CN100559420C - Parking guidance system based on computer vision - Google Patents

Abstract

A kind of parking guidance system based on computer vision, the parking guidance controller that comprises the parking stall recognition device that is installed in each parking lot, parking space information statistics distributing device and be used to receive and handle the parking space information in each parking lot, the parking guidance controller comprises parking space information receiver module, parking space information coding module, the parking space information that is used for gathering converts numerical coding to, described numerical coding is 28 figure places, the geographical location information in preceding 22 bit representation parking lots, the concrete parking space information of back 6 bit representations in this parking lot; And berth, parking lot information processing module, be used for each parking stall in each parking lot is produced a record, set up the inquiry inlet of each record, and the corresponding numerical coding of each record is outwards issued.The present invention can collect each parking lot information, and can induce, be convenient to the utilization factor that the driver can in time find the parking stall, increase the parking lot outside the venue.

Description

Parking guidance system based on computer vision

(1) technical field

The invention belongs to omnidirectional computer vision sensor technology, image recognition technology, database technology, urban road digital coding and the application of the network communications technology aspect parking guidance system, especially a kind of parking guidance device based on computer vision.

(2) background technology

Urban highway traffic is made up of dynamic traffic and static traffic two parts, and static traffic is meant that vehicle is to finish different trip purposes or take care of the state of parking in zones of different, different standages that produces.Static traffic is the same with dynamic traffic, is indivisible ingredient in the urban transportation, and dynamic traffic is starting point with the static traffic, and static traffic is the continuity of dynamic traffic.Dynamic traffic and static traffic had not only been mutually promoted but also mutual restriction, needed that coordinated development is common to constitute the urban transportation system.

During present urban traffic management, people often only pay attention to dredging and controlling urban dynamic traffic, planning, construction and management have been ignored to static traffics such as vehicle parkings, ignored inducing to vehicle parking, make hard nut to cracks such as urban traffic congestion, obstruction, accident take place frequently more apparent outstanding, people take a lot of control measures unable to do what one wishes painstakingly, produce little effect.The construction deficiency that a main crux that causes this problem is the urban parking area storehouse, mismanagement and the parking guidance system that lacks the advanced person, caused the time of blindly flowing, made parking offense and phenomenons such as taking road happen occasionally in order to seek the parking stall.

Enquiry data shows both at home and abroad, the driver is not owing to understand the parking stall, arbitrarily, arbitrarily, do not have the destination to go for the parking stall, increased extra burden to road traffic, the vehicle in the urban pavement wagon flow about nearly 12^15% is a vehicle of seeking parking position.Other has abroad and reports, for the gasoline of looking for the parking stall to expend in the urban district, accounts for 40% of whole driving gasoline in Paris, has increased the discharging of Vehicular exhaust, has increased considerably the environmental pollution that vehicle exhaust causes.

China's one side parking lot quantity can not adapt to the growth of automobile pollution at present, add the driver and blindly seek the parking lot, caused breaking rules and regulations to account for road parking phenomenon and increased, thereby greatly reduced road passage capability, cause traffic congestion and traffic hazard easily, this contradiction showed outstanding in positive day.

The storehouse, parking lot is the most important condition of urban static traffic, and the construction of this respect and the deficiency of management directly have influence on the normal operation of dynamic traffic, and dynamic traffic is not smooth, conversely the management of static traffic is exerted an influence again simultaneously.Produced the vicious cycle of traffic difficulties, make urban highway traffic safe, unimpeded, can't be protected in order, just seriously hamper city and expanding economy thereof.

At present China exists simultaneously that ratio differs greatly and the low problem of parking lot utilization factor between urban automobile volume and parking facility, shows as the parking stall on the one hand in the parking lot in and leaves unused the wasting of resources; On the other hand, therefore a large amount of nonlocal vehicles and the local vehicle of part spend the parking lot that driver's long time is sought parking position because the driver does not understand parking lot parking position situation.This has not only increased the urban road load, has a strong impact on the dynamic traffic of road, has increased considerably the environmental pollution that vehicle exhaust causes.External actual conditions show to have only by effective parking guidance infosystem and just can improve the situation that automobile is blindly sought the parking lot, and then reduce traffic hazard and reduce air pollution.Advanced parking guidance and infosystem key theory and technical application research are the important research contents of intelligent transportation, also are one of advanced subject of furtheing investigate in the world, are the key issues that China's urban transportation needs to be resolved hurrily.

The approach that solves parking problem mainly can be set about from following three aspects:

(1) makes rational planning for and develop static traffic infrastructure;

(2) adopt the advanced management means that parking is managed and controls;

(3) implement intelligent transport system.

Static traffic infrastructure mainly comprise the peripheral gateway in social parking lot, urban district Public Parking, join the detecting device, variable information display device, the network equipment etc. building the parking lot and parking is carried out the required outfit of scientific management.

Advanced parking guidance infosystem will utilize the unified variable electronic sign board in area for the driver provides different dynamic parking guidance information, comprises information such as parking position that position, parking lot, position, garage parking, curb parking position, driver select in advance and best travel route.Utilize these information, the guiding driver avoids crowded and finds comparatively desirable parking position fast.

Therefore advanced parking guidance infosystem must be made up of four parts such as parking information collection, information processing, information transmission and information issues, the effect of its each several part is as follows: 1) information acquisition system, system is by Long-Range Surveillance Unit, sensing device, each parking lot relevant information in the acquisition target zone has mainly comprised the information such as parking stall behaviour in service in parking lot; 2) information handling system, whether system becomes the information of the appropriate format that provides to the driver with the parking lot behaviour in service that collects and peripheral path information processing, block up as full sky (residue parking stall situation), the collector distributor road in parking lot etc.In addition, information handling system is also being undertaken the tasks such as changing pattern of the parking lot information of storing, processing processing parking lot operating position.These functions will lay the foundation for services such as the forecast of parking demand situation, parking stall reservation will be provided future; 3) information transmission system, the basic task of information transmission are guarantee from the information acquisition system to the information handling system again information issuing system unimpeded.Its form commonly used has forms such as Optical Transmission Network OTN, switched telephone network and optical access network; 4) information issuing system, the task of system are the information that information handling system was handled, and divide several levels to issue out to the external world by rights.Normally by control center, at any time the behaviour in service with each parking lot provides to the driver in the mode of vision or by the mode of broadcasting with the sense of hearing on the variable information display board, also can utilize modes such as internet, mobile phone and on-vehicle navigation apparatus to issue.Basis, issue form the most commonly used are the induction information plate that is arranged at trackside the most at present.

The ground induction coil that will detect parking stall behaviour in service in the parking lot in the prior art and be by being embedded in before the parking stall obtains whether occupied information of this parking stall by way of electromagnetic induction, though this mode can detect the behaviour in service of parking stall better, but when existing locality sense coil fault, the use that need stop the parking stall excavating pavement maintenance, increased maintenance personal's maintenance workload, increased maintenance cost, it is bigger to bury the ground induction coil one-time investment underground to each parking stall to the bigger parking lot of capacity simultaneously, and the many more complicacy of detection and the pressure of communication and calculating of also can bringing in parking stall.Problems such as the another kind of method that detects the parking stall is to adopt ultrasonic detector to obtain whether occupied information of this parking stall, and this mode exists also that one-time investment is bigger, the pressure of the complicacy of detection and communication and calculating is big.Processing by video image also can obtain whether occupied information of parking stall.

(3) summary of the invention

For the information that is confined to single parking lot that overcomes existing parking guidance device, can not induce outside the venue, deficiency that the parking lot utilization factor is not high, the invention provides a kind of information that can collect each parking lot, and can induce, be convenient to the parking guidance device that the driver can in time find the parking stall, increase the utilization factor in parking lot outside the venue based on computer vision.

The technical solution adopted for the present invention to solve the technical problems is:

A kind of parking guidance system based on computer vision, comprise the parking stall recognition device, the parking space information statistics distributing device that are installed in each parking lot, described parking guidance system also comprises the parking guidance controller of the parking space information that is used to receive and handle each parking lot, described parking guidance controller comprises: the parking space information receiver module is used to gather the parking space information in each parking lot; The parking space information coding module, the parking space information that is used for gathering converts numerical coding to, and described numerical coding is 28 figure places, the geographical location information in preceding 22 bit representation parking lots, the concrete parking space information of back 6 bit representations in this parking lot; In geographical location information, the absolute position in the most preceding 6 these cities of bit representation, then the road at place, 11 bit representation parking lot is from intown relative position, followed by the distance of 5 bit representation parking lots from the road starting end at place to the parking lot; In concrete parking space information, the floor attribute in the 1st definition parking lot, the position attribute in the 2nd definition parking lot, the parking stall attribute in the 3rd～5 definition parking lot, can the park cars attribute of size of this parking stall in the 6th definition parking lot; Berth, parking lot information processing module is used for each parking stall in each parking lot is produced a record, sets up the inquiry inlet of each record, and the corresponding numerical coding of each record is outwards issued.

As preferred a kind of scheme: in described parking space information coding module, represent that the 4th to the 6th bit representation latitude for digitally coded the 1st to the 3rd bit representation longitude with angle, represent with angle; With significant position, city center is initial point, East and West direction is the x axle, the north-south is the y axle, the city is divided into A, B, 4 quadrant districts of C, D, the quadrant district at the starting point place of the 7th bit representation road represents x, the y coordinate at two ends, street, road place with 4 bit digital, and x is 2 bit digital, y is 2 bit digital, and is unit with km; The 8th x coordinate to the 9th bit representation road starting point, the 10th y coordinate to the 11st bit representation street starting point, the quadrant district at the terminal point place of the 12nd bit representation road, the 13rd x coordinate to the 14th bit representation street terminal point, the 15th y coordinate to the 16th bit representation street terminal point, at a distance of in the 1km scope, the 17th figure place is distinguished with the English alphabet order as coordinate points; 5 of the 18th to the 22nd are encoded to the Native digits coding, from the ascending numbering of the origin-to-destination of road, begin from nearest intown crossing to compile for the side street lane, for road both sides street crossing is arranged, layout is carried out at single right two ends that extend to, a left side, increases 1 every 100m on the 21st, for have only one-sided runway if adopt the odd number layout at the left of road, in the right-hand employing even numbers layout of road, the 22nd bit representation both sides layout, left side layout or right side layout.

Further, described parking stall recognition device comprises the omnibearing vision sensor of overlooking parking position that is installed in the parking lot center of top, is used for the microprocessor according to the data identification parking stall of omnibearing vision sensor, and described omnibearing vision sensor connects microprocessor; Described omnibearing vision sensor comprises in order to the evagination catadioptric minute surface of object in the field in the reflection parking lot, in order to prevent dark circles cone, the transparent cylinder that anaclasis and light are saturated and to be used to take the camera of imaging body on the evagination mirror surface, described evagination catadioptric minute surface is positioned at the top of transparent cylinder, evagination catadioptric minute surface down, the dark circles cone is fixed on the bottom center of evagination catadioptric minute surface, and described camera facing to evagination catadioptric minute surface up;

Described microprocessor also comprises:

The view data read module is used to read the video image information of coming from the omnibearing vision sensor biography;

The image data file memory module, the video image information that is used for reading into is kept at storage unit by file mode;

The omnibearing vision sensor demarcating module is used to set up the material picture in space and the corresponding relation of the video image that is obtained;

Parking stall customization and mapping relations are set up module, are used to customize the parking stall frame on parking lot planimetric map and the comprehensive Video Detection figure, are used to set up the mapping relations of the parking stall that customizes on parking lot planimetric map and the comprehensive Video Detection figure;

The color space conversion module is used to finish the conversion of image rgb color space to the YCrCb color space, and conversion formula (17) provides:

Y＝0.29990*R+0.5870*G+0.1140*B

Cr＝0.5000*R-0.4187*G-0.0813*B+128 (17)

Cb＝-0.1787*R-0.3313*G+0.5000*B+128

Image pre-service and parking stall measure module, whether have car, adopt the background subtracting method if being used for detecting the parking stall frame, the computing formula of described background subtracting as the formula (18),

f _d(X，t ₀，t _i)＝f(X，t _i)-f(X，t ₀) (18)

In the formula: f _d(X, t ₀, t _i) be to photograph the colourity result who carries out image subtraction between panoramic picture and reference panoramic picture, f (X, t in real time _i) be to photograph image chroma value, f (X, t in real time ₀) be reference image chroma reference value;

And judge whether two components of image C r, Cb after subtracting each other surpass the preset threshold value scope, if surpass the threshold value of defined, then park in this parking stall of mark, otherwise the parking stall do not take, judgment formula as the formula (19):

$\left\{\begin{array}{c}1,\mathrm{if}(\mathrm{abs}(\mathrm{Cr}-\mathrm{ThresholdCr})>\mathrm{<figure-callout\; id="1"\; label="threshold"\; filenames="C20071006784500311.png"\; state="\{\{state\}\}">threshold</figure-callout>}1||\mathrm{abs}(\mathrm{Cb}-\mathrm{ThresholdCb})>\mathrm{threshold}2)\\ 0,\mathrm{else}\end{array}\right.---\left(19\right)$

In the formula: Cr represents the Cr mean value in the frame of detected current parking stall, Cb represents the Cb mean value in the frame of detected current parking stall, ThresholdCr represents the Cr mean value on ground in the frame of defined parking stall, ThresholdCb represents the Cb mean value on ground in the frame of defined parking stall, threshold1 represents the threshold range of Cr, and threshold2 represents the threshold range of Cb.

Or: described parking stall recognition device comprises the omnibearing vision sensor of overlooking parking position that is installed in the parking lot center of top, is used for the microprocessor according to the data identification parking stall of omnibearing vision sensor, and described omnibearing vision sensor connects microprocessor;

Described omnibearing vision sensor comprises in order to the evagination catadioptric minute surface of object in the field in the reflection parking lot, in order to prevent dark circles cone, the transparent cylinder that anaclasis and light are saturated and to be used to take the camera of imaging body on the evagination mirror surface, described evagination catadioptric minute surface is positioned at the top of transparent cylinder, evagination catadioptric minute surface down, the dark circles cone is fixed on the bottom center of evagination catadioptric minute surface, and described camera facing to evagination catadioptric minute surface up;

Described microprocessor also comprises:

The view data read module is used to read the video image information of coming from the omnibearing vision sensor biography;

The image data file memory module, the video image information that is used for reading into is kept at storage unit by file mode;

The omnibearing vision sensor demarcating module is used to set up the material picture in space and the corresponding relation of the video image that is obtained;

Parking stall customization and mapping relations are set up module, are used to customize the parking stall frame on parking lot planimetric map and the comprehensive Video Detection figure, are used to set up the mapping relations of the parking stall that customizes on parking lot planimetric map and the comprehensive Video Detection figure;

Rim detection parking stall module is used to adopt the rim detection parking stall whether car is arranged, and adopts Suo Beier (Sobel) operator as edge detection algorithm, and the Sobel operator adopts the template of 3*3 size, and the Sobel operator calculates partial derivative with following formula:

S _x＝(a ₂+ca ₃+a ₄)-(a ₀+ca ₇+a ₆)

S _y＝(a ₀+ca ₁+a ₂)-(a ₆+ca ₅+a ₄) (20)

Constant c is 2 in the formula, and the Sobel operator can be realized with following convolution template:

$S_x=\left[\begin{array}{ccc}-1& 0& 1\\ -2& 0& 2\\ -1& 0& 1\end{array}\right]$ $S_y=\left[\begin{array}{ccc}1& 2& 1\\ 0& 0& 0\\ -1& -2& -1\end{array}\right]---\left(21\right)$

Closed curve with direction chain representation scenery edge obtains parking stall image to be detected;

The vehicle image template of target setting object is T, size is M*N, and the image that parks cars in the parking stall to be detected is I, and size is L*W, managing a template T is superimposed upon on the image I, carry out images match and handle, compare the difference of the subimage of the I under T and its covering, if difference is less than preset threshold value, think that then T has coupling preferably at the subimage of this place and I, judging tentatively has car, greater than preset threshold value, judges that the parking stall does not take as difference.

Further, in the module of described rim detection parking stall, after judgement tentatively has car, turn to the image of M*N for spatial spreading, if i, j is each pixel coordinate in the parking stall scene of discretize, and (i j) is coordinate (i to f, the gray-scale value of pixel j), " central moment " of object is approached by following dual summation form in the image scene, with formula (23) expression

$D_{\mathrm{pq}}=\underset{i\&Element;R}{\mathrm{\&Sigma;}}\underset{j\&Element;R}{\mathrm{\&Sigma;}}{(i-\stackrel{\&OverBar;}{i})}^p{(j-\stackrel{\&OverBar;}{j})}^{q}f(i,j)---\left(23\right)$

Wherein: i=m ₁₀/ m ₀₀J=m ₀₁/ m ₀₀ $m_{00}=\underset{i\&Element;R}{\mathrm{\&Sigma;}}\underset{j\&Element;R}{\mathrm{\&Sigma;}}f(i,j);$ $m_{10}=\underset{i\&Element;R}{\mathrm{\&Sigma;}}\underset{j\&Element;R}{\mathrm{\&Sigma;}}i\&times;f(i,j);$ $m_{01}=\underset{i\&Element;R}{\mathrm{\&Sigma;}}\underset{j\&Element;R}{\mathrm{\&Sigma;}}j\&times;f(i,j);$

Carry out normalized, can get normalization permanent center square and be:

η _pq＝D _pq/D ₀₀，r＝(p+q)/2+1，

If p+q＜2 are derived two constant central moment functions of RST and are represented with formula (24),

The object of parking scene and the center invariant moments of template are compared, less than predetermined threshold value, judge that there is car this parking stall, greater than default template threshold value, judge that the parking stall does not take as the phase difference as the phase difference.

Further again, in the module of described rim detection parking stall, the centre of form of object in the scene of parking stall is aimed at the centre of form of template, utilize the rotational transform of coordinate to determine the anglec of rotation of object correspondence in the scene, the convergent-divergent algorithm:

x ₂＝N(x ₁-x ₀)+x ₀ (25)

y ₂＝N(y ₁-y ₀)+y ₀

The rotation algorithm:

x ₂＝(x ₁-x ₀)cosθ+(y ₁-y ₀)sinθ+x ₀ (26)

y ₂＝(y ₁-y ₀)cosθ+(x ₁-x ₀)sinθ+y ₀

In the formula, (x2 y2) is the new coordinate of pixel; (x1 y1) is former coordinate; (x0 y0) is the centre of form coordinate of object; N is a zoom factor; θ is the angle of rotation.

Or be: described parking stall recognition device comprises the omnibearing vision sensor of overlooking parking position that is installed in the parking lot center of top, is used for the microprocessor according to the data identification parking stall of omnibearing vision sensor, and described omnibearing vision sensor connects microprocessor;

Described omnibearing vision sensor comprises in order to the evagination catadioptric minute surface of object in the field in the reflection parking lot, in order to prevent dark circles cone, the transparent cylinder that anaclasis and light are saturated and to be used to take the camera of imaging body on the evagination mirror surface, described evagination catadioptric minute surface is positioned at the top of transparent cylinder, evagination catadioptric minute surface down, the dark circles cone is fixed on the bottom center of evagination catadioptric minute surface, and described camera facing to evagination catadioptric minute surface up;

Described microprocessor also comprises:

The view data read module is used to read the video image information of coming from the omnibearing vision sensor biography;

The image data file memory module, the video image information that is used for reading into is kept at storage unit by file mode;

The omnibearing vision sensor demarcating module is used to set up the material picture in space and the corresponding relation of the video image that is obtained;

Parking stall customization and mapping relations are set up module, are used to customize the parking stall frame on parking lot planimetric map and the comprehensive Video Detection figure, are used to set up the mapping relations of the parking stall that customizes on parking lot planimetric map and the comprehensive Video Detection figure;

The parking stall measure module is used for judging the similarity of this figure and rectangle after detecting edge contour point on the parking stall and link and closed curve interior pixel and filling, if the similarity of calculating in threshold range, just being judged to be is vehicle.

Further, set up in the module in customization of described parking stall and mapping relations, the parking stall frame on the comprehensive Video Detection figure of described customization is four points determining the parking stall by input equipment, defines with the straight line ways of connecting then and detects the parking stall.

Described catadioptric minute surface designs in order to following method: with the camera projection centre is that true origin is set up coordinate system, and the face shape of catoptron is used z (X) function representation; The pixel q of distance images central point ρ has accepted from horizontal scene O point (apart from Z axle d), at the light of mirror M point reflection in as the plane; Horizontal scene is undistorted to require the coordinate of the horizontal coordinate of scene object point and corresponding picture point linear;

d(ρ)＝αρ (1)

ρ is and the distance of the face shape central point of catoptron in the formula (1), and α is the magnification of imaging system;

If the normal that catoptron is ordered at M and the angle of Z axle are γ, the angle of incident ray and Z axle is Φ, and the angle of reflection ray and Z axle is θ, then

$\mathrm{tg}\left(x\right)=\frac{d\left(x\right)-x}{z\left(x\right)-h}---\left(2\right)$

$\mathrm{tg\&gamma;}=\frac{\mathrm{dz}\left(x\right)}{\mathrm{dx}}---\left(3\right)$

$\mathrm{tg}\left(2\mathrm{\&gamma;}\right)=\frac{2\frac{\mathrm{dz}\left(x\right)}{\mathrm{dx}}}{1-\frac{d^{2}z\left(x\right)}{{\mathrm{dx}}^2}}---\left(4\right)$

By reflection law:

2γ＝φ-θ (6)

$\stackrel{\&CenterDot;}{\&CenterDot;\&CenterDot;}\mathrm{tg}\left(2\mathrm{\&gamma;}\right)=\mathrm{tg}(\mathrm{\&phi;}-\mathrm{\&theta;})=\frac{\mathrm{tg\&phi;}-\mathrm{tg\&theta;}}{1+\mathrm{tg\&phi;tg\&theta;}}$

Obtain the differential equation (7) by formula (2), (4), (5) and (6)

$\frac{d^{2}z\left(x\right)}{{\mathrm{<figure-callout\; id="2"\; label="dx"\; filenames="C20071006784500311.png"\; state="\{\{state\}\}">dx</figure-callout>}}^2}+2k\frac{\mathrm{dz}\left(x\right)}{\mathrm{dx}}-1=0---\left(7\right)$

In the formula; $k=\frac{z\left(x\right)[z\left(x\right)-h]+x[d\left(x\right)-x]}{z\left(x\right)[d\left(x\right)-x]+x[z\left(x\right)-h]}---\left(8\right)$

Obtain the differential equation (9) by formula (7)

$\frac{\mathrm{dz}\left(x\right)}{\mathrm{dx}}+k-\sqrt{{\mathrm{<figure-callout\; id="2"\; label="k"\; filenames="C20071006784500311.png"\; state="\{\{state\}\}">k</figure-callout>}}^2+1}=0---\left(9\right)$

Obtain formula (10) by formula (1), (5)

$d\left(x\right)=\frac{\mathrm{afx}}{z\left(x\right)}---\left(10\right)$

By formula (8), (9), (10) and starting condition, separate the digital solution that the differential equation obtains reflecting mirror surface shape; Select camera according to application requirements, calibrate Rmin, the focal distance f of lens is determined the distance H o of catoptron from camera, calculates aperture of a mirror Do by (1) formula,

Determine systematic parameter af according to the visual field of using desired short transverse, obtain formula (11) by formula (1), (2) and (5), z (x) ≈ z ₀:

$\mathrm{tg\&phi;}=\frac{(\mathrm{af}-z_0)\frac{\mathrm{\&rho;}}{f}}{z_0-h}---\left(11\right)$

With the inconocenter point ρ=Rmin of largest circumference place in the center of circle as the plane ${\mathrm{\&omega;}}_{\max }=\frac{R_{\min }}{f}$

Corresponding visual field is ф max, obtains formula (12):

$\frac{\mathrm{\&rho;}}{f}=\frac{(z_0-h)\mathrm{tg}{\mathrm{\&phi;}}_{\max }}{{\mathrm{\&omega;}}_{\max }}+z_0---\left(12\right).$

Described parking stall recognition device comprises that being installed in the parking lot overhead view the common image sensor of parking position, is used for the microprocessor according to the data identification parking stall of common image sensor, and described common image sensor connects microprocessor.Vision sensor does not limit other any type of vision sensors, all is suitable for as long as can obtain the vision sensor of all parking space information in the parking lot, and the vision sensor number that is adopted depends on the size in parking lot.

Technical conceive of the present invention is: Flame Image Process and computer vision are constantly new technologies of development, adopt computer vision to observe four purposes in principle, i.e. the debating of the feature extraction of pre-service, the bottom, mid-level features known and by the explanation of image to senior sight.In general, computer vision comprises principal character, Flame Image Process and image understanding.Image is the extension of human vision.By machine vision, can hold the parking stall behaviour in service in the parking lot immediately exactly.The basis of image detection rapidity is that the information that vision is accepted is communication media with light; And image information is abundant and directly perceived, is that other present various Detection Techniques all can not provide so abundant and information intuitively.

The omnibearing vision sensor ODVS that developed recently gets up (OmniDirectional Vision Sensors) provide a kind of new solution for the panoramic picture that obtains scene in real time.The characteristics of ODVS are looking away (360 degree), can become piece image to the Information Compression in the hemisphere visual field, and the quantity of information of piece image is bigger; When obtaining a scene image, the riding position of ODVS in scene is free more; ODVS is without run-home during monitoring environment; Algorithm is simpler during moving object in the detection and tracking monitoring range; Can obtain the realtime graphic of scene.Therefore the fully-directional visual system based on ODVS developed rapidly in recent years, just becoming the key areas in the computer vision research, IEEE held the special symposial (IEEE workshop on Omni-directional vision) of annual omni-directional visual since 2000.Because parking stall measure need cover all parking stalls in the midfield, parking lot, therefore utilize omnibearing vision sensor can detect each parking stall at any time, as long as the centre that omnibearing vision sensor is installed in the top, parking lot does not also retrieve paper and the patent that omnibearing vision sensor is applied to parking guidance device technique field at present with regard to holding the parking stall behaviour in service in the parking lot easily.

Therefore, adopt omnibearing vision sensor ODVS also to utilize digital image processing techniques, whether in conjunction with some features that the parking stall in parking lot distributes and parks cars, it is occupied to detect each parking stall, induces in the field and induction information outside the venue for parking provides; When attaching most importance to, can monitor safety in the parking lot again, be equipped with the intelligentized insight of a pair of to the parking lot with parking stall measure.

Beneficial effect of the present invention mainly shows: 1, can collect each parking lot information, and can induce, be convenient to the utilization factor that the driver can in time find the parking stall, increase the parking lot outside the venue; 2, sensing range is wide, can detect with interior parking cars at 200 rice diameters the orientation; 3, installation and maintenance are noiseless, because video detector is installed on the top at middle part, parking lot often, therefore installation and maintenance can not influence the business in parking lot, do not need excavation yet, destroy the road surface; 4, low consumption easy to maintenance, traditional inductive coil detecting device needs excavated pavement to safeguard when damaging, and during video detecting device generation problem, can directly extract or repair facility, and has reduced maintenance cost; 5, detected parameters is abundant, not only the parking stall that can detect in the parking lot takies situation, by adding the various potential safety hazards that can also detect behind some new algorithms in the parking lot, such as accidents such as robber's difficulty of vehicle, fire, this is that general inductive coil detecting device is incomparable; 6, visuality can be passed to omnibearing realtime graphic the supvr in parking lot, realizes the function that monitors; 7, detecting reliability, accuracy height can equally with traditional inductive coil detecting device not have misoperation or flase drop and survey; 8, statistical computation is convenient, and algorithm is realized simple, is specially adapted to the management at large parking lot; 9, have good advance, extensibility, sustainable development, the video frequency car position detection technique is one of gordian technique of intelligent transportation system, itself just can become a system separately, by network can with advanced person's Vehicle Information System, etc. dynamic and intelligent traffic module be connected, realize more function.

(4) description of drawings

Fig. 1 reflexes to omni-directional visual planar imaging synoptic diagram for three-dimensional space;

Fig. 2 is the structural representation of omnibearing vision sensor;

Fig. 3 is the perspective projection imaging model synoptic diagram of omnibearing vision sensor and general perspective imaging model equivalence;

Fig. 4 is the omnibearing vision sensor undeformed simulation synoptic diagram of epigraph in the horizontal direction;

Fig. 5 is the structure function block diagram based on the parking guidance device of omnibearing vision sensor;

Fig. 6 is the functional block diagram based on the parking guidance function of omnibearing vision sensor;

Fig. 7 is the parking stall and the mapping relations synoptic diagram that detects the parking stall of parking lot video figure based on omnibearing vision sensor in the planimetric map of parking lot;

Fig. 8 is the numerical coding architectural schematic of the some parking space information in the geographical positional information in some parking lots and this parking lot in the expression city;

The Flame Image Process process flow diagram whether Fig. 9 stops for the Video Detection parking stall;

Figure 10 is the structural drawing with omnibearing vision sensor of remote access function.

(5) embodiment

Below in conjunction with accompanying drawing the present invention is further described.

Embodiment 1

With reference to Fig. 1～10, realize that parking guidance at first will solve two problems: 1) induce in the field, induce outside the venue; 2) parking stall takies the detection of situation.The embodiment of the invention is described in conjunction with solving above-mentioned two problems.

In inducing in described, at first be the situation that takies according to the parking stall, when entering the parking lot in parking vehicles, the parking space information that information issuing system just automatically can be parked offers the driver, so that the driver can find the parking stall that oneself parks cars rapidly; Such as the mode by the parking lot planimetric map shown in the accompanying drawing 7 or directly tell the modes such as number on parking stall.

Induce outside the venue described, except with the field in induce in the needed parking space information, also need relevant information, classification parking lot and space distribution information thereof, the parking lot affiliated facility information in parking lot.At first to solve supply and demand information matches problem for parking, promptly wish to know situation from nearest the provided parking space in trip purpose ground, this just relates to this parking stall that can park on (parking lot) on what orientation in city, which point in this orientation, at that point which parking stall, good city digital traffic coding scheme structure can be to the information of parking of issuing, inquiring about certain parking lot, and it all is very favorable carrying out empty wagons position statistics from the zones of different angle.

The numerical coding of the parking stall that the present invention proposes is mainly formed in two sub-sections with 28 figure places and is named preceding 22 geographical location information that are used for representing the parking lot, back 6 parking space information that are used for being illustrated in this parking lot.The meaning of each representative of coding scheme is shown in the following figure of accompanying drawing 8.Absolute position in the most preceding 6 these cities of bit representation, then 11 roads that are used to represent the place, parking lot are from intown relative position, followed by 5 be used to represent the distance of parking lot from the road starting end at place to the parking lot.The positional information that the parking lot has been arranged, traveler can find the parking lot that will park apace.

The represented meaning in each parking stall of city digital traffic coding scheme and parking lot is described, be used for representing that preceding 17 in preceding 22 of the geographical location information in parking lot are adopted the coordinate arranged modes to be, the most preceding 6 (the 1st to the 6th) represent the absolute position in this place, city, wherein go up 3 (the 1st to the 3rd) expression longitude (representing) with angle, following 3 (the 4th to the 6th) represents latitude (representing with angle), because all there are its longitude and latitude in each metropolis, this information is necessary to nonlocal driver and the sign on generalized information system; 11 (the 7th to 17) then are used for expression from intown relative position, its naming rule is: with significant position, city center is that (what there was the square in the city center is the center with the square to initial point, what do not have the square is center origin with the famous buildings in city center), East and West direction is the x axle, the north-south is the y axle, the city is divided into A, B, C, 4 quadrant districts of D, the quadrant district at the starting point place of the 7th bit representation road, consider that the megalopolis is in district radius 100km, the x (2 bit digital) that represents the two ends, street with 4 bit digital, y (2 bit digital) coordinate (is unit with km), the 8th x coordinate to the 9th bit representation road starting point, the 10th y coordinate to the 11st bit representation street starting point, the quadrant district at the terminal point place of the 12nd bit representation road, the 13rd x coordinate to the 14th bit representation street terminal point, the 15th y coordinate to the 16th bit representation street terminal point, for two parallel and two ends x, the street of y coordinate identical (coordinate points is at a distance of in the 1km scope) is aided with a, b, c....... order difference, represent with the 17th figure place determined in which city by above-mentioned coding definition, which orientation, distance and any bar road apart from the monumented point in this city; Ensuing the 18th to the 22nd 5 are encoded to the Native digits coding, from the ascending numbering of the origin-to-destination of road, begin from nearest intown crossing to compile for the side street lane, for road both sides street crossing is arranged, layout is carried out at single right two ends that extend to, a left side, on the 21st, increase 1 every 100m, therefore the distance that can represent 100Km from the 18th to the 21st; This coded system no matter driver enters the particular location that road can both be known the parking lot from the starting point of road or from the terminal point crossing of road; For have only one-sided runway if adopt the odd number layout at the left of road, in the right-hand employing even numbers layout of road, this message reflection is at the 22nd; Know in the time of making the driver on this road that travels by such layout that the parking lot is the Left or right at road, can avoid the unnecessary tune of driver.

After the driver enters the parking lot, with 6 last parking space information that are used for being illustrated in this parking lot in 28, carry out parking guidance in the field among the present invention.As parking guidance in the field, angle from the driver, at first to know floor, position, the parking stall that to stop, whether the type of judging the parking stall parked then is that the type with vehicle matches, for large-scale parking lot have also have in the multilayer, every layer also can be divided into a plurality of parking stalls in a plurality of positions, each position, each parking stall is specifically designed to a bit and parks oversize vehicle, and some parking stall is used to park dilly; The attribute of these floor attributes, position attribute, parking stall attribute and the size that can park cars all must define in the parking space information in the parking lot; Among the present invention with the floor attribute in the 1st of this 6 bit-encoded information definition parking lot, the position attribute in the 2nd definition parking lot, can the park cars attribute of vehicle size of size of the parking stall attribute in the 3rd～5 definition parking lot, a certain parking stall in the 6th definition parking lot.This attribute information of 6 all is to finish in the customization of accompanying drawing 7, all there is corresponding planimetric map each floor, each position according to the parking lot in accompanying drawing 2, just determined preceding 2 attribute in case downloaded this planimetric map, determine the parking stall attribute in the 3rd～5 definition parking lot then by this planimetric map customization parking stall, the attribute decision of this parking stall of the input attribute of vehicle size of size that can park cars behind the customization parking stall.

Among the present invention with 28 parking stall attribute informations of above-mentioned definition basic KEY as retrieval, statistics, such as knowing in certain city that the situation on parking stall just can obtain by preceding 17 Information Statistics in above-mentioned 28 on certain bar road, and the situation that will obtain the parking stall, the right on this road just can by in above-mentioned 28 preceding 17, the 22nd Information Statistics that are set at even number are obtained, just can obtain if obtain the situation on the parking stall in certain parking lot by preceding 22 Information Statistics in above-mentioned 28; The numerical coding system of this parking stall is very favorable to the parking information delivery system, and it can divide several level issue parking guidance information to the external world as required by rights.

As parking guidance except the attribute information of above-mentioned parking stall, it also is necessary also having two status informations, one is the occupied information of parking stall, another is the predetermined information of parking stall, about adopting the mode of video to detect among occupied information the present invention of parking stall, predetermined information about the parking stall is that traveler is scheduled to according to the needs time period of oneself by various means, therefore needing a start time in the present invention describes the seizure condition or the subscription state of parking stall, another one information is occupant's the number-plate number, so just can form a complete record.

Certain parking position state-detection table of table 1

Parking lot ID	Parking stall ID	Busy flag	Predetermined flag	Start time	License plate number

Certain parking position of table 2 utilizes the state history record sheet

Parking stall ID	Busy flag	Predetermined flag	Start time	License plate number

Parking lot ID in the table is exactly preceding 22 information in above-mentioned 28; Parking stall ID is exactly back 6 information in above-mentioned 28; If detect on this parking stall car is arranged, just busy flag is set to very, and start-of-record time while; If by the predetermined parking stall of network means, just predetermined flag is set to very, writes the schedule time and license plate number simultaneously in the start time.Two tables have been adopted among the present invention, be that parking stall state-detection table (shown in the table 1) and parking stall utilize state history record sheet (shown in the table 2), the data layout of these two tables is basic identical, has parking lot ID mainly to be provided with for aspect statistical regions empty wagons bit rate in the table 1; In case when detecting certain parking stall state and changing, the parking stall to be put in the record of certain parking stall of parking stall state-detection table and utilize the state history record sheet, form the historical service recorder of parking stall, parking stall shown in the table 2 utilizes the state history record sheet to use in the parking lot, therefore can remove parking stall ID to save storage space; Simultaneously certain new parking stall state is upgraded in the state-detection table of original parking stall state corresponding and this parking stall, so that in time reflect the online operating position in parking lot.

Solve the detection problem that the parking stall takies situation,, can adopt two kinds of detection modes: 1) based on the detection of color model on the Video Detection parking stall based on omnibearing vision sensor whether car being arranged; 2) based on the detection of image border profile (seeing that from depression angle the profile of vehicle all is rendered as the rectangle of rule).

The principle of work of described omnibearing vision sensor is: the manufacturing technology scheme of the opticator of ODVS camera head, ODVS camera head are mainly constituted by vertically downward catadioptric mirror with towards last camera.It is concrete that to constitute be to be fixed on bottom by the cylinder of transparent resin or glass by the image unit that collector lens and CCD constitute, the top of cylinder is fixed with the catadioptric mirror of a downward deep camber, the coniform body that between catadioptric mirror and collector lens, has a diameter to diminish gradually, this coniform body is fixed on the middle part of catadioptric mirror, and the purpose of coniform body is the light saturated phenomenon that causes in order to prevent superfluous light from injecting in cylinder inside.What Fig. 1 represented is the schematic diagram of the optical system of omnibearing vision sensor of the present invention.

Catadioptric omnidirectional imaging system can be carried out imaging analysis with the pin-hole imaging model, must must satisfy the requirement of real-time to the contrary projection of the real scene image of gathering but will obtain the perspective panorama picture.

The coordinate of the horizontal coordinate of object point and corresponding picture point is linear in the scene in parking lot just can guarantee that horizontal scene is undistorted, be installed in the height that can cover most of parking stalls in parking lot as the omnibearing vision sensor of parking guidance device, therefore monitor the parking stall situation on the horizontal direction in the whole parking lot, when the catadioptric minute surface of design omnibearing vision device, will guarantee in the horizontal direction indeformable.

At first select for use CCD (CMOS) device and imaging len to constitute camera in the design, preresearch estimates system physical dimension on the basis that the camera inner parameter is demarcated is determined the mirror surface shape parameter according to the visual field of short transverse then.

As shown in Figure 1, the projection centre C of camera is the horizontal scene h of distance place above the horizontal scene of road, and the summit of catoptron is above projection centre, apart from projection centre zo place.Be that true origin is set up coordinate system with the camera projection centre among the present invention, the face shape of catoptron is with z (X) function representation.The pixel q of distance images central point ρ has accepted from horizontal scene O point (apart from Z axle d), at the light of mirror M point reflection in as the plane.Therefore desirable detected status is undistorted on horizontal scene, requires the coordinate of the horizontal coordinate of scene object point and corresponding picture point linear;

d(ρ)＝αρ (1)

ρ is and the distance of the face shape central point of catoptron in the formula (1), and α is the magnification of imaging system.

If the normal that catoptron is ordered at M and the angle of Z axle are γ, the angle of incident ray and Z axle is Φ, and the angle of reflection ray and Z axle is θ.Then

$\mathrm{tg}\left(x\right)=\frac{d\left(x\right)-x}{z\left(x\right)-h}---\left(2\right)$

$\mathrm{tg\&gamma;}=\frac{\mathrm{dz}\left(x\right)}{\mathrm{dx}}---\left(3\right)$

$\mathrm{tg}\left(2\mathrm{\&gamma;}\right)=\frac{2\frac{\mathrm{dz}\left(x\right)}{\mathrm{dx}}}{1-\frac{d^{2}z\left(x\right)}{{\mathrm{dx}}^2}}---\left(4\right)$

By reflection law

2γ＝φ-θ

$\stackrel{\&CenterDot;}{\&CenterDot;\&CenterDot;}\mathrm{tg}\left(2\mathrm{\&gamma;}\right)=\mathrm{tg}(\mathrm{\&phi;}-\mathrm{\&theta;})=\frac{\mathrm{tg\&phi;}-\mathrm{tg\&theta;}}{1+\mathrm{tg\&phi;tg\&theta;}}---\left(6\right)$

Obtain the differential equation (7) by formula (2), (4), (5) and (6)

$\frac{d^{2}z\left(x\right)}{{\mathrm{<figure-callout\; id="2"\; label="dx"\; filenames="C20071006784500311.png"\; state="\{\{state\}\}">dx</figure-callout>}}^2}+2k\frac{\mathrm{dz}\left(x\right)}{\mathrm{dx}}-1=0---\left(7\right)$

In the formula; $k=\frac{z\left(x\right)[z\left(x\right)-h]+x[d\left(x\right)-x]}{z\left(x\right)[d\left(x\right)-x]+x[z\left(x\right)-h]}---\left(8\right)$

Obtain the differential equation (9) by formula (7)

$\frac{\mathrm{dz}\left(x\right)}{\mathrm{dx}}+k-\sqrt{{\mathrm{<figure-callout\; id="2"\; label="k"\; filenames="C20071006784500311.png"\; state="\{\{state\}\}">k</figure-callout>}}^2+1}=0---\left(9\right)$

Obtain formula (10) by formula (1), (5)

$d\left(x\right)=\frac{\mathrm{afx}}{z\left(x\right)}---\left(10\right)$

By formula (8), (9), (10) and starting condition, separate the digital solution that the differential equation can obtain reflecting mirror surface shape.The main digital reflex mirror of system's physical dimension is from the distance H o and the aperture of a mirror D of camera.Select suitable camera according to application requirements during the refractive and reflective panorama system design, calibrate Rmin, the focal distance f of lens is determined the distance H o of catoptron from camera, calculates aperture of a mirror Do by (1) formula.

Determining of systematic parameter:

Determine systematic parameter af according to the visual field of using desired short transverse.Obtain formula (11) by formula (1), (2) and (5), done some simplification here, with z (x) ≈ z ₀, main consideration is smaller with respect to the change in location of minute surface and camera for the height change of minute surface;

$\mathrm{tg\&phi;}=\frac{(\mathrm{af}-z_0)\frac{\mathrm{\&rho;}}{f}}{z_0-h}---\left(11\right)$

With the inconocenter point largest circumference place in the center of circle as the plane $\mathrm{\&rho;}=R\min \&RightArrow;{\mathrm{\&omega;}}_{\max }=\frac{R_{\min }}{f}$

Corresponding visual field is ф max.Then can obtain formula (12);

$\frac{\mathrm{\&rho;}}{f}=\frac{(z_0-h){\mathrm{tg\&phi;}}_{\max }}{{\mathrm{\&omega;}}_{\max }}+z_0---\left(12\right)$

The imaging simulation adopts the direction opposite with actual light to carry out.If light source is in the camera projection centre, equally spaced selected pixels point in the picture plane by the light of these pixels, intersects with surface level after mirror reflects, if intersection point is equally spaced, illustrates that then catoptron has the distortionless character of horizontal scene.The imaging simulation can be estimated the imaging character of catoptron on the one hand, can calculate aperture of a mirror and thickness exactly on the other hand.

Image transformation relates to the conversion between the different coordinates.In the imaging system of video camera, what relate to has following 4 coordinate systems; (1) real-world coordinates is XYZ; (2) with the video camera be the coordinate system x^y^z^ that formulate at the center; (3) photo coordinate system, formed photo coordinate system x*y*o* in video camera; (4) computer picture coordinate system, the coordinate system MN that the computer-internal digital picture is used is a unit with the pixel.

According to the different transformational relation of above several coordinate systems, just can obtain needed omnidirectional vision camera imaging model, converse the corresponding relation of two dimensional image to three-dimensional scenic.The approximate perspective imaging analytical approach that adopts catadioptric omnibearing imaging system among the present invention is with the formed corresponding relation that is converted to three-dimensional scenic as the planimetric coordinates two dimensional image in the video camera, Fig. 3 is general perspective imaging model, d is an object height, ρ is an image height, t is an object distance, and F is image distance (equivalent focal length).Can obtain formula (13)

$d=\frac{t}{F}\mathrm{\&rho;}---\left(13\right)$

When above-mentioned horizontal scene does not have the design of catadioptric omnibearing imaging system of distortion, require the coordinate of the horizontal coordinate of scene object point and corresponding picture point linear, represent suc as formula (1); Comparison expression (13), (1), horizontal as can be seen scene does not have the be imaged as perspective imaging of the catadioptric omnibearing imaging system of distortion to horizontal scene.Therefore with regard to horizontal scene imaging, the catadioptric omnibearing imaging system that horizontal scene can not had distortion is considered as having an X-rayed camera, and α is the magnification of imaging system.If the projection centre of this virtual perspective camera is C point (seeing accompanying drawing 3), its equivalent focal length is F.Comparison expression (13), (1) formula can obtain formula (14);

$\mathrm{\&alpha;}=\frac{t}{F};t=h---\left(14\right)$

Obtain formula (15) by formula (12), (14)

$F=\frac{{\mathrm{fh\&omega;}}_{\max }}{(z_0-h){\mathrm{tg\&phi;}}_{\max }+z_0{\mathrm{\&omega;}}_{\max 0}}---\left(15\right)$

Carry out the system imaging simulation according to above-mentioned omnidirectional vision camera imaging model, by the camera projection centre send through in the pixel planes equidistantly after the reflection of the light family of pixel, intersection point on the surface level in the parking lot of distance projection centre 5m is equally spaced basically, as shown in Figure 4.Therefore according in the above-mentioned design concept this patent relation between the coordinate of the coordinate of parking lot surface level and corresponding comprehensive picture point being reduced to linear relationship, that is to say that design by mirror surface be XYZ to the conversion of photo coordinate system with real-world coordinates can be the linear dependence of ratio with magnification α.Be conversion below from photo coordinate system to the used coordinate system of computer-internal digital picture, the image coordinate unit that uses in the computing machine is the number of discrete pixel in the storer, so also need round the imaging plane that conversion just can be mapped to computing machine to reality as the coordinate on plane, its conversion expression formula is for to be provided by formula (16);

$M=O_m-\frac{x^{*}}{S_x};$ $N=O_n-\frac{y^{*}}{S_y};---\left(16\right)$

In the formula: Om, On are respectively the line number and the columns at the some pixel place that the initial point of image plane shone upon on the computer picture plane; Sx, Sy are respectively scale factor in the x and y direction.Determining of Sx, Sy is by placing scaling board apart from the Z place between camera and mirror surface, video camera is demarcated the numerical value that obtains Sx, Sy, and unit is (pixel); Om, On.Determine it is that unit is (pixel) according to selected camera resolution pixel.

Fig. 2 is the structural drawing of omnibearing vision sensor, in this patent omnibearing vision sensor is installed in the place that can cover a large amount of parking stalls from 5 meters of ground, parking lot height, the video image that obtained.Described omnibearing vision sensor comprises in order to the evagination catadioptric minute surface 1 of object in the field in the reflection parking lot, in order to prevent the saturated dark circles cone 2 of anaclasis and light, transparent cylinder 3 and to be used to take the camera 4 of imaging body on the evagination mirror surface, described evagination catadioptric minute surface 1 is positioned at the top of transparent cylinder 3, evagination catadioptric minute surface 1 down, dark circles cone 2 is fixed on the bottom center of evagination catadioptric minute surface 1, and described camera 4 facing to evagination catadioptric minute surface up.

The Video Detection of parking stall is the key of total system, realize detection that the parking stall takies situation needing on the Flame Image Process to safeguard by the background image threshold value and parking stall that step such as background subtracting is confirmed to be detected on whether vehicle is arranged.In general, the color of ground color in the parking lot and vehicle has tangible difference.Utilize the color characteristic in the video image to judge whether the parking stall is shared by vehicle in this patent, accompanying drawing 9 is based on the treatment scheme of the video frequency car position detection of color model.

Whether parking stall on have car take, the first step is to obtain a more stable reference image in image recognition, utilizes it to do the background subtracting algorithm then if judging.Adopted the threshold value of in each parking stall frame, extracting an average chrominance in this patent, as the local color threshold value of each parking stall.The color of considering ground in the parking lot all is approaching, therefore can stop having in the parking stall image of vehicle at those, asks its mean value according to existing parking stall frame local color threshold value, obtains the global color threshold value of all parking stalls, whole parking lot.In actual use, some variations also can take place in the color model on ground, parking lot, therefore need constantly upgrade the color threshold data.

Whether the YCrCb color model of employing image is discerned certain parking stall occupied.Because the Y component is to light sensitive in the YCrCb color space, and these two components of CrCb are just relevant to color, and the color of distinguishing vehicle by these two components of CrCb and the color on ground, parking stall can reach whether occupied purpose of detection parking stall.By the Cr in the color space, the detection of Cb component can reach and the irrelevant effect of parking lot intraoral illumination light basically.Therefore from the color space of the video image that omnibearing vision sensor obtained is RGB, needs in program video image from the RGB color space conversion to the YCrCb color space, and conversion formula (17) provides,

Y＝0.29990*R+0.5870*G+0.1140*B

(17)

Cr＝0.5000*R-0.4187*G-0.0813*B+128

Cb＝-0.1787*R-0.3313*G+0.5000*B+128

Further, adopt the background subtracting algorithm to obtain parking space information, so-called background subtracting algorithm is also referred to as difference method, is a kind of image processing method that is usually used in detected image variation and moving object.

The computing formula of background subtracting as the formula (18),

f _d(X，t ₀，t _i)＝f(X，t _i)-f(X，t ₀) (18)

In the formula: f _d(X, t ₀, t _i) be to photograph the colourity result who carries out image subtraction between panoramic picture and reference panoramic picture in real time.F (X, t _i) be to photograph image chroma value, f (X, t in real time ₀) be reference image chroma reference value.In order to judge whether vehicle is arranged in the frame of parking, make background subtracting in the defined frame of parking in the figure in the lower right corner of the planimetric map that therefore stops and calculate, judge Cr, whether these two components of Cb have surpassed the threshold range of defined.Just park in this parking stall of mark if surpassed the words of the threshold range of defined, otherwise the parking stall does not take.Judgment formula as the formula (19),

$\left\{\begin{array}{c}1,\mathrm{if}(\mathrm{abs}(\mathrm{Cr}-\mathrm{ThresholdCr})>\mathrm{<figure-callout\; id="1"\; label="threshold"\; filenames="C20071006784500311.png"\; state="\{\{state\}\}">threshold</figure-callout>}1||\mathrm{abs}(\mathrm{Cb}-\mathrm{ThresholdCb})>\mathrm{threshold}2)\\ 0,\mathrm{else}\end{array}\right.---\left(19\right)$

In the formula: Cr represents the Cr mean value in the frame of detected current parking stall, Cb represents the Cb mean value in the frame of detected current parking stall, ThresholdCr represents the Cr mean value on ground in the frame of defined parking stall, ThresholdCb represents the Cb mean value on ground in the frame of defined parking stall, threshold1 represents the threshold range of Cr, and threshold2 represents the threshold range of Cb.

Adopt omnibearing vision sensor to detect resulting parking stall figure shown in the figure on parking planimetric map the right, though the situation that this detection figure parks in seeing in the parking lot in a big way, but also exist problems such as being not easy to understand, can not covering interior all parking stalls of whole large parking lot, therefore adopt the parking stall mapping mode to set up the contact that detects between figure and the issue figure (planimetric map) among the present invention, the specific implementation method is:

Based on the parking guidance device of comprehensive sensor when coming into operation, at first to arrange concrete condition and finish the preceding system customization preliminary work of use according to the parking stall in parking lot, specific practice: the planimetric map that at first is written into the parking lot, planimetric map is made can select various graphic making softwares for use, by the size on the actual parking stall parking frame that draws in proportion, and in parking frame definition parking stall number, as shown in Figure 7; Then want the surveyed area of customization parking stall frame on the comprehensive video figure, and the mapping relations of the parking stall on the planimetric map shown in foundation and the accompanying drawing 7, the mapping relations of the parking stall on the planimetric map shown in comprehensive video figure and the accompanying drawing 7.System user interface shown in opening, frame video image in the left side at this interface is the parking lot that obtains from omnibearing vision sensor, the right is that its one duplicates, and on the image on the right the parking stall is edited, and purpose is to set up the surveyed area of parking stall frame.The user can directly draw the parking stall frame on panorama sketch after pressing interpolation parking stall frame, the drafting means adopt the mouse drag and drop to realize.

Because some deformation can take place in the captured comprehensive panorama sketch of omnibearing vision sensor, it is an irregular quadrilateral that the quadrilateral parking stall of rule is reflected on the image, therefore when drawing detection parking stall frame, need adopt irregular quadrilateral definition, the means that adopt in this patent are to determine four points of parking stall by mouse, define with the straight line ways of connecting then and detect the parking stall.

After defining the detection parking stall, mouse is placed on the right button of pressing mouse in this parking stall then, can eject a parking stall property window in the program, in this window, be chosen in the car item that is defined among the definition parking lot inner plane figure, then in the type of selecting to park cars.At this moment the definition that detects the parking stall, the coding of an interior parking space information and the mapping relations of having set up the parking stall on detection parking stall and the planimetric map have just been finished.

Just can detect and monitor, when press begin to monitor button after, system can eject the window identical with accompanying drawing 7, has mapping relations owing to detect the parking stall with parking stall on the planimetric map shown in the accompanying drawing 7, so can reflect the situation of actual detected on planimetric map.Equally also this planimetric map can be distributed on the WEB page, make traveler be directly acquainted with the situation of parking in certain parking lot, for the remote reservation parking stall provides new means by the internet.

Recited above mainly is the information acquisition problem in parking lot, as information processing in the parking guidance also is a very important link, whether the purpose of information processing is the information that the appropriate format that provides to the driver is provided for the parking lot behaviour in service that will collect and peripheral path information processing, block up as full sky (residue parking stall situation), the collector distributor road in parking lot etc.In addition, also undertaking the tasks such as changing pattern of the parking lot information of storing, processing processing parking lot operating position.These functions will lay the foundation for services such as the forecast of parking demand situation, parking stall reservation will be provided future; The ability size of information processing, effect quality depend on the data structure of descriptor to a great extent, and parking stall numerical coding system shown in Figure 8 has improved the information processing capability in the parking guidance.

Also can adopt other vision sensors to replace the detection that omnibearing vision sensor carries out the parking stall.

Embodiment 2

With reference to Fig. 1～Figure 10, in the frame of Video Detection parking stall, whether have the image processing techniques of car, all the other are identical with embodiment 1.This method mainly is at the situation in parking lot out of doors, out of doors since the background in parking lot can be because the variation of weather flies upward thing (leaf, the polybag etc.) influence of face color model over the ground in (rain, snow etc.), the parking lot, whether have car can bring bigger false recognition rate, therefore can adopt the edge of image detection method on the parking stall measure of motor pool with color model if carrying out the parking stall;

The border of car body is the very important descriptor of a class of describing the car body feature, and these borders may produce marginal information in imaging process.The edge is meant the combination that those pixels of significant change are arranged in its surrounding pixel gray scale.The edge is the vector with amplitude and direction, and it shows as the sudden change of gray scale in image.Rim detection is exactly the noncontinuity that will detect this gray scale in the image.

There is several method to select to rim detection at present, because what expectation obtained in this patent is the edge of the car body on certain parking stall, and it is less demanding to the integrality and the slickness of edge wheel hub, therefore it is simple that we adopt calculating wherein, classical edge detection method-the method for differential operator of fast operation, this method relies on image differentiated and tries to achieve gradient and carry out rim detection, main from marginal point often corresponding to the big point of single order differential amplitude, while also sets out corresponding to the zero cross point of second-order differential, design some single orders or second-order differential operator, try to achieve its gradient or second derivative zero crossing, select certain threshold value to extract the border again.

Described edge detection method can be divided into following four steps haply:

1. filtering: edge detection algorithm mainly is based on the first order derivative and the second derivative of image intensity, but the calculating of derivative is very sensitive to noise, therefore must use wave filter to improve the performance of the edge detection method relevant with noise.It may be noted that most of wave filters have also caused the loss of edge strength when reducing noise.Therefore the edge strengthens and reduces between the picture noise needs to obtain a kind of balance.

2. strengthen: the basis that strengthens the edge is a changing value of determining each vertex neighborhood intensity in the image.Enhancement algorithms can be given prominence to the point that the neighborhood intensity level has significant change.The edge strengthens generally to be finished by the compute gradient amplitude.

3. detect: in image, have the gradient magnitude of many points bigger, and these might not all be the edges under specific situation, so should be with coming someway to determine that those points are marginal points.The simplest rim detection criterion is a gradient magnitude A value criterion.

4. locate: determine the pixel at place, edge, if more accurate definite marginal position also can come the estimated edge position on subpixel resolution, the direction at edge also can be estimated.

Adopt Suo Beier (Sobel) operator as edge detection algorithm in the present invention, the Sobel operator adopts the template of 3*3 size, has so just avoided compute gradient on the interpolated point between the pixel.The Sobel operator calculates partial derivative with following formula:

S _x＝(a ₂+ca ₃+a ₄)-(a ₀+ca ₇+a ₆)

(20)

S _y＝(a ₀+ca ₁+a ₂)-(a ₆+ca ₅+a ₄)

Constant c is 2 in the formula.The Sobel operator can be realized with following convolution template:

$S_x=\left[\begin{array}{ccc}-1& 0& 1\\ -2& 0& 2\\ -1& 0& 1\end{array}\right]$ $S_y=\left[\begin{array}{ccc}1& 2& 1\\ 0& 0& 0\\ -1& -2& -1\end{array}\right]---\left(21\right)$

The method that can be used for detecting car body in the scene of the parking stall in parking lot can be a template matching method.This method is that the object of all positions in the model of the car body that will will detect and the parking stall scene compares, and investigates the object that whether exists the model with car body to be complementary.The matching algorithm that the present invention adopts is: the vehicle image template of establishing the known target object is T, and size be M*N, and vehicle image template T leaves in the computing machine, makes different templates according to different vehicle needs, so that use when comparing; The image that can park cars in the parking stall to be detected is I, and size is L*W (L＞M, W＞N).The process of coupling is to manage a template T to be superimposed upon on the image I, and compares the difference of the subimage of the I under T and its covering.If difference is less than certain prior preset threshold, then think T this place and.The subimage of I has coupling preferably, has promptly found destination object.Entire image to be matched is pressed individual element scanning and implemented aforesaid operations, then can determine whether exist the determined destination object of template T, i.e. certain type vehicle in the image I.The mathematical description of matching process can be represented by formula (22):

$D(i,j)=\underset{m=1}{\overset{M}{\mathrm{\&Sigma;}}}\underset{n=1}{\overset{N}{\mathrm{\&Sigma;}}}{[I_{\mathrm{\&eta;}}(m,n)-T(m,n)]}^2---\left(22\right)$

Before images match is handled, need the pre-service of image, comprise and carry out edge extraction, binaryzation, profile is linked to be closed curve and the curve inner region is carried out white pixel fill.Obtain the centre of form coordinate and the center invariant moments of the binary picture of the car body template that is stored in the computing machine and the object in the scene of parking stall respectively and (suppose object zero lap in the scene of parking stall here, omni-directional visual passed after device is installed in certain altitude the vehicle on the parking stall is not overlapped), with the invariant moments of template respectively with the parking stall scene in each object invariant moments relatively, difference can predicate the object shown in the template-be car body less than certain preset threshold.Then the centre of form of object in the scene of parking stall is aimed at the centre of form of template, utilize " convergent-divergent " and image processing methods such as " rotations " that template is alignd with object in the scene, to determine whether object really is the represented object of template in the scene, if just be judged to be on this parking stall car arranged, then according to the result of template matches, obtain the type of parking vehicle on this parking stall.

Turn to the image of M*N for a spatial spreading, if i, j is each pixel coordinate in the parking stall scene of discretize, and (i is (i for coordinate j) to f, the gray-scale value of pixel j), according to the notion and the acquiring method of above-mentioned invariant moments, according to the disposal route of the digital picture of discretize, " central moment " of object can be approached by following dual summation form in the then visual scene, represent with formula (23)

$D_{\mathrm{pq}}=\underset{i\&Element;R}{\mathrm{\&Sigma;}}\underset{j\&Element;R}{\mathrm{\&Sigma;}}{(i-\stackrel{\&OverBar;}{i})}^p{(j-\stackrel{\&OverBar;}{j})}^{q}f(i,j)---\left(23\right)$

Wherein: i=m ₁₀/ m ₀₀J=m ₀₁/ m ₀₀ $m_{00}=\underset{i\&Element;R}{\mathrm{\&Sigma;}}\underset{j\&Element;R}{\mathrm{\&Sigma;}}f(i,j);$ $m_{10}=\underset{i\&Element;R}{\mathrm{\&Sigma;}}\underset{j\&Element;R}{\mathrm{\&Sigma;}}i\&times;f(i,j);$ $m_{01}=\underset{i\&Element;R}{\mathrm{\&Sigma;}}\underset{j\&Element;R}{\mathrm{\&Sigma;}}j\&times;f(i,j).$

For the center invariant moments is not become with proportional zoom, carry out normalized, can get normalization permanent center square and be: η _Pq=D _Pq/ D ₀₀, r=(p+q)/2+1,

If p+q＜2 then can be derived two constant central moment functions of RST and be represented with formula (24),

The concrete steps of described template matching algorithm are as follows:

1) each object in template and the parking stall scene is carried out pre-service: utilize " Roberts " operator respectively the object in the scene of parking stall to be carried out edge sharpening, and select appropriate threshold to carry out binaryzation.Marginal information to the object of closure is carried out the skeleton refinement, with the closed curve at direction chain code (or array) expression scenery edge, with the connectedness of the closed curve of the object edge that guarantees to represent with single pixel chain.Carry out the zone and fill, with the zone in the maximum gradation value filling object edge closed curve.If the region area of asking is greater than a threshold value (projected area of minimum vehicle is this threshold value), calculate the centre coordinate (barycentric coordinates) of scenery in the parking stall then according to formula (23), ask for second order normalization center invariant moments according to formula (24), otherwise do not carry out to judge.

2) with the center invariant moments of the object of parking scene and template relatively, and get the phase difference and can tentatively think the object that template is represented less than a certain threshold value.The centre of form with template overlaps with the centre of form coordinate of object in the scene then.

3) template is carried out convergent-divergent so that the consistent size of the essentially identical object of invariant moments with it in itself and the scene.Utilize the rotational transform of coordinate to determine the anglec of rotation of object correspondence in the scene.The algorithm that convergent-divergent and rotation are adopted is respectively:

The convergent-divergent algorithm:

x ₂＝N(x ₁-x ₀)+x ₀

(25)

y ₂＝N(y ₁-y ₀)+y ₀

The rotation algorithm:

x ₂＝(x ₁-x ₀)cosθ+(y ₁-y ₀)sinθ+x ₀

(26)

y ₂＝(y ₁-y ₀)cosθ+(x ₁-x ₀)sinθ+y ₀

More than in two groups of formulas only the edge contour point to template carry out convergent-divergent and rotation processing, also need link afterwards and the closed curve interior pixel is filled.(x in the formula ₂, y ₂) be the new coordinate of pixel; (x ₁, y ₁) be former coordinate; (x ₀, y ₀) be the centre of form coordinate of object; N is a zoom factor; θ is the angle of rotation.

Whether 4) binary picture that above-mentioned two width of cloth have been aimed at is carried out exclusive-OR operation, really be the type of car body shown in the template and vehicle with this object that how much confirms of residual pixel.

Sometimes do not need to know the type that parks cars on the parking position, therefore also can adopt easier mode to detect, after the edge contour point and link and the filling of closed curve interior pixel that detect on the parking stall, judge the similarity of this figure and rectangle, if the similarity of calculating is in threshold range, just being judged to be is vehicle, (because the profile of vehicle can be approximated to be a rectangle).

The microprocessor that is connected with omnibearing vision sensor is an embedded system, and its structure as shown in figure 10; Computer-centric system among Fig. 5 can adopt general PC, can adopt server for large parking lot, omnibearing vision sensor that is used for Video Detection in the parking lot and above-mentioned PC or server constitute LAN (Local Area Network), by PC or server and outside formation wide area network, to realize the inside and outside parking guidance in field as shown in Figure 6, the implementation algorithm among the present invention is realized by Java language.

Claims (3)

1, a kind of parking guidance system based on omnidirectional computer vision, comprise the parking stall recognition device, the parking space information statistics distributing device that are installed in each parking lot, it is characterized in that: described parking guidance system also comprises the parking guidance controller of the parking space information that is used to receive and handle each parking lot, and described parking guidance controller comprises:

The parking space information receiver module is used to gather the parking space information in each parking lot;

The parking space information coding module, the parking space information that is used for gathering converts numerical coding to, and described numerical coding is 28 figure places, the geographical location information in preceding 22 bit representation parking lots, the concrete parking space information of back 6 bit representations in this parking lot; In geographical location information, the absolute position in the most preceding 6 these cities of bit representation, then the road at place, 11 bit representation parking lot is from intown relative position, followed by the distance of 5 bit representation parking lots from the road starting end at place to the parking lot; In concrete parking space information, the floor attribute in the 1st definition parking lot, the position attribute in the 2nd definition parking lot, the parking stall attribute in the 3rd～5 definition parking lot, can the park cars attribute of size of this parking stall in the 6th definition parking lot;

Berth, parking lot information processing module is used for each parking stall in each parking lot is produced a record, sets up the inquiry inlet of each record, and the corresponding numerical coding of each record is outwards issued;

In described parking space information coding module, represent that the 4th to the 6th bit representation latitude for digitally coded the 1st with angle to the 3rd bit representation longitude, represent with angle; With significant position, city center is initial point, East and West direction is the x axle, the north-south is the y axle, the city is divided into A, B, C, a D4 quadrant district, the quadrant district at the starting point place of the 7th bit representation road represents x, the y coordinate at two ends, street, road place with 4 bit digital, and x is 2 bit digital, y is 2 bit digital, and is unit with km; The 8th x coordinate to the 9th bit representation road starting point, the 10th y coordinate to the 11st bit representation street starting point, the quadrant district at the terminal point place of the 12nd bit representation road, the 13rd x coordinate to the 14th bit representation street terminal point, the 15th y coordinate to the 16th bit representation street terminal point, at a distance of in the 1km scope, the 17th figure place is distinguished with the English alphabet order as coordinate points; 5 of the 18th to the 22nd are encoded to the Native digits coding, from the ascending numbering of the origin-to-destination of road, begin from nearest intown crossing to compile for the side street lane, for road both sides street crossing is arranged, layout is carried out at single right two ends that extend to, a left side, increases 1 every 100m on the 21st, for have only one-sided runway if adopt the odd number layout at the left of road, in the right-hand employing even numbers layout of road, the 22nd bit representation both sides layout, left side layout or right side layout;

Described parking stall recognition device comprises the omnibearing vision sensor of overlooking parking position that is installed in the top, parking lot, is used for the microprocessor according to the data identification parking stall of omnibearing vision sensor, and described omnibearing vision sensor connects microprocessor;

Described omnibearing vision sensor comprises in order to the evagination catadioptric minute surface of object in the reflection field, parking lot, in order to prevent dark circles cone, the transparent cylinder that anaclasis and light are saturated and to be used to take the camera of imaging body on the evagination mirror surface, described evagination catadioptric minute surface is positioned at the top of transparent cylinder, evagination catadioptric minute surface down, the dark circles cone is fixed on the bottom center of evagination catadioptric minute surface, and described camera facing to evagination catadioptric minute surface up;

Described microprocessor also comprises:

The view data read module is used to read the video image information of coming from the omnibearing vision sensor biography;

The image data file memory module, the video image information that is used for reading into is kept at storage unit by file mode;

The omnibearing vision sensor demarcating module is used to set up the material picture in space and the corresponding relation of the video image that is obtained;

Parking stall customization and mapping relations are set up module, are used to customize the parking stall frame on parking lot planimetric map and the comprehensive Video Detection figure, are used to set up the mapping relations of the parking stall that customizes on parking lot planimetric map and the comprehensive Video Detection figure;

The color space conversion module is used to finish the conversion of image rgb color space to the YCrCb color space, and conversion formula (17) provides:

Y＝0.29990*R+0.5870*G+0.1140*B

Cr＝0.5000*R-0.4187*G-0.0813*B+128 (17)

Cb＝-0.1787*R-0.3313*G+0.5000*B+128

Image pre-service and parking stall measure module, whether have car, adopt the background subtracting method if being used for detecting the parking stall frame, the computing formula of described background subtracting as the formula (18),

f _d(X，t ₀，t _i)＝f(X，t _i)-f(X，t ₀)

(18)

In the formula: f _d(X, t ₀, t _i) be to photograph the colourity result who carries out image subtraction between panoramic picture and reference panoramic picture, f (X, t in real time _i) be to photograph image chroma value, f (X, t in real time ₀) be reference image chroma reference value;

And judge whether two components of image C r, Cb after subtracting each other surpass the preset threshold value scope, if surpass the threshold value of defined, then park in this parking stall of mark, otherwise the parking stall do not take, judgment formula as the formula (19):

$\left\{\begin{array}{c}1,\mathrm{if}(\mathrm{abs}(\mathrm{Cr}-\mathrm{ThresholdCr})>\mathrm{threshold}1||\mathrm{abs}(\mathrm{Cb}-\mathrm{ThresholdCb})>\mathrm{threshold}2)\\ 0,\mathrm{else}\end{array}\right.---\left(19\right)$

In the formula: Cr represents the Cr mean value in the frame of detected current parking stall, Cb represents the Cb mean value in the frame of detected current parking stall, ThresholdCr represents the Cr mean value on ground in the frame of defined parking stall, ThresholdCb represents the Cb mean value on ground in the frame of defined parking stall, threshold1 represents the threshold range of Cr, and threshold2 represents the threshold range of Cb.

2, a kind of parking guidance system based on omnidirectional computer vision, comprise the parking stall recognition device, the parking space information statistics distributing device that are installed in each parking lot, it is characterized in that: described parking guidance system also comprises the parking guidance controller of the parking space information that is used to receive and handle each parking lot, and described parking guidance controller comprises:

The parking space information receiver module is used to gather the parking space information in each parking lot;

The parking space information coding module, the parking space information that is used for gathering converts numerical coding to, and described numerical coding is 28 figure places, the geographical location information in preceding 22 bit representation parking lots, the concrete parking space information of back 6 bit representations in this parking lot; In geographical location information, the absolute position in the most preceding 6 these cities of bit representation, then the road at place, 11 bit representation parking lot is from intown relative position, followed by the distance of 5 bit representation parking lots from the road starting end at place to the parking lot; In concrete parking space information, the floor attribute in the 1st definition parking lot, the position attribute in the 2nd definition parking lot, the parking stall attribute in the 3rd～5 definition parking lot, can the park cars attribute of size of this parking stall in the 6th definition parking lot;

Berth, parking lot information processing module is used for each parking stall in each parking lot is produced a record, sets up the inquiry inlet of each record, and the corresponding numerical coding of each record is outwards issued;

In described parking space information coding module, represent that the 4th to the 6th bit representation latitude for digitally coded the 1st with angle to the 3rd bit representation longitude, represent with angle; With significant position, city center is initial point, East and West direction is the x axle, the north-south is the y axle, the city is divided into A, B, 4 quadrant districts of C, D, the quadrant district at the starting point place of the 7th bit representation road represents x, the y coordinate at two ends, street, road place with 4 bit digital, and x is 2 bit digital, y is 2 bit digital, and is unit with km; The 8th x coordinate to the 9th bit representation road starting point, the 10th y coordinate to the 11st bit representation street starting point, the quadrant district at the terminal point place of the 12nd bit representation road, the 13rd x coordinate to the 14th bit representation street terminal point, the 15th y coordinate to the 16th bit representation street terminal point, at a distance of in the 1km scope, the 17th figure place is distinguished with the English alphabet order as coordinate points; 5 of the 18th to the 22nd are encoded to the Native digits coding, from the ascending numbering of the origin-to-destination of road, begin from nearest intown crossing to compile for the side street lane, for road both sides street crossing is arranged, layout is carried out at single right two ends that extend to, a left side, increases 1 every 100m on the 21st, for have only one-sided runway if adopt the odd number layout at the left of road, in the right-hand employing even numbers layout of road, the 22nd bit representation both sides layout, left side layout or right side layout;

Described parking stall recognition device comprises the omnibearing vision sensor of overlooking parking position that is installed in the parking lot center of top, is used for the microprocessor according to the data identification parking stall of omnibearing vision sensor, and described omnibearing vision sensor connects microprocessor;

Described omnibearing vision sensor comprises in order to the evagination catadioptric minute surface of object in the field in the reflection parking lot, in order to prevent dark circles cone, the transparent cylinder that anaclasis and light are saturated and to be used to take the camera of imaging body on the evagination mirror surface, described evagination catadioptric minute surface is positioned at the top of transparent cylinder, evagination catadioptric minute surface down, the dark circles cone is fixed on the bottom center of evagination catadioptric minute surface, and described camera facing to evagination catadioptric minute surface up;

Described microprocessor also comprises:

The view data read module is used to read the video image information of coming from the omnibearing vision sensor biography;

The image data file memory module, the video image information that is used for reading into is kept at storage unit by file mode;

The omnibearing vision sensor demarcating module is used to set up the material picture in space and the corresponding relation of the video image that is obtained;

Parking stall customization and mapping relations are set up module, are used to customize the parking stall frame on parking lot planimetric map and the comprehensive Video Detection figure, are used to set up the mapping relations of the parking stall that customizes on parking lot planimetric map and the comprehensive Video Detection figure;

The parking stall measure module is used for judging the similarity of this figure and rectangle after detecting edge contour point on the parking stall and link and closed curve interior pixel and filling, if the similarity of calculating in threshold range, just being judged to be is vehicle.

3, the parking guidance system based on omnidirectional computer vision as claimed in claim 2, it is characterized in that: set up in the module in customization of described parking stall and mapping relations, parking stall frame on the comprehensive Video Detection figure of described customization, be four points determining the parking stall by input equipment, define with the straight line ways of connecting then and detect the parking stall.

Images