CN201825037U - Lane departure alarm device for vehicles on highway - Google Patents

Abstract

The utility model discloses a lane departure alarm device for vehicles on a highway, which is composed of a vision acquiring and processing system and an alarm system. The vision acquiring and processing system consists of a camera, an image acquiring card and a computer, an output end of the image acquiring card is connected with an input end of the computer, and an output end of the computer is connected with the alarm system. The camera acquires image information and stores the image information in the image acquiring card, the image acquiring card transmits the image information to the computer, and the computer processes the image information and sends out alarm signals to the alarm system according to processed results. The alarm system is composed of a sound installation and an indicator lamp, and input ends of the sound installation and the indicator lamp are connected to the output end of the computer. The lane departure alarm device realizes mutual complementation of information based on area division and edge detection, improves lane boundary extraction precision, can detect lanes in various complicated road environments in a robust manner, uses the distance from a central line of a vehicle to a left lane and the distance from the central line of the vehicle to a right lane as alarm basis to realize lane departure alarm effectively.

Description

The deviation warning device of vehicle on highway

Technical field

The utility model belongs to the automobile assistant driving field of research, is specifically related to a kind of deviation warning device of vehicle on highway.

Background technology

Along with the develop rapidly of China's express highway, the active safety of vehicle in express highway more and more is subjected to people's attention.Particularly in passenger and freight transportation because driver tired driving or other cause automotive run-off-road with other factor, knock into the back, side impact, knock the intermediate isolating band or sweep away traffic accident such as road, cause the loss of goods and personnel's injures and deaths.The research and development of vehicle lane deviation warning device become very necessary and urgent.

The correct extraction of lane information is the prerequisite and the basis of whole device normal operation in the vehicle lane deviation warning device.At present, in the vehicle lane deviation warning device of existing exploitation, most gray level images that adopt are handled, and the extraction of lane information is subjected to road surface shade, greasy dirt easily, traffic lane line is imperfect and the interference of factor such as road surface breakage, can not extract lane information complete and accurate.Therefore, study and a kind ofly can effectively reflect lane information, thereby the deviation warning device that can effectively send vehicle is very to be necessary.

Summary of the invention

Defective or deficiency at above-mentioned prior art existence, the purpose of this utility model is, a kind of deviation warning device of vehicle on highway is provided, after this device obtains the position and direction information in track by vision collecting and disposal system, compare, discern with setting value, extract the difference factor,, surpass threshold value and carry out the deviation warning this factor and deviation control threshold.

In order to achieve the above object, the utility model adopts following technical solution:

A kind of deviation warning device of vehicle on highway, this device is made up of vision collecting and disposal system, warning:

Vision collecting and disposal system be used for collection vehicle position and direction graphicinformation and detect the track; This system comprises camera, image pick-up card, computing machine, and the mouth of described image pick-up card connects the computing machine input end, and the mouth of described computing machine connects warning; The graphicinformation of described camera real-time collection vehicle position and direction, and deposit the graphicinformation that collects in image pick-up card, image pick-up card is real-time transmitted to computing machine with graphicinformation, computer real-time is handled the graphicinformation that receives, and sends alerting signal according to result to warning;

Warning is made up of indicator lamp and sound equipment, and the input end of described sound equipment and indicator lamp is connected the mouth of computing machine, and described indicator lamp and sound equipment are used for alerting signal that real-time receiving computer sends and send sound and light alarm according to alerting signal respectively.

Described image pick-up card is the FPGA image pick-up card.

The utility model also comprises following other technologies feature:

Described image pick-up card is the FPGA image pick-up card.

The utility model will replenish mutually based on the information of Region Segmentation and the information of rim detection, improved the precision that lane boundary is extracted, for can detected track, robust ground under the road environment of multiple complexity, the distance of utilizing vehicle centre-line and track, the left and right sides is reported to the police but actv. carries out deviation as alarm criteria.

Description of drawings

Fig. 1 is the structural representation of vehicle lane deviation warning device of the present utility model.

Fig. 2 is a system equipment interface scheme drawing.

Fig. 3 is the workflow diagram of warning device of the present utility model.

Fig. 4 is the diagram of circuit that carries out lane detection in the warning device working process.

Fig. 5 is that the lane shift geometric distance concerns scheme drawing.Wherein the label among the figure is represented respectively: 1, camera, 2, image, 3, boundary line, the left side, track.

The specific embodiment

As shown in Figure 1, warning device of the present utility model is made up of vision collecting and disposal system, warning:

Vision collecting and disposal system be used for collection vehicle position and direction graphicinformation and detect the track; This system comprises camera, image pick-up card, computing machine, and the mouth of image pick-up card connects the computing machine input end, and the mouth of computing machine connects warning; The graphicinformation of camera real-time collection vehicle position and direction, and deposit the graphicinformation that collects in image pick-up card, image pick-up card is real-time transmitted to computing machine with graphicinformation, computer real-time is handled the graphicinformation that receives, and sends alerting signal according to result to warning;

Warning is made up of indicator lamp and sound equipment, and the input end of sound equipment and indicator lamp is connected the mouth of computing machine, and indicator lamp and sound equipment are used for alerting signal that real-time receiving computer sends and send sound and light alarm according to alerting signal respectively.

Described image pick-up card is the FPGA image pick-up card.Sound equipment and gauge panel light that sound and light alarm adopts vehicle itself to carry.Computing machine adopts the Y460 of association notebook PC, processor model Intel Pentium Dual Core P6000, internal memory 2G.

During practical application, camera is installed in the anterior centre of operator's compartment position, camera links to each other with image pick-up card, computing machine links to each other with image pick-up card, computing machine is connected with auto audio equipment with gauge panel light, computing machine use MATLAB with the image processing software bag handle the image that is collected.

As shown in Figure 2, what the interface of system's connection camera was used is Camera Link interface, by Camera Link interface the realtime graphic high speed information is transferred to and carries out data in real time in the FPGA image pick-up card and handle, and realize communication between capture card and the computing machine by pci interface.

As shown in Figure 3 and Figure 4, workflow of the present utility model is as follows:

When the vehicle that the deviation warning device is housed when advance in the track, camera deposits the graphicinformation of the road surface characteristic of collection in image pick-up card in real time, image pick-up card sends to computing machine in real time with graphicinformation, by computing machine the image that collects is handled, computing machine adopts lane detection to obtain the position and the direction information in track, and compare, discern with setting value, extract the difference factor, when the difference factor surpasses threshold value, system reports to the police by acousto-optic, and the concrete workflow of this device is as follows:

The first step: lane detection; Determine lane position by lane detection, this lane detection method has provided the direction angle in the locus and the track in track, can be applied directly in the judgement to vehicle and track relative position.The track is extracted and is undertaken by following steps:

1) extracts the lane boundary template

Utilize colouring information to carry out Region Segmentation and obtain the lane boundary template, promptly utilize the information in difference, brightness and the colourity of the brightness in hsv color space and degree of saturation specific humidity of each pixel to set up Gauss model respectively, carry out cluster, the Region Segmentation result who merges on three components determines the zone, track, at last the mathematical morphology operation is carried out in the zone, track and is obtained lane boundary extraction template.Its step is expressed as follows:

(1) on the colourity color component, carries out Region Segmentation;

The image of a given width of cloth m * n utilizes formula 1 to carry out the S as a result of Region Segmentation on the colourity color component _HFor:

(formula 1)

Wherein: (i, j) expression pixel coordinate; H (i, j) expression pixel (i, chromatic value j); u _HAnd σ _HAverage and the variance of representing chromatic component respectively, u _HAnd σ _HUtilize formula 2, formula 3 to calculate respectively:

$u_H=\frac{1}{\mathrm{mm}}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{j=1}{\overset{n}{\mathrm{\Σ}}}H(i,j)$ (formula 2)

${\mathrm{\σ}}_H={\left(\frac{1}{\mathrm{mn}}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{(H(i,j)-u_H)}^2\right)}^{1/2}$ (formula 3)

In the formula, D is for putting the letter parameter, D=2 in the present embodiment;

(2) on the brightness color component, carry out Region Segmentation;

Utilize formula 4 on the brightness color component, to carry out Region Segmentation S as a result _V:

(formula 4)

Wherein: (i, j) expression pixel coordinate; V (i, j) expression pixel (i, chromatic value j); u _VAnd σ _VAverage and the variance of representing luminance component respectively, u _VAnd σ _VUtilize formula 5, formula 6 to calculate respectively:

$u_V=\frac{1}{\mathrm{mn}}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{j=1}{\overset{n}{\mathrm{\Σ}}}V(i,j)$ (formula 5)

${\mathrm{\σ}}_V={\left(\frac{1}{\mathrm{mn}}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{(V(i,j)-u_V)}^2\right)}^{1/2}$ (formula 6)

In the formula, V (i, j) expression pixel (i, chromatic value j);

(3) utilize the difference information of brightness and degree of saturation specific humidity to carry out Region Segmentation;

According to brightness value and intensity value, utilize formula 7 to calculate difference C:

C (i, j)=| V (i, j)-S (i, j) | (formula 7)

Wherein, S (i, j) expression pixel (i, intensity value j);

According to formula 8, utilize the S as a result of the difference information Region Segmentation of brightness and degree of saturation specific humidity _C:

(formula 8)

Wherein: $u_C=\frac{1}{\mathrm{mn}}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{j=1}{\overset{n}{\mathrm{\Σ}}}C(i,j)$ (formula 9)

${\mathrm{\σ}}_C={\left(\frac{1}{\mathrm{mn}}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{(C(i,j)-u_C)}^2\right)}^{1/2}$ (formula 10)

(4) utilize formula 11 to adopt the logical "or" computing to merge S _CAnd S _V, obtain preliminary road area segmentation result L _S:

(formula 11)

(5) utilize formula 12 to adopt the logical computing to merge L _SAnd S _H, obtain road area segmentation result L:

(formula 12)

(6) scanning obtains track template M _L

At first with the Region Segmentation result from the center to both sides, scan from bottom to top, every scanning line mind-set both sides from image, first pixel value of running into of record is 0 point respectively, forms the template M in track from the center to the pixel of this point _L

(7) utilize formula 13 to carry out the mathematical morphology operation and obtain lane boundary template E ₁

$E_1=(M_L\⊕b)-\left(M_L\mathrm{\Θb}\right)$ (formula 13)

Wherein, b=[1] _{1 * 1}The expression structural element; Θ represents the mathematical morphology erosion operation;

Expression mathematical morphology dilation operation;

2) detect the road edge point

Adopt the Sobel operator to carry out rim detection respectively on colourity, brightness, three color components of degree of saturation specific humidity, merge the edge detection results on three components, obtain the lane detection result based on the edge, step is as follows:

(1) adopt the Sobel operator to carry out rim detection respectively on V, S, H component, edge detection results is designated as E respectively _V, E _SAnd E _H

(2) utilize formula 14 to adopt the logical "or" computings to merge edge detection results on three components, obtain road-edge detection E as a result ₂, formula is expressed as follows:

(formula 14)

3) obtain the lane boundary point

Integration region segmentation result and edge detection results promptly utilize formula 15 to adopt logical computing integration region to cut apart the lane boundary template E that obtains ₁With rim detection E ₂Obtain lane boundary point E:

(formula 15)

4) use the Hough conversion to determine lane line

Use the Hough conversion to determine the boundary line in track to the lane boundary point that obtains;

The Hough conversion is a kind of method of representative type detection of straight lines, and its main thought is that the problem with detection of straight lines changes at the Hough parameter space and seeks peak point, supposes that the straight line of unknown boundary position in the image can be described as with equation:

Xcos (θ)+ysin (θ)=r (formula 16)

Belong to any point (x on this straight line, y) can be at parameter space (r, θ) be described as a sine curve, point on the same straight line corresponding to the same point in the parameter space (r, θ), by detecting (r, cumulative distribution θ) is tried to achieve has peaked (r, θ) point, and then try to achieve parametric equation, promptly obtain the parametric description of this straight line.

Second step: deviation is reported to the police

1, the relative lane position of vehicle determines

As shown in Figure 5, suppose that camera is perpendicular to the ground, the controlled variable departure distance is determined by formula (17):

$\frac{f}{f+H}=\frac{\mathrm{\Δ}{b}_l}{\mathrm{\Δ}{B}_l}$ (formula 17)

Wherein the camera focal length is represented with f (mm), and the camera lens front end represents that with H (mm) track left margin line imaging end points is apart from the transverse distance of vehicle center Δ b on the camera projection plane apart from ground-surface height _lExpression, Δ B _lBe track left margin line imaging end points correspondence position on the ground with lateral excursion distance vehicle centre-line reality.K (mm/ pixel) expression camera plane and frame are deposited the ratio that concerns between resolution (pixel).s _lIt is the lateral excursion distance of track left margin line imaging end points and picture centre line.Then utilize formula 18 can get departure distance to be:

$\mathrm{\Δ}{B}_l=(1+\frac{H}{f})\×k\×s_l$ (formula 18)

In like manner, can obtain the lateral excursion of boundary line, the right, track and vehicle centre-line apart from Δ B _r

2, deviation is reported to the police

Judge according to formula 19, as Δ B _lOr Δ B _rDuring less than given threshold value Δ B, computing machine sends alarm control signal to indicator lamp and sound equipment, and vehicle and sound equipment carry out sound and light alarm according to the alarm control signal that receives,

Δ B _r≤ Δ B or Δ B _l≤ Δ B (formula 19)

Wherein, the value of Δ B is relevant with the width that the warning device vehicle is installed, if the width of vehicle is W, the value of Δ B should be W/2.

Claims (2)

1. the deviation warning device of a vehicle on highway is characterized in that, this device is made up of vision collecting and disposal system, warning:

Vision collecting and disposal system be used for collection vehicle position and direction graphicinformation and detect the track; This system comprises camera, image pick-up card, computing machine, and the mouth of described image pick-up card connects the computing machine input end, and the mouth of described computing machine connects warning; The graphicinformation of described camera real-time collection vehicle position and direction, and deposit the graphicinformation that collects in image pick-up card, image pick-up card is real-time transmitted to computing machine with graphicinformation, computer real-time is handled the graphicinformation that receives, and sends alerting signal according to result to warning;

Warning is made up of indicator lamp and sound equipment, and the input end of described sound equipment and indicator lamp is connected the mouth of computing machine, and described indicator lamp and sound equipment are used for alerting signal that real-time receiving computer sends and send sound and light alarm according to alerting signal respectively.

2. vehicle on highway deviation warning device as claimed in claim 1 is characterized in that described image pick-up card is the FPGA image pick-up card.

Images