CN204422039U - The pinpoint True k Scale Weighing System of vehicle based on license plate - Google Patents

Abstract

The utility model provides a kind of vehicle based on license plate pinpoint True k Scale Weighing System, comprising: two video cameras being positioned at weighing table two ends; Comprise the truck scale of weighing controller, weighing table and LOAD CELLS, weighing controller comprises the picture recognition module, image processing module, position determination module and the Weighing control module that connect successively.Picture recognition module obtains front truck License Plate Image and rear car License Plate Image; Image processing module determines the path coordinate of the intersection point of pedal line and the road plane in forward and backward license plate image, license plate centrally put according to image and path coordinate system transformational relation, thus centrally puts the path coordinate on road plane on determining; Position determination module based on the position of this path coordinate determination vehicle, to carry out Weighing control based on this vehicle location to vehicle.Based on the location of license plate image being realized to the accurate location of vehicle in weighing area, the accurate control of guarantee vehicle weighing and weighing results are accurately and reliably.

Description

The pinpoint True k Scale Weighing System of vehicle based on license plate

Technical field

The utility model belongs to truck scale technical field, especially relates to the pinpoint True k Scale Weighing System of a kind of vehicle based on license plate.

Background technology

Truck scale is weighed and to be obtained a wide range of applications in China's trade settlement.Affect truck scale and weigh the factor of accuracy except the environment for use of truck scale, the subjectivity interference also having driver to implement in truck scale weighing process.Driver disturbs the weigh major way of accuracy of truck scale to be require vehicle to be parked on the assigned address of weighing table not in accordance with weighing of truck scale, and the modes such as weighing apparatus illegally gone up by weighing apparatus (as wheel flanging, wheel keep to the side) as not exclusively upper in vehicle and front (afterwards) face vehicle affects weighing precision.Therefore, realize vehicle location detection to have important practical significance to the precise of truck scale.

At present, conventional vehicle location checkout equipment is vehicle separator, adopts infrared mode more, is made up of infrared transmitter and infrared ray receiver.When vehicle does not sail weighing table completely into, far infrared will be blocked in the front end of vehicle or rear end, make infrared remote receiver not receive infrared signal.At this moment, vehicle separator does not send " permission " weighing-up wave to weighing controller, and truck scale can not be weighed, and staff also cannot carry out the work such as logging data printing reports.Only have after vehicle sails weighing table into completely, vehicle separator sends " permission " weighing-up wave, truck scale just can be weighed.Thus, the phenomenon that the weighting error that causes because vehicle does not sail weighing table into is completely large can be avoided, improve the accuracy of weighing data.

But above-mentioned employing infrared ray carries out weighing the mode of location of vehicle, can be subject to the impact such as surrounding environment and temperature, make position testing result not accurate enough.Such as, if infrared remote receiver is that will face two kinds of interference: 1. environmental interference, and the crowd outside glass door and window and the flowing of vehicle all can have influence on the erroneous judgement of infrared remote receiver to vehicle location signal facing to glass door and window; 2. white light interference, infrared remote receiver has very strong inhibit feature to white light, can cause infrared remote receiver sensitivity decrease-wrong report.For another example, the cold and hot change of ambient air temperature causes the wrong report of infrared remote receiver.And these all will cause the inaccurate of vehicle location positioning result, thus cause the inaccurate, unreliable of vehicle weighing result the most at last.

Utility model content

For above-mentioned Problems existing, the utility model proposes the pinpoint True k Scale Weighing System of a kind of vehicle based on license plate, easily by the impact of environment, the inaccurate defect of truck scale weighing results is caused in order to overcome vehicle location in existing True k Scale Weighing System.

The utility model provides the pinpoint True k Scale Weighing System of a kind of vehicle based on license plate, comprising:

First video camera, the second video camera and truck scale, described truck scale comprises weighing controller, weighing table and LOAD CELLS;

Described first position for video camera is in described weighing table endpiece predeterminable range place, described second position for video camera is in described weighing table inlet end predeterminable range place, be respectively used to the collection described vehicle being carried out to the first image and the second image after described vehicle enters the weighing area of described weighing table, wherein, described first image and described second image are the image collected the same time;

Described weighing controller comprises the picture recognition module, image processing module, position determination module and the Weighing control module that connect successively;

Described first video camera is connected with described picture recognition module respectively with described second video camera, described LOAD CELLS and described Weighing control model calling;

Described picture recognition module, for obtaining the front truck License Plate Image in described first image of described first video camera transmission, and obtains the rear car License Plate Image in described second image of described second video camera transmission;

Described image processing module, for respectively using described front truck License Plate Image and described rear car License Plate Image as pending license plate image, according to the ordinate transform relation of the camera image plane set up in advance and road plane, determine the path coordinate (x in described pending license plate image, license plate centrally being put the pedal line CD of C and the intersection point D of road plane _d, y _d, 0), and according to described ordinate transform relation, determine the height H described license plate centrally put between C and described intersection point D;

Described image processing module, also for according to the path coordinate of described intersection point D and described height H, determine described on centrally to put the path coordinate of C on described road plane be (x _d, y _d, H);

Described position determination module, for centrally putting path coordinate on described road plane according to front license plate upper and the path coordinate on described road plane is centrally put in rear the upper of license plate, determines the position of described vehicle in described weighing area;

Described Weighing control module, if determine that described vehicle meets default weighing positions condition in the position of described weighing area for described position determination module, then controls described LOAD CELLS and weighs to described vehicle.

Further, described image processing module, specifically for:

Adopt dilating window method, the top of fixing dilating window in described pending license plate image on license plate edge transverse horizontal line on, the bottom of flexible described dilating window, determines the present image coordinate (u of two end points M and N in bottom of described dilating window _m,v _m) and (u _n,v _n);

According to described ordinate transform relation, determine the present image coordinate (u of described two end points M and N respectively _m,v _m) and (u _n,v _n) path coordinate (x of correspondence on described road plane _m,y _m,0) and (x _n,y _n,0);

According to described path coordinate (x _m,y _m,0) and (x _n,y _n,0), determine that the width w of described license plate self is: further, the path coordinate (x of described intersection point D is determined _d, y _d, 0) be: (x _d, y _d, 0) and=((x _m+ x _n)/2, (y _m+ y _n)/2,0);

According to described ordinate transform relation, determine the height H in described pending license plate image, license plate centrally put between C and described intersection point D;

Determine whether described w and described H meets and preset flexible cut-off condition, described cut-off condition of stretching of presetting is:

min f(u,v)＝(1-a)(|w-w ₀|/w ₀)+a(|H-H ₀|/H ₀)

S.t. the width of license plate self: w ∈ [42cm, 46cm]

Front license plate setting height(from bottom): H ∈ [42cm, 62cm]

Rear license plate setting height(from bottom): H ∈ [68cm, 108cm]

If meet, then determine that in described pending license plate image, license plate centrally putting the path coordinate of C on described road plane is (x _d, y _d, H);

Wherein, (u, v) refers to the image coordinate variable of two end points M and N in bottom of dilating window; w ₀for license plate normal width; The distributed density values of a corresponding to H described in license plate setting height(from bottom); H ₀in distribution density for license plate setting height(from bottom), the license plate setting height(from bottom) corresponding to maximum distribution density points.

Further, described image processing module, also for:

According to ordinate transform formula (1), determine the present image coordinate (u of described two end points M and N respectively _m,v _m) and (u _n,v _n) path coordinate (x of correspondence on described road plane _m,y _m,0) and (x _n,y _n,0):

$\left[\begin{array}{ccc}{r}_{11}& r_{12}& -\frac{u_p-u_0}{f_u}\\ r_{21}& r_{22}& \frac{v_p-v_0}{f_v}\\ r_{31}& r_{32}& -1\end{array}\right]\left[\begin{array}{c}{x}_P\\ y_P\\ z^c\end{array}\right]=\left[\begin{array}{ccc}{r}_{11}& r_{12}& r_{13}\\ r_{21}& r_{22}& r_{23}\\ r_{31}& r_{32}& r_{33}\end{array}\right]\left[\begin{array}{c}{t}_x\\ t_y\\ t_z\end{array}\right]---\left(1\right)$

According to ordinate transform formula (2), determine the height H in described pending license plate image, license plate centrally put between C and described intersection point D:

$\left[\begin{array}{cc}-r_{13}& \frac{u_q-u_0}{f_u}\\ -r_{23}& -\frac{v_q-v_0}{f_v}\\ -r_{33}& 1\end{array}\right]\left[\begin{array}{c}{z}_Q-t_z\\ z^c\end{array}\right]=\left[\begin{array}{cc}{r}_{11}& r_{12}\\ r_{21}& r_{22}\\ r_{31}& r_{32}\end{array}\right]\left[\begin{array}{c}{x}_P-t_x\\ y_P-t_y\end{array}\right]---\left(2\right)$

Wherein, make $A=\left[\begin{array}{ccc}{f}_u& 0& u_0\\ 0& f_v& v_0\\ 0& 0& 1\end{array}\right]$ For intrinsic parameters of the camera matrix, wherein, f _uand f _vfor focal length of camera, u ₀and v ₀for the origin of the plane of delineation; R and T is made to be video camera external parameter, wherein, $R=\left[\begin{array}{ccc}{r}_{11}& r_{12}& r_{13}\\ r_{21}& r_{22}& r_{23}\\ r_{31}& r_{32}& r_{33}\end{array}\right]$ For rotation matrix, $T=\left[\begin{array}{c}{t}_x\\ t_y\\ t_z\end{array}\right]$ For translation vector; z ^cfor the coordinate in camera coordinate system, determine according to following pin-hole imaging model:

$\left[\begin{array}{c}u-u_0\\ v-v_0\end{array}\right]=\left[\begin{array}{c}{f}_u\frac{x^c}{z^c}\\ -f_v\frac{y^c}{z^c}\end{array}\right],\left[\begin{array}{c}{x}^c\\ y^c\\ z^c\end{array}\right]=R\left(\begin{array}{c}\left[\begin{array}{c}x\\ y\\ z\end{array}\right]-T\end{array}\right)$

Wherein, (u _p,v _p) be the image coordinate of any point p in the plane of delineation, (x _p,y _p,0) be the path coordinate point P of image coordinate correspondence in road plane of p point; Q point (x above road plane _q, y _q, z _q) pedal line and road plane intersect at described P point, then x _q=x _p, y _q=y _p, wherein, (u _q, v _q) image coordinate of q point on the plane of delineation corresponding to the described Q point on road plane.

Further, described picture recognition module also for:

Identify the front truck license number in described front truck License Plate Image and the rear car license number in described rear car License Plate Image;

Also comprise in described weighing controller: comparison module and alarm appliance;

Described comparison module, for determining that whether described front truck license number is identical with described rear car license number;

Described Weighing control module, if also different with described rear car license number for described front truck license number, then control described alarm appliance and sends warning;

Described Weighing control module, if also identical with described rear car license number for described front truck license number, then controlling described image processing module performs described respectively using described front truck License Plate Image and described rear car License Plate Image as pending license plate image, according to the ordinate transform relation of the camera image plane set up in advance and road plane, determine the path coordinate (x in described pending license plate image, license plate centrally being put the pedal line CD of C and the intersection point D of road plane _d, y _d, 0), and according to described ordinate transform relation, determine the step described license plate centrally being put the height H between C and described intersection point D.

Further, also comprise in described weighing controller:

Memory module, for receive and store described first video camera send the first video image, receives and store described second video camera transmission the second video image.

The pinpoint True k Scale Weighing System of the vehicle based on license plate that the utility model provides, before treating by two video cameras being arranged on truck scale weighing table entrance and exit two ends vehicle of weighing, rear-mounted license plate carries out the image acquisition of same time, identify the license plate image in image at weighing controller after, carry out the coordinate transformation relation based on pre-set image plane and road plane, to the deterministic process of license plate on the plane of delineation of two dimension centrally carrying out path coordinate corresponding in three-dimensional road plane, thus, after obtaining the upper path coordinate centrally put on road plane of former and later two license plates respectively, can according to the physical location of path coordinate determination vehicle in weighing area of these two central points, such as vehicle is relative to a left side for truck scale weighing table, right, before, rear position, thus the accurate location of vehicle in weighing area can be realized based on to the analysis of license plate image, thus the accurate control to vehicle weighing can be ensured, ensure weighing results accurately and reliably.

Accompanying drawing explanation

Fig. 1 is that the utility model is based on the structural representation of vehicle license plate according to the True k Scale Weighing System embodiment one of location;

Fig. 2 is the transformational relation schematic diagram of the plane of delineation and road plane;

Fig. 3 is the principle schematic of window telescopic method;

Fig. 4 is that the utility model is based on the structural representation of vehicle license plate according to the True k Scale Weighing System embodiment two of location.

Embodiment

Fig. 1 is the utility model shines the True k Scale Weighing System embodiment one of location structural representation based on vehicle license plate, and as shown in Figure 1, this system comprises:

First video camera 1, second video camera 2 and truck scale 3, described truck scale 3 comprises weighing controller 31, weighing table 32 and LOAD CELLS 33;

Described first video camera 1 is positioned at described weighing table 32 endpiece predeterminable range place, described second video camera 2 is positioned at described weighing table 32 inlet end predeterminable range place, be respectively used to the collection described vehicle being carried out to the first image and the second image after described vehicle enters the weighing area of described weighing table 32, wherein, described first image and described second image are the image collected the same time;

Described weighing controller 31 comprises the picture recognition module 311, image processing module 312, position determination module 313 and the Weighing control module 314 that connect successively;

Described first video camera 1 is connected with described picture recognition module 311 respectively with described second video camera 2, and described LOAD CELLS 33 is connected with described Weighing control module 314;

Described picture recognition module 311, for obtaining the front truck License Plate Image in described first image of described first video camera 1 transmission, and obtains the rear car License Plate Image in described second image of described second video camera 2 transmission;

Described image processing module 312, for respectively using described front truck License Plate Image and described rear car License Plate Image as pending license plate image, according to the ordinate transform relation of the camera image plane set up in advance and road plane, determine the path coordinate (x in described pending license plate image, license plate centrally being put the pedal line CD of C and the intersection point D of road plane _d, y _d, 0), and according to described ordinate transform relation, determine the height H described license plate centrally put between C and described intersection point D;

Described image processing module 312, also for according to the path coordinate of described intersection point D and described height H, determine described on centrally to put the path coordinate of C on described road plane be (x _d, y _d, H);

Described position determination module 313, for centrally putting path coordinate on described road plane according to front license plate upper and the path coordinate on described road plane is centrally put in rear the upper of license plate, determines the position of described vehicle in described weighing area;

Described Weighing control module 314, if determine that described vehicle meets default weighing positions condition in the position of described weighing area for described position determination module 313, then controls described LOAD CELLS 33 and weighs to described vehicle.

In the present embodiment, in a distance, two ends of truck scale weighing table 32, be respectively arranged with video camera, these two video cameras are used for gathering the image of the front/rear side of its facing vehicle, wherein, the image of the front/rear license plate of vehicle is included in the image collected.Thus after when these two video cameras the image of the vehicle collected being sent to respectively the picture recognition module 311 in weighing controller, picture recognition module 311 can adopt multiple image-recognizing method in prior art, therefrom identifies license plate image.

What deserves to be explained is, although the riding position of the first video camera and the second video camera is in the both sides of truck scale weighing table, be understandable that, its riding position can not stop the current of vehicle.Such as can be placed on a L-type bar or on the bar of gantry.Concrete riding position is not specifically limited, and only need ensure that the front/rear license plate collected just shows it.

In the present embodiment, in order to realize the accurate location of the physical location of vehicle in weighing area, the strategy that have employed based on license plate graphical analysis carries out the location of vehicle physical location.

Only for the analysis to front truck License Plate Image in the present embodiment, similar with it to the analysis of rear License Plate Image, not repeat specification.

Specifically, for front truck License Plate Image, coordinate due to the pixel in the plane of delineation is two-dimensional coordinate, and the physical location of license plate on road plane is three-dimensional coordinate, directly can not obtain corresponding three-dimensional coordinate by two-dimensional coordinate, therefore, in the present embodiment, two main process are divided into for the determination of license plate in front truck License Plate Image centrally being put the path coordinate of C on road plane: first the first determines the path coordinate of the pedal line of this central point C and the intersection point D of road plane, now, z in the path coordinate of intersection point D is 0 to coordinate, namely the path coordinate of D point is (x _d, y _d, 0), it two is determine the height H between this central point C and intersection point D, thus the path coordinate of central point C on road plane is (x _d, y _d, H).And in these two processes, need to realize the conversion of image coordinate to path coordinate by the coordinate transformation relation between the plane of delineation and road plane, detailed coordinate transformation relation will elaborate below.

In the present embodiment, after obtaining the upper path coordinate centrally put on road plane of forward and backward license plate respectively, position determination module 313 just can determine the physical location of vehicle in weighing area.

Specifically, in most actual conditions, the license plate of vehicle hangs on the axis of vehicle, therefore, these two central points are equivalent on the axis of vehicle, be under the prerequisite of drawing the symmetrical region that fixed center line is axis of symmetry in advance in weighing area, position determination module 313 can by the path coordinate of these two central points, determine that the front and rear of vehicle is relative to the left-right deviation of drawing fixed center line, if deviation is larger, very possible vehicle flanging is then described, does not have complete on weighing table.On the other hand, position determination module 313 is parallel to the distance of lane line by calculating two central point path coordinates, the vehicle commander of vehicle can be known, further, according to the path coordinate of one of them central point, the position at headstock/tailstock place that this central point is corresponding can be known, thus, under the prerequisite that position and the vehicle commander at headstock/tailstock place determine, according to the length of weighing table, position determination module 313 can know that whether vehicle is complete on weighing table.Thus only meet in the physical location of vehicle and preset weighing positions condition, when namely meeting the condition of vehicle completely on weighing table, Weighing control module 314 just confirms that truck scale to vehicle weighing data effectively, thus obtains weighing results accurately.

Below to determining that the process of the path coordinate before in front truck License Plate Image, license plate centrally put on road plane is described in detail.

First, first introduce the transformational relation of the plane of delineation and the road plane adopted in the present embodiment, as shown in Figure 2, Fig. 2 is the transformational relation schematic diagram of the plane of delineation and road plane.If road surface is smooth, the coordinate under certain any alive boundary coordinate system, camera coordinate system and image coordinate system on road plane is expressed as (x, y, z), (x ^c, y ^c, z ^c) and (u, v).

The pedal line of camera coordinate system initial point and road plane intersect at o point, if this point is the initial point of video camera world coordinate system, definition road plane is the xoy plane of video camera world coordinate system.In fig. 2, the x direction of video camera world coordinate system is parallel to lane line direction, and y direction is perpendicular to lane line direction, and z direction perpendicular to road plane upwards, forms right hand rectangular coordinate system.To try to achieve in monocular-camera arbitrfary point (u, v) and the world coordinates of road plane on image by constraint condition and fasten the corresponding relation of point (x, y, z).

If $A=\left[\begin{array}{ccc}{f}_u& 0& u_0\\ 0& f_v& v_0\\ 0& 0& 1\end{array}\right]$ For intrinsic parameters of the camera matrix, R and T is video camera external parameter, wherein, $R=\left[\begin{array}{ccc}{r}_{11}& r_{12}& r_{13}\\ r_{21}& r_{22}& r_{23}\\ r_{31}& r_{32}& r_{33}\end{array}\right]$ For rotation matrix, $T=\left[\begin{array}{c}{t}_x\\ t_y\\ t_z\end{array}\right]$ For translation vector, wherein: f _uand f _vfor focal length of camera, u ₀and v ₀for centre coordinate and the initial point of the plane of delineation, r ₁₁... r ₃₃for the component of R.According to following pin-hole imaging model, there is following relation:

$\left[\begin{array}{c}u-u_0\\ v-v_0\end{array}\right]=\left[\begin{array}{c}{f}_u\frac{x^c}{z^c}\\ -f_v\frac{y^c}{z^c}\end{array}\right],\left[\begin{array}{c}{x}^c\\ y^c\\ z^c\end{array}\right]=R\left(\begin{array}{c}\left[\begin{array}{c}x\\ y\\ z\end{array}\right]-T\end{array}\right)$

In Fig. 2, p point (u on the plane of delineation _p, v _p) back projection to camera coordinate system, and intersects at P point (x with road plane _p, y _p, z _p), then z _p=0.According to above-mentioned relation formula cancellation x ^cand y ^c, then p point coordinate in known image plane, to ask on road plane the equation in coordinates of corresponding P point to be:

$z^c\left[\begin{array}{c}\frac{u_p-u_0}{f_u}\\ -\frac{v_p-v_0}{f_v}\\ 1\end{array}\right]=R\left(\begin{array}{c}\left[\begin{array}{c}{x}_P\\ y_P\\ 0\end{array}\right]\end{array},-,T\right)$

Arrange above formula and obtain following ordinate transform formula (1), thus the x of P point coordinate can be solved _pand y _p.

$\left[\begin{array}{ccc}{r}_{11}& r_{12}& -\frac{u_p-u_0}{f_u}\\ r_{21}& r_{22}& \frac{v_p-v_0}{f_v}\\ r_{31}& r_{32}& -1\end{array}\right]\left[\begin{array}{c}{x}_P\\ y_P\\ z^c\end{array}\right]=\left[\begin{array}{ccc}{r}_{11}& r_{12}& r_{13}\\ r_{21}& r_{22}& r_{23}\\ r_{31}& r_{32}& r_{33}\end{array}\right]\left[\begin{array}{c}{t}_x\\ t_y\\ t_z\end{array}\right]---\left(1\right)$

And then, if Q point (x above road plane _q, y _q, z _q) pedal line and road plane intersect at above-mentioned P point, then x _q=x _p, y _q=y _pif Q point corresponds to the q point (u on the plane of delineation _q, v _q), as shown in Figure 2.According to above-mentioned ordinate transform formula (1) cancellation x ^cand y ^c, p and the q point in known image plane, ask Q point on road plane relative to the height equation of P point to be:

$z^c\left[\begin{array}{c}\frac{u_q-u_0}{f_u}\\ -\frac{v_q-v_0}{f_v}\\ 1\end{array}\right]=R\left(\begin{array}{c}\left[\begin{array}{c}{x}_P\\ y_P\\ z_Q\end{array}\right]\end{array}-T\right)$

Arrange above formula height equation and obtain following ordinate transform formula (2), simultaneously according to formula (1), the z of Q point coordinate on road plane can be solved _q.

$\left[\begin{array}{cc}-r_{13}& \frac{u_q-u_0}{f_u}\\ -r_{23}& -\frac{v_q-v_0}{f_v}\\ -r_{33}& 1\end{array}\right]\left[\begin{array}{c}{z}_Q-t_z\\ z^c\end{array}\right]=\left[\begin{array}{cc}{r}_{11}& r_{12}\\ r_{21}& r_{22}\\ r_{31}& r_{32}\end{array}\right]\left[\begin{array}{c}{x}_P-t_x\\ y_P-t_y\end{array}\right]---\left(2\right)$

Based on the ordinate transform relation of above-mentioned camera image plane and road plane, the process realizing determining before in front truck License Plate Image, license plate centrally to be put the path coordinate of C on road plane is specific as follows, the present embodiment only former license plate is that example is described, same mode is applicable to the situation of rear license plate, not repeat specification.

First, under brief description, the present embodiment is based on the determination of license plate centrally being put the path coordinate of C on road plane, realizes the pinpoint ultimate principle of vehicle.

According to the analysis of license plate image, calculate width and the height of license plate, and in conjunction with license plate normal width and constraint condition highly, and the constraint condition that license plate plane is vertical with road plane, set up license plate model.By calculating the path coordinate of pedal line and the road plane joining that license plate is centrally put, based on the path coordinate of this intersection point, obtain the upper of license plate and centrally put the path coordinate on road plane, thus realize by license plate position coordinates, realize the accurate location to vehicle.

Specifically, as shown in Figure 3, if license plate plane is vertical with road plane, make AB be the intersecting lens of license plate plane and road plane, and to be parallel on license plate transversely horizontal line.If license plate vertical projection to AB line is MN line segment, therefore, license plate is centrally put the pedal line CD to road plane joining D of C also on MN line segment.Adopt dilating window method to estimate the path coordinate of D point in the present embodiment, thus obtain the physical location of C point on road plane, realize the accurate location to vehicle.What deserves to be explained is, during practical operation, need fix on the transverse horizontal line on edge on license plate, top of dilating window, the bottom of this window of free-extension.The process of dilating window bottom is mainly in order to determine the true intersection position of license plate plane and road plane, so-called true intersection position, can be understood as is when bottom is on true intersection, the height of the license plate now calculated and width meet the requirement of flexible cut-off condition, namely meet the requirement of license plate model.

Particularly, described image processing module 312 adopts dilating window method, on the transverse horizontal line on edge on license plate, top of fixing dilating window, the bottom of flexible described dilating window, suppose that bottom is line segment MN, determine that the present image coordinate of two end points in bottom of described dilating window is respectively (u _m,v _m) and (u _n,v _n).Now, the present image coordinate (u of these two end points M and N can be obtained respectively according to above-mentioned ordinate transform formula (1) _m,v _m) and (u _n,v _n) path coordinate (x of correspondence on road plane _m,y _m,0) and (x _n,y _n,0) distance, obtained between M and N point is also referred to as the width of license plate self.

And then because above-mentioned intersection point D is the mid point of line segment MN, therefore, the path coordinate of D point is (x _d, y _d, 0) and=((x _m+ x _n)/2, (y _m+ y _n)/2,0), image processing module 312 can calculate by ordinate transform formula (2) height H between pedal line and intersection point D license plate centrally being put C, namely license plate is centrally put height (being called for short: license plate setting height(from bottom)) overhead.Equally, it is H' that image processing module 312 also can calculate the height centrally put between the pedal line of E and intersection point D point under license plate by formula (2), namely centrally put height overhead under license plate, thus obtain the path coordinate (x of E point _e, y _e, z _e)=(x _d, y _d, H').Therefore, the height of license plate self is h=H-H'.What deserves to be explained is, height overhead is centrally put at this place under can calculating license plate, thus can obtain the height of license plate, and this highly may be used in other embodiment, and the present embodiment does not explain.

Afterwards, image processing module 312, according to the following flexible cut-off condition pre-set, determines whether w and H now calculated satisfies condition:

min f(u,v)＝(1-a)(|w-w ₀|/w ₀)+a(|H-H ₀|/H ₀)

S.t. the width of license plate self: w ∈ [42cm, 46cm]

Front license plate setting height(from bottom): H ∈ [42cm, 62cm]

Rear license plate setting height(from bottom): H ∈ [68cm, 108cm]

Wherein, (u, v) refers to the image coordinate variable of two end points M and N in bottom of dilating window; w ₀for license plate normal width; The distributed density values of a corresponding to H described in license plate setting height(from bottom); H ₀in distribution density for license plate setting height(from bottom), the license plate setting height(from bottom) corresponding to maximum distribution density points.The distribution density function of this license plate setting height(from bottom) rule of thumb can be added up in advance and obtain.

In fact, automotive number plate contour dimension makes according to GA36-2007 standard-required: before and after medium-sized and following automobile, before license plate, full-sized car, the contour dimension of license plate is 44cm*14cm, and after full-sized car, the contour dimension of license plate, photograph of hanging number plate is 44cm*22cm.Therefore, the judgement interval of above-mentioned width and height is determined with reference to this standard.

What deserves to be explained is, in above-mentioned flexible cut-off condition, if be front truck License Plate Image when pre-treatment, then before, license plate setting height(from bottom) is as the judgement reference of H, if when pre-treatment is rear car License Plate Image, then after, license plate setting height(from bottom) is as the judgement reference of H.

If meet, then determine the path coordinate (x of described intersection point D _d, y _d, 0) be: (x _d, y _d, 0) and=((x _m+ x _n)/2, (y _m+ y _n)/2,0), and determine that it is (x that license plate centrally puts the path coordinate of C on road plane _c, y _c, z _c)=(x _d, y _d, H).

What deserves to be explained is, if above-mentioned w and H calculated in the position at current place, the bottom of dilating window does not meet above-mentioned flexible cut-off condition, then image processing module 312 need continue the bottom of this window flexible, until find the position of the MN line segment meeting above-mentioned flexible cut-off condition.

In the present embodiment, before treating by two video cameras being arranged on truck scale two ends vehicle of weighing, the license plate that rear side is hung carries out the image acquisition of same time, after picture recognition module in weighing controller identifies the license plate image in image, image processing module carries out the coordinate transformation relation based on pre-set image plane and road plane, to the deterministic process of license plate on the plane of delineation of two dimension centrally carrying out path coordinate corresponding in three-dimensional road plane, thus, after obtaining the upper path coordinate centrally put on road plane of former and later two license plates respectively, position determination module can according to the physical location of path coordinate determination vehicle in weighing area of these two central points, such as vehicle is relative to a left side for truck scale weighing table, right, before, rear position, thus the accurate location of vehicle in weighing area can be realized based on to the analysis of license plate image, thus Weighing control module can ensure the accurate control to vehicle weighing, ensure weighing results accurately and reliably.

Fig. 4 is the utility model based on the structural representation of vehicle license plate according to the True k Scale Weighing System embodiment two of location, as shown in Figure 4, on basis embodiment illustrated in fig. 1, described picture recognition module 311 also for:

Identify the front truck license number in described front truck License Plate Image and the rear car license number in described rear car License Plate Image;

Also comprise in described weighing controller 31: comparison module 315 and alarm appliance 316;

Described comparison module 315, for determining that whether described front truck license number is identical with described rear car license number;

Described Weighing control module 314, if also different with described rear car license number for described front truck license number, then control described alarm appliance 316 and sends warning;

Described Weighing control module 314, if also identical with described rear car license number for described front truck license number, then controlling described image processing module 312 performs described respectively using described front truck License Plate Image and described rear car License Plate Image as pending license plate image, according to the ordinate transform relation of the camera image plane set up in advance and road plane, determine the path coordinate (x in described pending license plate image, license plate centrally being put the pedal line CD of C and the intersection point D of road plane _d, y _d, 0), and according to described ordinate transform relation, determine the step described license plate centrally being put the height H between C and described intersection point D.

Further, also comprise in described weighing controller 31:

Memory module 317, for receive and store described first video camera 1 send the first video image, receives and store described second video camera 2 transmission the second video image.

In the present embodiment, by judging the consistance of front and back license plate number, can effectively prevent from blocking license plate or deck.

In addition, in the present embodiment, in order to effectively prevent other Stakeholders to the interference of vehicle weighing, the first video camera and the second video camera can be made to take video image around truck scale continuously.

In fact, detector can also be set on weighing table, this detector can be magnetic induction loop, after detector detects that vehicle enters weighing area, namely trigger the process that the first video camera and the second video camera are taken pictures continuously simultaneously, thus obtain the first video image and the second video image.Be understandable that, based on this, above-mentioned first image and the second image can be a frame of corresponding same time point in the first video image and the second video image respectively.Only have when in the multiframe correspondence image of weighing controller at the first video image and the second video image, when determining the picture frame meeting above-mentioned default weighing positions condition, just control to weigh to vehicle.

Last it is noted that above each embodiment is only in order to illustrate the technical solution of the utility model, be not intended to limit; Although be described in detail the utility model with reference to foregoing embodiments, those of ordinary skill in the art is to be understood that: it still can be modified to the technical scheme described in foregoing embodiments, or carries out equivalent replacement to wherein some or all of technical characteristic; And these amendments or replacement, do not make the essence of appropriate technical solution depart from the scope of each embodiment technical scheme of the utility model.

Claims (3)

1., based on the pinpoint True k Scale Weighing System of vehicle of license plate, it is characterized in that, comprising:

First video camera, the second video camera and truck scale, described truck scale comprises weighing controller, weighing table and LOAD CELLS;

Described first position for video camera is in described weighing table endpiece predeterminable range place, described second position for video camera is in described weighing table inlet end predeterminable range place, be respectively used to the collection described vehicle being carried out to the first image and the second image after described vehicle enters the weighing area of described weighing table, wherein, described first image and described second image are the image collected the same time;

Described weighing controller comprises the picture recognition module, image processing module, position determination module and the Weighing control module that connect successively;

Described first video camera is connected with described picture recognition module respectively with described second video camera, described LOAD CELLS and described Weighing control model calling;

Described picture recognition module, for obtaining the front truck License Plate Image in described first image of described first video camera transmission, and obtains the rear car License Plate Image in described second image of described second video camera transmission;

Described image processing module, for respectively using described front truck License Plate Image and described rear car License Plate Image as pending license plate image, according to the ordinate transform relation of the camera image plane set up in advance and road plane, determine the path coordinate (x in described pending license plate image, license plate centrally being put the pedal line CD of C and the intersection point D of road plane _d, y _d, 0), and according to described ordinate transform relation, determine the height H described license plate centrally put between C and described intersection point D;

Described image processing module, also for according to the path coordinate of described intersection point D and described height H, determine described on centrally to put the path coordinate of C on described road plane be (x _d, y _d, H);

Described position determination module, for centrally putting path coordinate on described road plane according to front license plate upper and the path coordinate on described road plane is centrally put in rear the upper of license plate, determines the position of described vehicle in described weighing area;

Described Weighing control module, if determine that described vehicle meets default weighing positions condition in the position of described weighing area for described position determination module, then controls described LOAD CELLS and weighs to described vehicle.

2. system according to claim 1, is characterized in that, described picture recognition module also for:

Identify the front truck license number in described front truck License Plate Image and the rear car license number in described rear car License Plate Image;

Also comprise in described weighing controller: comparison module and alarm appliance;

Described comparison module, for determining that whether described front truck license number is identical with described rear car license number;

Described Weighing control module, if also different with described rear car license number for described front truck license number, then control described alarm appliance and sends warning;

Described Weighing control module, if also identical with described rear car license number for described front truck license number, then controlling described image processing module performs described respectively using described front truck License Plate Image and described rear car License Plate Image as pending license plate image, according to the ordinate transform relation of the camera image plane set up in advance and road plane, determine the path coordinate (x in described pending license plate image, license plate centrally being put the pedal line CD of C and the intersection point D of road plane _d, y _d, 0), and according to described ordinate transform relation, determine the step described license plate centrally being put the height H between C and described intersection point D.

3. system according to claim 1, is characterized in that, also comprises in described weighing controller:

Memory module, for receive and store described first video camera send the first video image, receives and store described second video camera transmission the second video image.