the module according to the present invention based on video and the second embodiment of the vehicle detecting system of radar
Block diagram.As shown in figure 15, the vehicle detecting system based on video and radar that second embodiment provides is basic with first embodiment
Identical, difference is, is additionally arranged radar process plate 50 and Video processing plate 60.
Wherein, radar process plate 50 and radar installations 10 communicate, and obtain radar data and are processed.Radar process plate 50
Can be whether there is based on radar data detection vehicle, the vehicle holding time, vehicle location, the real-time indicators such as automobile's instant velocity.
Video processing plate 60 is communicated with video acquisition device 20, obtains video data and is processed.Video processing plate 60
Can be whether there is based on video data detection environment visibility, vehicle, the vehicle holding time, the real-time indicators such as queue length.
Fusion detection device 30 is connected with this radar process plate 50 and Video processing plate 60, and the main fusion radar that passes through is processed
Plate 50 and the real-time detector data of Video processing plate 60 are corresponding with vehicle on section to be detected by the vehicle graphic on video image
Radar data match, and draw the value of calculation of effective vehicle detection index.Wherein radar process plate 50 can by with
Too net and fusion detection device 30 communicate.Video processing plate 60 can adopt but be not limited to rs232 interface and fusion detection device
30 communications.Vehicle be whether there is, the vehicle holding time, lane information, car speed, the car such as congestion in road situation and flow
Testing index, this fusion detection device 30 can be according to the computational methods of each vehicle detection index aforementioned, at radar
Effective data is chosen or calculated to reason plate 50 and the result of calculation that draws of Video processing plate 60, or in the statistics week setting
Related data is counted in phase.The vehicle detection index that just can calculate for only radar data or video data, for example
The vehicle location being calculated based on radar data, and the environment visibility that calculated based on video data and queue length, then
Directly the result of calculation of radar process plate 50 or Video processing plate 60 can be exported.
It should be appreciated that, the vehicle detecting system based on video and radar of the present invention and the principle of method and realized
Cheng Xiangtong, therefore to the embodiment of the vehicle checking method based on video and radar elaborate be also applied for based on video and
The vehicle detecting system of radar.
In sum, the present invention adopts video and two kinds of non-contact type vehicle detecting sensors of radar, on hardware, software
Completely integrated, constitute a kind of real round-the-clock, visualization, comprehensive Testing index, high detection rate, high reliability
New vehicle detecting system.There is provided most basic detection data for solving Urban Traffic Jam Based.This vehicle detecting system exists
The respective advantage of video and radar sensor is combined on hardware device, solving radar monitoring data can not visual ask
Topic, also compensate for Video Detection affected by environment it is impossible to the problem of all weather operations, two kinds of biographies have also been merged on software algorithm
The advantage of sensor is so that the accuracy rate of vehicle detection also greatly improves.
Finally it is noted that above example, only in order to technical scheme to be described, is not intended to limit；Although
With reference to the foregoing embodiments the present invention is described in detail, it will be understood by those within the art that: it still may be used
To modify to the technical scheme described in foregoing embodiments, or equivalent is carried out to wherein some technical characteristics；
And these modification or replace, do not make appropriate technical solution essence depart from various embodiments of the present invention technical scheme spirit and
Scope.
Claims (10)
1. a kind of vehicle checking method based on video and radar is it is characterised in that comprise the following steps:
Gather the radar data of section to be detected vehicle；
Gather the video image in section to be detected；
Based on described radar data and video image, the vehicle graphic on video image is corresponding with vehicle on section to be detected
Radar data matches, and calculates vehicle detection index；
The result of calculation of vehicle detection index is superimposed upon on video image and is shown.
2. the vehicle checking method based on video and radar according to claim 1 it is characterised in that described by video figure
As the step that upper vehicle graphic is matched with the corresponding radar data of vehicle on section to be detected includes:
Whether there is vehicle graphic based in video image identification region to be detected, be then to obtain vehicle graphic on the video images
Vehicle graphic coordinate, described vehicle graphic coordinate includes abscissa rx' and vertical coordinate ry', when according to vertical coordinate ry' judge
When this vehicle graphic is in the start line delimit in advance in described region to be detected, according to the abscissa r of vehicle graphicx' with
And on the region to be detected of measured in advance each lane line lateral separation xa1' to xan' judge this vehicle graphic lane information,
If xai’≤|rx’|<xai+1', then judge that this vehicle graphic belongs to i-th track, 1≤i < n；Wherein n is on region to be detected
The quantity in track；
Extract the actual position coordinate of vehicle based on radar data, described actual position coordinate includes abscissa rxAnd vertical coordinate
ry, when according to vertical coordinate ryWhen judging that this vehicle is in the start line delimit in advance in section to be detected, according to the horizontal seat of vehicle
Mark rxAnd on the section to be detected of measured in advance each lane line lateral separation xa1To xanJudge the track letter of this vehicle
Breath, if xaj≤|rx|<xaj+1, then judge that this vehicle belongs to j-th track, 1≤j < n；Wherein n is track on section to be detected
Quantity；
As i=j, vehicle graphic and the radar data of this vehicle that will identify that match.
3. the vehicle checking method based on video and radar according to claim 1 and 2 is it is characterised in that described vehicle
Testing index includes environment visibility, vehicle holding time, vehicle location, lane information, car speed, queue length, flow
And/or road situation.
4. the vehicle checking method based on video and radar according to claim 3 is it is characterised in that described calculating vehicle
The step of Testing index includes following vehicle holding time calculation procedure:
Based on video image computing environment visibility；
Judge that environment visibility, whether higher than default visibility, is to whether there is based on video images detection vehicle, and record
Holding time；Otherwise it is based on radar data and detect the vehicle holding time.
5. the vehicle checking method based on video and radar according to claim 4 it is characterised in that described based on video
The step that image detection vehicle whether there is includes:
Using the lbp operator extraction vehicle textural characteristics of 9 × 9 neighborhoods, obtain after sample image is trained with lbp feature
Strong classifier, judged in video image with the presence or absence of vehicle graphic using strong classifier that this carries lbp feature, whether obtain
There is the primary testing result of car；
When described primary testing result detects and has car, using haar feature extraction vehicle textural characteristics, sample image is carried out
Obtain the strong classifier with haar feature after training, using this carry haar feature strong classifier judge in video image be
The no secondary detection result that there is vehicle graphic, whether had car.
6. the vehicle checking method based on video and radar according to claim 4 it is characterised in that described based on video
Image detection vehicle is judged herein in connection with radar data when whether there is, and comprises the following steps:
Using the lbp operator extraction vehicle textural characteristics of 9 × 9 neighborhoods, obtain after sample image is trained with lbp feature
Strong classifier, judged in video image with the presence or absence of vehicle graphic using strong classifier that this carries lbp feature, whether obtain
There is the primary testing result of car；
When described primary testing result detects and has car, using haar feature extraction vehicle textural characteristics, sample image is carried out
Obtain the strong classifier with haar feature after training, using this carry haar feature strong classifier judge in video image be
The no secondary detection result that there is vehicle graphic, whether had car and confidence level；
Judged whether higher than default value according to confidence level, be then to judge that vehicle exists；Otherwise vehicle is judged based on radar data
Whether there is, be then to judge that vehicle exists, otherwise judge that vehicle does not exist.
7. the vehicle checking method based on video and radar according to claim 3 is it is characterised in that described calculating vehicle
The step of Testing index includes following car speed calculation procedure:
Extract the automobile's instant velocity of vehicle based on radar data, judge whether automobile's instant velocity is higher than preset vehicle speed, be then by described wink
When speed as car speed, be otherwise based on video image and calculate average speed as car speed.
8. the vehicle checking method based on video and radar according to claim 7 it is characterised in that described based on video
It is the average speed calculating vehicle by below equation that image calculates average speed:
Wherein, d2 and d1 is respectively the corresponding actual road of vehicles identifications horizontal line on n-th frame and the 1st frame video image in video data
The fore-and-aft distance on road, t is the picture frame interval time of video image, vaUnit be kilometer per hour.
9. the vehicle checking method based on video and radar according to claim 3 is it is characterised in that described calculating vehicle
The step of Testing index includes following queue length calculation procedure:
When preset vehicle speed is less than based on the automobile's instant velocity that radar data judges afterbody vehicle on section to be detected, start based on video
The method of data separate telescopic window calculates queue length, and judges the head vehicle on described section to be detected in radar data
Moment speed is more than withdrawal telescopic window when zero, cancels the queueing condition of vehicle.
10. a kind of vehicle detecting system based on video and radar is it is characterised in that include:
Radar installations, for gathering the radar data of section to be detected vehicle；
Video acquisition device, for gathering the video image in section to be detected；
Fusion detection device, based on described radar data and video image, by the vehicle graphic on video image and road to be detected
In section, the corresponding radar data of vehicle matches, and calculates vehicle detection index；
Display device, is shown for being superimposed upon the result of calculation of vehicle detection index on video image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610816421.0A CN106373394B (en) | 2016-09-12 | 2016-09-12 | Vehicle detection method and system based on video and radar |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610816421.0A CN106373394B (en) | 2016-09-12 | 2016-09-12 | Vehicle detection method and system based on video and radar |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106373394A true CN106373394A (en) | 2017-02-01 |
| CN106373394B CN106373394B (en) | 2019-01-04 |
Family
ID=57899336
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610816421.0A Active CN106373394B (en) | 2016-09-12 | 2016-09-12 | Vehicle detection method and system based on video and radar |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106373394B (en) |
Cited By (37)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106710240A (en) * | 2017-03-02 | 2017-05-24 | 公安部交通管理科学研究所 | Passing vehicle tracking and speed measuring method integrating multiple-target radar and video information |
| CN108200708A (en) * | 2018-02-07 | 2018-06-22 | 宁夏乐智博科技有限公司 | Control method and intelligent road-lamp |
| CN108337786A (en) * | 2018-02-07 | 2018-07-27 | 宁夏乐智博科技有限公司 | Illumination control method and intelligent road-lamp |
| CN108337785A (en) * | 2018-02-07 | 2018-07-27 | 宁夏乐智博科技有限公司 | Illumination control method and intelligent illuminating system |
| CN108734965A (en) * | 2017-04-14 | 2018-11-02 | 北京嘀嘀无限科技发展有限公司 | A kind of method and device of vehicle peccancy detection |
| CN108734959A (en) * | 2018-04-28 | 2018-11-02 | 扬州远铭光电有限公司 | A kind of embedded vision train flow analysis method and system |
| CN108749708A (en) * | 2018-05-11 | 2018-11-06 | 长安大学 | The device and method of vehicle fog lamp is automatically turned under a kind of low visibility |
| CN108896062A (en) * | 2018-05-11 | 2018-11-27 | 长安大学 | A kind of device and method using the electronic map group's of prompting mist section in real time |
| CN109085570A (en) * | 2018-06-10 | 2018-12-25 | 南京理工大学 | Automobile detecting following algorithm based on data fusion |
| CN109102702A (en) * | 2018-08-24 | 2018-12-28 | 南京理工大学 | Vehicle speed measuring method based on video encoder server and Radar Signal Fusion |
| CN109255857A (en) * | 2018-08-23 | 2019-01-22 | 江苏罗思韦尔电气有限公司 | A kind of recorder with safety precaution function |
| CN109613537A (en) * | 2019-01-16 | 2019-04-12 | 南京奥杰智能科技有限公司 | A kind of hologram radar |
| CN109658715A (en) * | 2019-01-31 | 2019-04-19 | 厦门精益远达智能科技有限公司 | Statistical method of traffic flow, device, equipment and the storage medium of multilane |
| CN109646796A (en) * | 2019-01-17 | 2019-04-19 | 浙江大学 | Channel wireless radio multi closed loop stimulation system for epilepsy therapy |
| CN110288832A (en) * | 2019-07-10 | 2019-09-27 | 南京慧尔视智能科技有限公司 | It is merged based on microwave with the multiple-object information of video and visual presentation method |
| CN110428647A (en) * | 2019-07-26 | 2019-11-08 | 百度在线网络技术（北京）有限公司 | The vehicle collaboration at crossing current method, apparatus, equipment and storage medium |
| CN110444026A (en) * | 2019-08-06 | 2019-11-12 | 北京万集科技股份有限公司 | The triggering grasp shoot method and system of vehicle |
| CN110515073A (en) * | 2019-08-19 | 2019-11-29 | 南京慧尔视智能科技有限公司 | The trans-regional networking multiple target tracking recognition methods of more radars and device |
| CN110533921A (en) * | 2019-08-06 | 2019-12-03 | 北京万集科技股份有限公司 | The triggering grasp shoot method and system of vehicle |
| CN110596696A (en) * | 2018-06-13 | 2019-12-20 | 通用汽车环球科技运作有限责任公司 | Apparatus and method for improved radar beamforming |
| CN110647806A (en) * | 2019-08-13 | 2020-01-03 | 浙江大华技术股份有限公司 | Object behavior monitoring method, device, equipment, system and storage medium |
| CN110728839A (en) * | 2019-10-23 | 2020-01-24 | 昆山宝创新能源科技有限公司 | Vehicle and illegal punishment method and device thereof |
| CN110827542A (en) * | 2019-11-11 | 2020-02-21 | 江苏中路工程技术研究院有限公司 | Highway safety vehicle distance early warning system |
| CN111126444A (en) * | 2019-11-28 | 2020-05-08 | 天津津航技术物理研究所 | Classifier integration method |
| CN111178215A (en) * | 2019-12-23 | 2020-05-19 | 深圳成谷科技有限公司 | Sensor data fusion processing method and device |
| CN111526646A (en) * | 2020-05-06 | 2020-08-11 | 辽宁百思特达半导体科技有限公司 | Highway service compartment external lighting system based on wisdom lamp pole |
| CN111554111A (en) * | 2020-04-21 | 2020-08-18 | 河北万方中天科技有限公司 | Signal timing optimization method and device based on multi-source data fusion and terminal |
| CN112017448A (en) * | 2019-05-29 | 2020-12-01 | 浙江宇视科技有限公司 | Traffic detection method, signal lamp control method, device, equipment and storage medium |
| WO2020253227A1 (en) * | 2019-06-17 | 2020-12-24 | 创新先进技术有限公司 | Video processing method, apparatus and device, and video monitoring system |
| CN112150799A (en) * | 2020-08-19 | 2020-12-29 | 上海图丽信息技术有限公司 | Method for collecting road vehicle traffic big data by fusing radar videos |
| CN112419712A (en) * | 2020-11-04 | 2021-02-26 | 同盾控股有限公司 | Road section vehicle speed detection method and system |
| US10942267B2 (en) | 2019-06-17 | 2021-03-09 | Advanced New Technologies Co., Ltd. | Video object processing |
| CN112991774A (en) * | 2021-02-02 | 2021-06-18 | 深圳市天鼎微波科技有限公司 | Traffic flow detection method and device based on millimeter waves |
| CN113255708A (en) * | 2020-02-10 | 2021-08-13 | 富士通株式会社 | Data fusion method and device and data processing equipment |
| CN113438449A (en) * | 2021-06-07 | 2021-09-24 | 西安恒盛安信智能技术有限公司 | Video image transmission method |
| CN115331458A (en) * | 2021-05-11 | 2022-11-11 | 北京万集科技股份有限公司 | Signal lamp control method, device, equipment and storage medium |
| CN116740944A (en) * | 2023-08-10 | 2023-09-12 | 云南省交通投资建设集团有限公司 | Driving safety early warning method, device and storage medium in road tunnel |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000348284A (en) * | 1999-06-07 | 2000-12-15 | Mitsubishi Electric Corp | Device and method for vehicle detection and computer- readable recording medium where vehicle detecting program is recorded |
| CN102447911A (en) * | 2010-10-01 | 2012-05-09 | 魏载荣 | Image acquisition unit, acquisition method, and associated control unit |
| JP2013168178A (en) * | 2013-05-15 | 2013-08-29 | Mitsubishi Electric Corp | Vehicle detection system and vehicle detection method for vehicle detection system |
| CN103581627A (en) * | 2013-11-07 | 2014-02-12 | 北京环境特性研究所 | Image and information fusion display method for high-definition video |
| CN104424804A (en) * | 2013-09-10 | 2015-03-18 | 上海弘视通信技术有限公司 | Intelligent speed-measuring method with single radar and multiple lanes and system thereof |
| CN105480227A (en) * | 2015-12-29 | 2016-04-13 | 大连楼兰科技股份有限公司 | Information fusion method based on infrared radar and video image in active driving technique |
-
2016
- 2016-09-12 CN CN201610816421.0A patent/CN106373394B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000348284A (en) * | 1999-06-07 | 2000-12-15 | Mitsubishi Electric Corp | Device and method for vehicle detection and computer- readable recording medium where vehicle detecting program is recorded |
| CN102447911A (en) * | 2010-10-01 | 2012-05-09 | 魏载荣 | Image acquisition unit, acquisition method, and associated control unit |
| JP2013168178A (en) * | 2013-05-15 | 2013-08-29 | Mitsubishi Electric Corp | Vehicle detection system and vehicle detection method for vehicle detection system |
| CN104424804A (en) * | 2013-09-10 | 2015-03-18 | 上海弘视通信技术有限公司 | Intelligent speed-measuring method with single radar and multiple lanes and system thereof |
| CN103581627A (en) * | 2013-11-07 | 2014-02-12 | 北京环境特性研究所 | Image and information fusion display method for high-definition video |
| CN105480227A (en) * | 2015-12-29 | 2016-04-13 | 大连楼兰科技股份有限公司 | Information fusion method based on infrared radar and video image in active driving technique |
Cited By (46)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106710240B (en) * | 2017-03-02 | 2019-09-27 | 公安部交通管理科学研究所 | The passing vehicle for merging multiple target radar and video information tracks speed-measuring method |
| CN106710240A (en) * | 2017-03-02 | 2017-05-24 | 公安部交通管理科学研究所 | Passing vehicle tracking and speed measuring method integrating multiple-target radar and video information |
| CN108734965A (en) * | 2017-04-14 | 2018-11-02 | 北京嘀嘀无限科技发展有限公司 | A kind of method and device of vehicle peccancy detection |
| CN108200708A (en) * | 2018-02-07 | 2018-06-22 | 宁夏乐智博科技有限公司 | Control method and intelligent road-lamp |
| CN108337786A (en) * | 2018-02-07 | 2018-07-27 | 宁夏乐智博科技有限公司 | Illumination control method and intelligent road-lamp |
| CN108337785A (en) * | 2018-02-07 | 2018-07-27 | 宁夏乐智博科技有限公司 | Illumination control method and intelligent illuminating system |
| CN108734959A (en) * | 2018-04-28 | 2018-11-02 | 扬州远铭光电有限公司 | A kind of embedded vision train flow analysis method and system |
| CN108896062A (en) * | 2018-05-11 | 2018-11-27 | 长安大学 | A kind of device and method using the electronic map group's of prompting mist section in real time |
| CN108749708B (en) * | 2018-05-11 | 2021-10-22 | 长安大学 | Device and method for automatically turning on fog lamp of vehicle under low visibility |
| CN108749708A (en) * | 2018-05-11 | 2018-11-06 | 长安大学 | The device and method of vehicle fog lamp is automatically turned under a kind of low visibility |
| CN109085570A (en) * | 2018-06-10 | 2018-12-25 | 南京理工大学 | Automobile detecting following algorithm based on data fusion |
| CN110596696A (en) * | 2018-06-13 | 2019-12-20 | 通用汽车环球科技运作有限责任公司 | Apparatus and method for improved radar beamforming |
| CN110596696B (en) * | 2018-06-13 | 2024-03-15 | 通用汽车环球科技运作有限责任公司 | Apparatus and method for improved radar beamforming |
| CN109255857A (en) * | 2018-08-23 | 2019-01-22 | 江苏罗思韦尔电气有限公司 | A kind of recorder with safety precaution function |
| CN109102702A (en) * | 2018-08-24 | 2018-12-28 | 南京理工大学 | Vehicle speed measuring method based on video encoder server and Radar Signal Fusion |
| CN109613537A (en) * | 2019-01-16 | 2019-04-12 | 南京奥杰智能科技有限公司 | A kind of hologram radar |
| CN109646796A (en) * | 2019-01-17 | 2019-04-19 | 浙江大学 | Channel wireless radio multi closed loop stimulation system for epilepsy therapy |
| CN109658715A (en) * | 2019-01-31 | 2019-04-19 | 厦门精益远达智能科技有限公司 | Statistical method of traffic flow, device, equipment and the storage medium of multilane |
| CN109658715B (en) * | 2019-01-31 | 2021-01-22 | 厦门精益远达智能科技有限公司 | Multilane traffic flow statistical method, device, equipment and storage medium |
| CN112017448A (en) * | 2019-05-29 | 2020-12-01 | 浙江宇视科技有限公司 | Traffic detection method, signal lamp control method, device, equipment and storage medium |
| WO2020253227A1 (en) * | 2019-06-17 | 2020-12-24 | 创新先进技术有限公司 | Video processing method, apparatus and device, and video monitoring system |
| US10942267B2 (en) | 2019-06-17 | 2021-03-09 | Advanced New Technologies Co., Ltd. | Video object processing |
| WO2021003896A1 (en) * | 2019-07-10 | 2021-01-14 | 南京慧尔视智能科技有限公司 | Multi-target information fusion and visual presentation method based on microwaves and videos |
| CN110288832A (en) * | 2019-07-10 | 2019-09-27 | 南京慧尔视智能科技有限公司 | It is merged based on microwave with the multiple-object information of video and visual presentation method |
| CN110428647A (en) * | 2019-07-26 | 2019-11-08 | 百度在线网络技术（北京）有限公司 | The vehicle collaboration at crossing current method, apparatus, equipment and storage medium |
| CN110444026A (en) * | 2019-08-06 | 2019-11-12 | 北京万集科技股份有限公司 | The triggering grasp shoot method and system of vehicle |
| CN110444026B (en) * | 2019-08-06 | 2021-07-09 | 北京万集科技股份有限公司 | Triggering snapshot method and system for vehicle |
| CN110533921A (en) * | 2019-08-06 | 2019-12-03 | 北京万集科技股份有限公司 | The triggering grasp shoot method and system of vehicle |
| CN110647806A (en) * | 2019-08-13 | 2020-01-03 | 浙江大华技术股份有限公司 | Object behavior monitoring method, device, equipment, system and storage medium |
| CN110515073A (en) * | 2019-08-19 | 2019-11-29 | 南京慧尔视智能科技有限公司 | The trans-regional networking multiple target tracking recognition methods of more radars and device |
| CN110728839A (en) * | 2019-10-23 | 2020-01-24 | 昆山宝创新能源科技有限公司 | Vehicle and illegal punishment method and device thereof |
| CN110827542A (en) * | 2019-11-11 | 2020-02-21 | 江苏中路工程技术研究院有限公司 | Highway safety vehicle distance early warning system |
| CN111126444A (en) * | 2019-11-28 | 2020-05-08 | 天津津航技术物理研究所 | Classifier integration method |
| CN111178215A (en) * | 2019-12-23 | 2020-05-19 | 深圳成谷科技有限公司 | Sensor data fusion processing method and device |
| CN111178215B (en) * | 2019-12-23 | 2024-03-08 | 深圳成谷科技有限公司 | Sensor data fusion processing method and device |
| CN113255708A (en) * | 2020-02-10 | 2021-08-13 | 富士通株式会社 | Data fusion method and device and data processing equipment |
| CN111554111B (en) * | 2020-04-21 | 2021-04-20 | 河北万方中天科技有限公司 | Signal timing optimization method and device based on multi-source data fusion and terminal |
| CN111554111A (en) * | 2020-04-21 | 2020-08-18 | 河北万方中天科技有限公司 | Signal timing optimization method and device based on multi-source data fusion and terminal |
| CN111526646A (en) * | 2020-05-06 | 2020-08-11 | 辽宁百思特达半导体科技有限公司 | Highway service compartment external lighting system based on wisdom lamp pole |
| CN112150799A (en) * | 2020-08-19 | 2020-12-29 | 上海图丽信息技术有限公司 | Method for collecting road vehicle traffic big data by fusing radar videos |
| CN112419712A (en) * | 2020-11-04 | 2021-02-26 | 同盾控股有限公司 | Road section vehicle speed detection method and system |
| CN112991774A (en) * | 2021-02-02 | 2021-06-18 | 深圳市天鼎微波科技有限公司 | Traffic flow detection method and device based on millimeter waves |
| CN115331458A (en) * | 2021-05-11 | 2022-11-11 | 北京万集科技股份有限公司 | Signal lamp control method, device, equipment and storage medium |
| CN113438449A (en) * | 2021-06-07 | 2021-09-24 | 西安恒盛安信智能技术有限公司 | Video image transmission method |
| CN116740944A (en) * | 2023-08-10 | 2023-09-12 | 云南省交通投资建设集团有限公司 | Driving safety early warning method, device and storage medium in road tunnel |
| CN116740944B (en) * | 2023-08-10 | 2023-10-27 | 云南省交通投资建设集团有限公司 | Driving safety early warning method, device and storage medium in road tunnel |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106373394B (en) | 2019-01-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106373394A (en) | Vehicle detection method and system based on video and radar | |
| CN106919915B (en) | Map road marking and road quality acquisition device and method based on ADAS system | |
| WO2020000251A1 (en) | Method for identifying video involving violation at intersection based on coordinated relay of video cameras | |
| CN104616502B (en) | Car license recognition and alignment system based on combination type bus or train route video network | |
| CN105512720B (en) | A kind of public transit vehicle passenger flow statistics method and system | |
| CN101458871B (en) | Intelligent traffic analysis system and application system thereof | |
| CN104537841B (en) | Unlicensed vehicle violation detection method and detection system thereof | |
| Fernández-Caballero et al. | Road-traffic monitoring by knowledge-driven static and dynamic image analysis | |
| CN106326893A (en) | Vehicle color recognition method based on area discrimination | |
| CN109064495A (en) | A kind of bridge floor vehicle space time information acquisition methods based on Faster R-CNN and video technique | |
| CN112700470A (en) | Target detection and track extraction method based on traffic video stream | |
| CN105608431A (en) | Vehicle number and traffic flow speed based highway congestion detection method | |
| CN104029680A (en) | Lane departure warning system and method based on monocular camera | |
| CN106096525A (en) | A kind of compound lane recognition system and method | |
| CN106778593A (en) | A kind of track level localization method based on the fusion of many surface marks | |
| CN107886055A (en) | A kind of retrograde detection method judged for direction of vehicle movement | |
| CN103164958B (en) | Method and system for vehicle monitoring | |
| CN102902983B (en) | A kind of taxi identification method based on support vector machine | |
| CN106778540A (en) | Parking detection is accurately based on the parking event detecting method of background double layer | |
| CN105957341A (en) | Wide area traffic jam detection method based on unmanned plane airborne platform | |
| CN104267209B (en) | Method and system for expressway video speed measurement based on virtual coils | |
| CN103605960A (en) | Traffic state identification method based on fusion of video images with different focal lengths | |
| CN110909607B (en) | Passenger flow sensing device system in intelligent subway operation | |
| CN115440071B (en) | Automatic driving illegal parking detection method | |
| CN105069407A (en) | Video-based traffic flow acquisition method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20220509 Address after: 518000 d303, Shenzhen Aerospace Science and Technology Innovation Research Institute building, No. 6, science and technology south 10th Road, high tech Zone, Yuehai street, Nanshan District, Shenzhen, Guangdong Patentee after: SHENZHEN SHANGQIAO INFORMATION TECHNOLOGY CO.,LTD. Address before: 518057 Room 201, building a, building 1, Qianwan 1st Road, Qianhai Shenzhen Hong Kong cooperation zone, Shenzhen, Guangdong Patentee before: SHENZHEN SHANGQIAO TRAFFIC TECHNOLOGY CO.,LTD. |
|
| TR01 | Transfer of patent right |