The content of the invention
The embodiment of the present invention technical problem to be solved is how in the case where additional laser radar is not increased, there is provided oneThe system and method for planting accurate identification vehicle axles.
A kind of vehicle axles identifying system based on laser radar is embodiments provided, including：
Laser radar, speculum and data processor；
The laser radar is installed on trackside top preset height, and perpendicular to road surface vehicle to be measured is scanned, and obtains the first viaFace information of vehicles, and first road surface information of vehicles is sent to the data processor；
The speculum is located at laser radar lower section, with road surface into default angle, through treating described in speculum reflectionMeasuring car causes the laser radar to obtain the second road vehicles information, and causes the laser radar by second road surface carInformation is sent to the data processor；
The data processor is used for first road surface information of vehicles and the second road vehicles informationReason, recognizes the axle information of the vehicle to be measured.
Alternatively, the scanning area of the laser radar includes the first scanning area and the second scanning area, directly scansThe region of the vehicle described to be measured for arriving is the first scanning area, and the laser radar reflects the institute scanned through the speculumThe region for stating vehicle to be measured is the second scanning area；
The section of the section of first scanning area and second scanning area is in same plane.
Alternatively, the preset height is less than 20 meters more than 2 meters, and the predetermined angle is less than 90 degree more than 20 degree.
A kind of vehicle axles recognition methods based on laser radar of employing said system is embodiments provided, is wrappedInclude：
During the scanning area that vehicle to be measured passes through laser radar along road traveling, data processor receives laser thunderThe scanning frame of the vehicle odometry information multiple to be measured obtained up to scanning；The vehicle odometry information to be measured includes the laser radarDirectly scan the second tunnel that the first road surface information of vehicles for getting and the reflected mirror reflection scanning of the laser radar getFace information of vehicles；
First road surface information of vehicles and the second road vehicles information are processed, the vehicle to be measured is obtainedVehicle's contour information and vehicle axles information；
Registration is carried out to the vehicle's contour information and vehicle axles information to judge and identification, is determined and the plurality of is treated measuring carThe scanning frame of ranging information includes the targeted scans frame of vehicle axles；
According to the scanning frame and the targeted scans frame of the plurality of vehicle odometry information to be measured to the vehicle to be measuredWheel shaft is identified.
Alternatively, it is described that first road surface information of vehicles and the second road vehicles information are carried out by process and included：
Set up rectangular coordinate system；Perpendicular to the x-axis for the rectangular coordinate system of vehicle heading, vertical direction is instituteState the y-axis of rectangular coordinate system, the origin that be projected as the rectangular coordinate system of the laser radar on road surface；
Carry out coordinate to the scanning element in first road surface information of vehicles and the second road vehicles information respectively to turnChange, obtain vehicle's contour information sequence x1 (m) and vehicle axles information sequence x2 (n).
Alternatively, the scanning element in first road surface information of vehicles and the second road vehicles information carried out respectivelyCoordinate Conversion, including：
Scanning element in 1 pair of first road surface information of vehicles of formula carries out Coordinate Conversion：
X=L × sin α, y=H-L × cos α formula 1
Scanning element in 2 pairs of the second road vehicles information of formula carries out Coordinate Conversion：
X=L × cos α-(H-H1), y=H1-L × sin α formula 2
Wherein, x is the abscissa of scanning element, and y is the ordinate of scanning element；L is the laser radar to the scanning elementThe distance between, H is distance of the laser radar to road surface, and H1 is the distance of the speculum and y-axis intersection point to road surface, αFor laser scanning line and the angle of y-axis.
Alternatively, it is described that registration judgement and identification are carried out to the vehicle's contour information and vehicle axles information, including：
Vehicle's contour information sequence x1 (m) and the feature in vehicle axles information sequence x2 (n) are extracted respectivelyInformation；
According to the characteristic information to vehicle's contour information sequence x1 (m) and the vehicle axles information sequence x2N () is matched, determine the scanning frame of the plurality of vehicle odometry information to be measured and include the targeted scans frame of vehicle axles.
Alternatively, the characteristic information includes：
Tire, wheel hub, tire and vehicle body gap and mud guard.
Alternatively, the scanning frame and the targeted scans frame according to the plurality of vehicle odometry information to be measured is to describedThe wheel shaft of vehicle to be measured is identified, including：
Using the wheel shaft central point in the targeted scans frame as signature identification point, according to the vehicle's contour information sequenceX1 (m) and vehicle axles information sequence x2 (n) determine the abscissa of each signature identification point；
Characteristic point sequence t (k) is determined according to the abscissa of each signature identification point；
The wheel shaft of the vehicle to be measured is identified according to characteristic point sequence t (k).
Alternatively, it is described the wheel shaft of the vehicle to be measured is identified according to characteristic point sequence t (k), including：
Process is filtered to characteristic point sequence t (k), according to characteristic point sequence t (k) after filtering process to describedThe wheel shaft of vehicle to be measured is identified.
Vehicle axles identifying system and method based on laser radar provided in an embodiment of the present invention, vehicle to be measured is along road surfaceDuring driving through the scanning area of laser radar, the vehicle odometry information multiple to be measured that Laser Radar Scanning is obtained is receivedScanning frame；Vehicle odometry information to be measured directly scans the first road surface information of vehicles and laser thunder for getting including laser radarThe second road vehicles information that scanning gets is reflected up to reflected mirror；To the first road surface information of vehicles and the second road vehicles letterBreath is processed, and obtains the vehicle's contour information and vehicle axles information of vehicle to be measured；To vehicle's contour information and vehicle axlesInformation carries out registration and judges and identification that determining the scanning frame of multiple vehicle odometry informations to be measured includes that the target of vehicle axles is sweptRetouch frame；It is identified according to the wheel shaft that the scanning frame and targeted scans frame of multiple vehicle odometry informations to be measured treat measuring car.ThisInventive embodiments are treated the wheel shaft of measuring car in the case where additional laser radar is not increased and are accurately recognized, install simple,Easy to maintenance, cost price is relatively low.
Specific embodiment
To make purpose, technical scheme and the advantage of the embodiment of the present invention clearer, below in conjunction with the embodiment of the present inventionIn accompanying drawing, clear, complete description is carried out to the technical scheme in the embodiment of the present invention, it is clear that described embodiment isThe a part of embodiment of the present invention, rather than the embodiment of whole.Based on the embodiment in the present invention, those of ordinary skill in the artThe every other embodiment obtained under the premise of creative work is not made, belongs to the scope of protection of the invention.
Fig. 1 is the structural representation of the vehicle axles identifying system based on laser radar of one embodiment of the invention.Such asShown in Fig. 1, the system of the embodiment of the present invention includes：
Laser radar 11, speculum 12 and data processor 13；
Laser radar 11 is installed on trackside top preset height, and perpendicular to road surface vehicle V to be measured is scanned, and obtains the first road surfaceInformation of vehicles, and first road surface information of vehicles is sent to data processor；
Speculum 12 is located at laser radar lower section, and with road surface into default angle, reflecting vehicle V to be measured through speculum 12 makesObtain the laser radar and obtain the second road vehicles information, and laser radar 11 is sent the second road vehicles informationTo the data processor；
The data processor is used for first road surface information of vehicles and the second road vehicles informationReason, recognizes the axle information of vehicle V to be measured.
The vehicle axles identifying system based on laser radar of the embodiment of the present invention, by below laser radar with road surfaceSpeculum is set into default angle, the scanning area of laser radar is expanded, it is right in the case where additional laser radar is not increasedThe wheel shaft of vehicle to be measured is accurately recognized that install simple, easy to maintenance, cost price is relatively low.
Further, the scanning area of the laser radar includes the first scanning area and the second scanning area, directly sweepsThe region of the vehicle described to be measured retouched is the first scanning area, and the laser radar is scanned through speculum reflectionThe region of the vehicle to be measured is the second scanning area；
The section of the section of first scanning area and second scanning area is in same plane.
It will be appreciated that the section of the first scanning area can simplify with the section of second scanning area in same planeThe process of Coordinate Conversion, will be specifically described below in embodiment of the method.
In actual applications, laser radar 11, the scanning of laser radar 11 are installed on the edge-on bar or portal frame of tracksidePerpendicular to vehicle heading, the preset height is less than 20 meters more than 2 meters, with road surface into fixed angles below laser radar in faceDegree installs speculum 12, and the predetermined angle is less than 90 degree more than 20 degree, and the setting height(from bottom) scope of speculum 12 is 0.1 to 2 meters, is madeThe scanning plane part for obtaining laser radar directly detects vehicle to be measured, and a part detects vehicle to be measured by speculum, swashsOptical radar can be accurately measured the distance of each scanning element of vehicle to be measured.The reflectivity of speculum is not less than 80%.
Fig. 2 is the schematic flow sheet of the vehicle axles recognition methods based on laser radar of one embodiment of the invention.Such asShown in Fig. 2, the system of the vehicle axles recognition methods based on laser radar of the embodiment of the present invention based on above-described embodiment, bagInclude：
S21：During the scanning area that vehicle to be measured passes through laser radar along road traveling, Laser Radar Scanning is receivedThe scanning frame of the vehicle odometry information multiple to be measured for obtaining；The vehicle odometry information to be measured is directly swept including the laser radarRetouch the second road vehicles that the first road surface information of vehicles for getting and the reflected mirror reflection scanning of the laser radar getInformation；
It should be noted that laser radar carries out rotation sweep according to fixed frequency, and usually, the range finding of laser radarAbility is that between 20 meters to 60 meters, the angular resolution of scanning is 0.125 ° to 0.5 ° adjustable, and the angle of scanning is 90 ° to 360 °Between, the motor of laser radar often rotates a circle, and obtains the scanning frame of a vehicle odometry information to be measured.Measuring car is treated according to eachThe scanning frame of ranging information, if several scanning frames continuously occur is not detected by effective overall height information, it is believed that treat measuring carScanning area is passed through completely.
S22：First road surface information of vehicles and the second road vehicles information are processed, is obtained described to be measuredThe vehicle's contour information and vehicle axles information of vehicle；
S23：Registration is carried out to the vehicle's contour information and vehicle axles information to judge and identification, determines the plurality for the treatment ofSurveying the scanning frame of vehicle odometry information includes the targeted scans frame of vehicle axles；
S24：Measuring car is treated to described according to the scanning frame and the targeted scans frame of the plurality of vehicle odometry information to be measuredWheel shaft be identified.
Vehicle axles recognition methods based on laser radar provided in an embodiment of the present invention, is not increasing additional laser radarIn the case of treat the wheel shaft of measuring car and accurately recognized, install simple, easy to maintenance, cost price is relatively low.
In a kind of optional embodiment of the embodiment of the present invention, step S22 includes：
Set up rectangular coordinate system；Perpendicular to the x-axis for the rectangular coordinate system of vehicle heading, vertical direction is instituteState the y-axis of rectangular coordinate system, the origin that be projected as the rectangular coordinate system of the laser radar on road surface；
Carry out coordinate to the scanning element in first road surface information of vehicles and the second road vehicles information respectively to turnChange, obtain vehicle's contour information sequence x1 (m) and vehicle axles information sequence x2 (n).
As shown in figure 3, send a little origin O is projected as on road surface with the laser of described laser radar, along roadThe direction in section is x-axis, and the direction along laser radar is y-axis.Then the coordinate of laser radar laser emission point S is (0, H), thenThe coordinate (- (H-H1), H1) of mirror point S '.
It is the first scanning area of the vehicle described to be measured being directly scanned out from scan line 1 to the region scan line 4,Hereinafter referred to as scanning area 1, is to scan through speculum reflection from scan line 5 to the region scan line 6Second scanning area of the vehicle to be measured, hereinafter referred to as scanning area 2.
If the section of the section of scanning area 1 and scanning area 2 is at grade, the coordinate that can simplify scanning element turnsThe process changed.In actual applications, the installation of speculum is if there is tilting, section and the scanning area 2 of scanning area 1Section not at grade, it is also possible to carry out the identification of vehicle axles, now needing to increase inclination angle carries out the conversion of coordinate.
Specifically, the scanning element in first road surface information of vehicles and the second road vehicles information carried out respectivelyCoordinate Conversion, including：
Scanning element in 1 pair of first road surface information of vehicles of formula carries out Coordinate Conversion：
X=L × sin α, y=H-L × cos α formula 1
Scanning element in 2 pairs of the second road vehicles information of formula carries out Coordinate Conversion：
X=L × cos α-(H-H1), y=H1-L × sin α formula 2
Wherein, x is the abscissa of scanning element, and y is the ordinate of scanning element；L is the laser radar to the scanning elementThe distance between, H is distance of the laser radar to road surface, and H1 is the distance of the speculum and y-axis intersection point to road surface, αFor laser scanning line and the angle of y-axis.
It will be appreciated that scanning element is in x-axis positive axis, the value of α is just, conversely, negative semiaxis of the scanning element in x-axis, the value of αIt is negative.After coordinate transform, the sequence that scanning area 1 is obtained be vehicle's contour information sequence x1 (m), the sequence that scanning area 2 is obtainedIt is classified as vehicle axles information sequence x2 (n).
Further, step S23 includes：
Vehicle's contour information sequence x1 (m) and the feature in vehicle axles information sequence x2 (n) are extracted respectivelyInformation；
According to the characteristic information to vehicle's contour information sequence x1 (m) and the vehicle axles information sequence x2N () is matched, determine the scanning frame of the plurality of vehicle odometry information to be measured and include the targeted scans frame of vehicle axles.
Specifically, according to sequence x1 (m), in the range of road width, the judgement of height is carried out；Limit sequence x1 (m)X coordinate scope, for the scanning element in the range of pavement of road, the judgement of y-axis value is carried out, in the event of continuous scanningThe y values of point exceed threshold value, such as 300 millimeters, it is believed that detect vehicle, carry out single frames axletree recognition logic.
According to all x1 (m) information of a car, it is distributed according to the priority of sweep time, it can be deduced that the three-dimensional letter of vehicleBreath, according to three-D profile, can distinguish passenger-cargo.
Specifically, the characteristic information includes：
Tire, wheel hub, tire and vehicle body gap and mud guard.
According to the characteristic information to vehicle's contour information sequence x1 (m) and the vehicle axles information sequence x2N () is matched, specially：
The contour shape of analytical sequence x1 (m), investigates the saltus step situation of less than 1 meter of scanning element, according to the jump of y-axis valueBecome, tentatively find the profile of axletree, then the information inside the wheel that scan is analyzed, if x-axis change is notable, y-axisValue gradually increases, and it is single tire situation to tire or tire that explanation is scanning, if the feature that x reduces after occurring increasing, y-axis is graduallyThe situation of increase, illustrates the twins scanned to tire type；
The contour shape of analytical sequence x2 (n), according to the preliminary judgement in sequence x1 (m), it is determined that being suspected to be the ground of wheel shaftSide whether there is wheel shaft detailed features, and such as Fig. 4 a and 4b are to detect vehicle but the frame does not have the situation of axletree information, whereinFig. 4 a are the curve map of sequence x2 (n), and Fig. 4 b are the curve map of sequence x1 (m)；Fig. 4 c and Fig. 4 d is to detect car and tire isThe situation of twins, Fig. 4 c are the curve map of sequence x2 (n), and Fig. 4 d are the curve map of sequence x1 (m)；Fig. 4 e and Fig. 4 f is to detectCar and tire are the situation of single tire, Fig. 4 e for sequence x2 (n) curve map, Fig. 4 f for sequence x1 (m) curve map.
Further, step S24 includes：
Using the wheel shaft central point in the targeted scans frame as signature identification point, according to the vehicle's contour information sequenceX1 (m) and vehicle axles information sequence x2 (n) determine the abscissa of each signature identification point；
Characteristic point sequence t (k) is determined according to the abscissa of each signature identification point；
The wheel shaft of the vehicle to be measured is identified according to characteristic point sequence t (k).
If it should be noted that vehicle's contour information sequence x1 (m) and vehicle axles information sequence x2 (n) areCar is to axle information and match point information is consistent for inspection, it is believed that detect wheel shaft, and will detect the central point of wheel shaft asSignature identification point, according to the abscissa of each signature identification point characteristic point sequence t (k) is determined.
Specifically, it is described the wheel shaft of the vehicle to be measured is identified according to characteristic point sequence t (k), including：
Process is filtered to characteristic point sequence t (k), according to characteristic point sequence t (k) after filtering process to describedThe wheel shaft of vehicle to be measured is identified.
As shown in figure 5, the distribution situation for wheel shaft characteristic point over time, it can be seen that signature identification point is presented two ripplesPeak, illustrates that the car has two wheel shafts, due to each frame estimate of situation not necessarily very accurately, there is the situation of erroneous judgement, can be withAccording to the coordinate trend of overall signature identification point, process is filtered, so as to accurately obtain Vehicle Axles information.
The vehicle axles recognition methods based on laser radar of the embodiment of the present invention, using laser radar road section is obtainedRanging information, the ranging information is comprising being directly scanned out the region and the area through scanning mirror to road side on road surfaceDomain；Data processor, according to the difference in region, carries out respectively coordinate transform to the ranging information for receiving, and two are scannedThe information of vehicles in region is detected, if detecting information of vehicles, carries out matching detection to the axletree characteristic of the frame, and by axletreeFeature is preserved；Treating all scanning frame informations of measuring car carries out comprehensive analysis, rejects single frames and judges the wrong miscellaneous of informationScatterplot, obtains the precise information of Vehicle Axles.Using a laser radar carry out vehicle's contour analysis carry out car lorry areaWhile dividing, the axletree information of vehicle can be accurately obtained so that the judgement of lorry is finer.
Vehicle axles identifying system and method based on laser radar provided in an embodiment of the present invention, vehicle to be measured is along road surfaceDuring driving through the scanning area of laser radar, the vehicle odometry information multiple to be measured that Laser Radar Scanning is obtained is receivedScanning frame；Vehicle odometry information to be measured directly scans the first road surface information of vehicles and laser thunder for getting including laser radarThe second road vehicles information that scanning gets is reflected up to reflected mirror；To the first road surface information of vehicles and the second road vehicles letterBreath is processed, and obtains the vehicle's contour information and vehicle axles information of vehicle to be measured；To vehicle's contour information and vehicle axlesInformation carries out registration and judges and identification that determining the scanning frame of multiple vehicle odometry informations to be measured includes that the target of vehicle axles is sweptRetouch frame；It is identified according to the wheel shaft that the scanning frame and targeted scans frame of multiple vehicle odometry informations to be measured treat measuring car.ThisInventive embodiments are treated the wheel shaft of measuring car in the case where additional laser radar is not increased and are accurately recognized, install simple,Easy to maintenance, cost price is relatively low.
Those skilled in the art are it should be appreciated that embodiments of the invention can be provided as method, system or computer programProduct.Therefore, the present invention can be using complete hardware embodiment, complete software embodiment or with reference to the reality in terms of software and hardwareApply the form of example.And, the present invention can be adopted and wherein include the computer of computer usable program code at one or moreThe computer program implemented in usable storage medium (including but not limited to magnetic disc store, CD-ROM, optical memory etc.) is producedThe form of product.
The present invention is the flow process with reference to method according to embodiments of the present invention, equipment (system) and computer programFigure and/or block diagram are describing.It should be understood that can be by computer program instructions flowchart and/or each stream in block diagramThe combination of journey and/or square frame and flow chart and/or the flow process in block diagram and/or square frame.These computer programs can be providedThe processor of all-purpose computer, special-purpose computer, Embedded Processor or other programmable data processing devices is instructed to produceA raw machine so that produced for reality by the instruction of computer or the computing device of other programmable data processing devicesThe device of the function of specifying in present one flow process of flow chart or one square frame of multiple flow processs and/or block diagram or multiple square frames.
It should be noted that term " including ", "comprising" or its any other variant are intended to the bag of nonexcludabilityContain, so that a series of process, method, article or equipment including key elements is not only including those key elements, but also includingOther key elements being not expressly set out, or also include the key element intrinsic for this process, method, article or equipment.In the absence of more restrictions, the key element for being limited by sentence "including a ...", it is not excluded that including the key elementProcess, method, article or equipment in also there is other identical element.