KR20090131824A - Image type wandering measurement system and traffic parameter computing method using thereof - Google Patents

Abstract

PURPOSE: An image type wandering measurement system and a traffic parameter computing method using the same are provided to obtain a speed automatic correction value using the difference between the vehicle obtained through an image processing of a digital camera and the vehicle obtained by the piezo sensor buried under the road. CONSTITUTION: A digital camera(100) is installed in an upper part of the road and obtains the image of the vehicle traveling on the road. A piezo sensor(200) is buried on the road of an image sensing region(11) obtaining the image of the vehicle by the digital camera and senses the load applied to the wheels of the vehicle. The piezo sensor is installed to cross from an edge of an entrance(A) of the image sensing region with a fixed angle diagonally without crossing the road direction vertically. An operator(300) determines whether the vehicle passes through the image sensing region using a wheel load signal transmitted from the piezo sensor and the digital image transmitted from the digital camera.

Description

Image Type Wandering Measurement System and Traffic Parameter Computing Method using same

The present invention is installed on the top of the lane digital camera 100 for acquiring the image of the vehicle traveling on the road; And a piezoelectric sensor for detecting a load acting on a wheel of a vehicle, which is embedded in an oblique direction (constant angle θ) on a lane in an image detection area 11 through which the digital camera 100 acquires an image of the vehicle. It relates to an image-type wandering measurement device for calculating a traffic parameter using the (200); and a traffic parameter calculation method using the same.

Currently used image detectors acquire digital images from digital cameras installed on the road, and general image detectors that detect the presence and speed of traffic through image processing of digital images, such as fog or In bad weather conditions, images cannot be acquired properly, and the performance is significantly degraded, which can be useless, and even if it is not a musical instrument, there is a high possibility of error due to the shadow of a vehicle passing through adjacent lanes.

For example, as shown in FIG. 1, when the shadow of a large vehicle passing through the fourth lane occurs in the third lane, the conventional image detector recognizes the shadow of the third lane moving at the same speed as the vehicle of the fourth lane as a separate vehicle. You can make an error.

In addition, when relying on the existing image detector, since the classification of the vehicle depends only on the length of the vehicle, there are only three types of distinguishable vehicles, and the accuracy thereof is inferior, and the load of the passing vehicle cannot be measured.

In the case of the existing buried WIM equipment installed so that the axis detection sensor (piezo sensor) crosses the lane vertically (perpendicular to the direction of movement of the lane), the error of recognizing the shadow as a vehicle can be prevented and the load by the axis of the vehicle Measurement is possible, but it is impossible to measure the load of each wheel of the passing vehicle, but only the load of both wheels attached to the same axle.

The object of the present invention created to solve the above problems is as follows.

First, it is an object of the present invention to provide a means for minimizing the added sensor and at the same time significantly complement the shortcomings of the existing image detector.

Second, another object of the present invention is to provide a measuring device capable of calculating various traffic parameters and classifying vehicle types in various ways.

Third, another object of the present invention is to provide a means for obtaining a speed automatic correction value by using a difference between a lubrication obtained using a piezo sensor embedded on a road and a digit obtained through image processing of a digital camera. .

Fourthly, another object of the present invention is to provide a means for estimating the traffic volume per unit time for each lane even in a musical instrument.

Fifth, another object of the present invention is to provide a means for transmitting information on a speed or a load of a vehicle to a vehicle driver driving a road using a road sign.

Sixth, the purpose of this study is to obtain and analyze the mean and standard deviation values for the frequency of passage of vehicle wheels by road by collecting the wandering data of vehicles used as road pavement design factors.

The configuration of the present invention created to achieve the above object is as follows.

The present invention is installed on the top of the lane digital camera 100 for acquiring the image of the vehicle traveling on the road; A piezo sensor 200 which detects a load acting on a wheel of a vehicle, which is buried on a road in the image detection area 11 through which the digital camera 100 acquires an image of the vehicle; And determining whether the vehicle passes through the image detection area 11 by using the digital image transmitted from the digital camera 100 and the wheel load signal transmitted from the piezo sensor 200, and passing through image processing. Computing unit 300 for calculating the speed of the vehicle to be configured, wherein the piezo sensor 200 is not installed to traverse perpendicularly to the direction of the lane, the predetermined angle (θ) in the image detection area 11 for each lane It is characterized in that it is installed across the diagonal line as much as).

Technical effects according to the configuration of the present invention are as follows.

First, it is possible to minimize the added sensor and at the same time significantly compensate for the shortcomings of the existing image detector.

In other words, the piezo sensor 200 for detecting the load of the vehicle wheel is installed separately from the digital camera 100 for detecting each lane to solve the problem of misunderstanding the shadow of the vehicle driving the surrounding lane as a separate vehicle. Can be.

Second, it is possible to calculate various traffic parameters and classify the vehicle more diversely.

In other words, in the conventional image sensor, only three types of vehicles can be classified and classified with respect to the vehicle length based on the length of the vehicle. However, in the present invention, the digital camera 100 detects each lane and By using the piezo sensor 200 buried in the diagonal direction of the lane, the traffic parameters such as vehicle speed, vehicle lease, wheelbase, left margin of the vehicle, right margin of the vehicle, and speed automatic correction value are more accurately By using the various traffic parameters, the vehicle types may be classified in more detail, and the traffic volume according to the vehicle types may be accurately calculated.

Third, the speed automatic correction value can be obtained by correcting the speed of the vehicle by the image processing by using the difference between the lubrication acquired by using the piezo sensor embedded on the road and the image obtained by image processing of the digital camera. .

Fourth, it is possible to estimate the traffic volume per unit time by lane even at all times.

In other words, dividing the cumulative axis per unit time for each lane by the traffic volume per unit time for each lane to obtain the axis correction coefficient for each lane, if the traffic detection using the digital camera 100 is impossible on the instrument, the piezo sensor 200 The amount of traffic per unit time for each lane can be calculated by dividing the cumulative number of axes per unit time for each lane detected by.

Fifth, information about the speed or the load of the vehicle can be transmitted to the vehicle driver driving the road using the road sign.

In other words, the information about the speed and the load of the vehicle calculated by the calculating unit 300 may be transmitted to the road sign 110 and transmitted to the driver of the vehicle to prevent the accident.

Hereinafter, specific embodiments of the present invention will be described in detail with reference to the accompanying drawings.

FIG. 2 is a perspective view showing the overall configuration of the present invention, and FIG. 3 shows factors for calculating the image detection area 11, the embedded state of the piezo sensor 200, the left and right lanes, and the traffic parameters in individual lanes. It is a top view.

As shown in FIG. 2, the present invention includes a digital camera 100, a piezo sensor 200, and a calculation unit 300, and a road light sign 400 may be added as necessary.

The digital camera 100 is installed on one side of the lane and acquires an image of the vehicle moving on the lane and transmits the image to the operation unit 300. The pillar is installed around the lane and the digital camera 100 is held. It is installed on the upper portion of the image detection area 11, or not shown separately in the drawing, but may be installed in a structure that crosses the upper portion of the lane like a speeding prevention camera.

One or more digital cameras 100 are not necessarily installed, and two or more digital cameras 100 may be installed in consideration of the area (length and width) of the image detection area 11.

However, since it is possible to acquire images of multiple lanes with one digital camera 100, a separate digital camera 100 is not necessarily required for each lane.

The digital image acquired by the digital camera 100 is transmitted to the calculator 300 in a wired or wireless manner and subjected to image processing by an algorithm embedded in the calculator 300.

The specific process of such image processing is similar to that described in the prior art, and a separate description thereof is omitted.

Piezo sensor 200 is buried on the road surface of each lane as in the case of the conventional buried detector detects the axle of the vehicle passing through it and transmits the detected signal to the calculation unit 300 in a wired or wireless manner. One piezo sensor 200 is provided in each lane, and unlike the conventional buried type detector, the piezo sensor 200 of the present invention is located in the image detection area 11 as shown in FIG. It is installed in the direction of the entrance (A) side of the image detection area 11 without being installed transversely perpendicular to the lane direction It is installed to cross a diagonal line by a certain angle (θ).

Although only one lane is shown in FIG. 3, even when there are other lanes adjacent to each other in the same direction, the same direction as in FIG. 3 is installed in the image detection area 11 so as to traverse diagonally by the same angle θ.

As such, when the piezo sensor 200 is installed diagonally, as shown in FIG. 3, as the vehicle proceeds, the right wheel of the vehicle first passes through the piezo sensor 200, and then the left wheel of the vehicle is It passes through the piezo sensor 200.

That is, the left wheel and the right wheel do not pass through the piezo sensor 200 at the same time, but the wheel located on the right side in the drawing passes through the piezo sensor 200 first, and using the unusual configuration of the present invention As described below, only one digital camera 100 and the piezo sensor 200 may be used to accurately calculate various traffic parameters.

The calculation unit 300 recognizes that the actual vehicle is on the lane according to the signal transmitted from the piezo sensor 200, and the pixel value inside the image detection area 11 through the digital camera 100. Even if a change is detected, if the wheel load signal is not transmitted from the piezo sensor 200, the vehicle passes through the corresponding lane.

In this way, by combining the signals of the digital camera 100 and the piezo sensor 200, it is possible to prevent the error that the shadow of the vehicle passing through the peripheral lane in the conventional image detector was recognized as a separate vehicle.

The calculation unit 300 is generated when the digital image transmitted from the digital camera 100 and the wheel load signal transmitted from the piezo sensor 200 (the left wheel and the right wheel pass through the piezo sensor 200, respectively). To determine whether the vehicle passes through the image detection region 11 and extract various traffic parameters.

The calculation unit 300 classifies the digital image transmitted from the digital camera 100 and extracts a feature to classify the presence or absence of the vehicle and the vehicle type, and calculates the speed of the vehicle. Various traffic parameters for the vehicle may be calculated by combining the signals transmitted from the piezo sensor 200.

The calculation unit 300 may calculate the weight of the vehicle by summing the loads of the wheels transmitted from the piezo sensor 200, and the information about the wheels is combined with the speed of the vehicle through the road light sign 400. It can be delivered to the driver of the vehicle.

A specific method for calculating various traffic parameters using a specific embodiment of the present invention as shown in FIG. 3 is as follows. The calculation unit 300 is detected by the digital camera 100 and the piezo sensor 200. After receiving the signal, the traffic parameter is calculated as follows according to the algorithm stored therein.

(1) Speed of vehicle by image processing (applied to non-instrumental)

When the vehicle enters the image detection area 11 where image processing is performed by using the digital camera 100 and the calculation unit 300, the time T0 when the vehicle reaches the entrance A of the image detection area 11, The vehicle detects the time T1 at which the vehicle reaches the exit B of the image detection region 11 and calculates the speed of the vehicle using the following relational expression.

V = D / ( T0  - T1 )

V: vehicle speed by image processing

D: distance (m) between the entrance (A) and the exit (B) of the image detection area (11)

T0: time when the vehicle reaches the entrance A of the image detection area 11 (sec)

T1: time (sec) when the vehicle reaches the exit B of the image detection area 11

(2) Right margin of the vehicle

The distance between the right wheel of the vehicle and the right lane is obtained using the following relationship using the angle θ between the piezo sensor 200 and the vertical line crossing the lane and the speed of the vehicle by image processing.

The average value and standard deviation of these data are used as design factors for calculating the thickness for future pavement design along with the left margin of the vehicle.

L3  = D1  Of tan (θ)

   = V x ( T0  - T11 ) Of tan (θ)

L3: Right margin of the vehicle (m)

θ: angle between the piezo sensor 200 and the vertical line crossing the lane

V: vehicle speed by image processing

T0: time when the vehicle reaches the entrance A of the image detection area 11 (sec)

T11: time when the right wheel of the first axis of the vehicle stepped on the piezo sensor 200 (sec)

      That is, the signal generation time (sec) of the first axis sensor

(3) Vehicle Ethics

The distance between the left wheel and the right wheel of the vehicle (vehicle vehicle) is calculated using the piezo sensor 200 as follows.

L2  = ( D1  + D2 ) Of tan (θ)- L3

   = V x (( T12  - T0 ) Of tan (θ)- L3

L2: Vehicle Arbitration (m)

V: vehicle speed by image processing

T12: Time when the left wheel of the first axis of the vehicle stepped on the piezo sensor 200 (sec)

T0: time when the vehicle reaches the entrance A of the image detection area 11 (sec)

θ: angle between the piezo sensor 200 and the vertical line crossing the lane

(4) Left margin of the vehicle

The left margin of the vehicle is obtained from the following relation using the distance between the right and left lanes of the lane, the right margin and the vehicle lease.

L1  = L-( L2  + L3  )

L1: Left margin of vehicle (m)

L: Distance between right and left lanes (m)

L2: Vehicle Arbitration (m)

L3: Right margin of the vehicle (m)

(5) wheelbase

The wheelbase, which is the distance between the axles of the vehicle (between the vehicle's axles), is obtained from the following equation.

Ds = V x 1/2 x [( T11  - T21  ) + ( T12  - T22  )]

Ds: Wheelbase (m)

V: vehicle speed by image processing

T11: time when the right wheel of the first axis of the vehicle stepped on the piezo sensor 200 (sec)

T12: Time when the left wheel of the first axis of the vehicle stepped on the piezo sensor 200 (sec)

T21: Time when the right wheel of the 2nd axis of the vehicle stepped on the piezo sensor 200 (sec)

T22: time when the left wheel of the second axis of the vehicle stepped on the piezo sensor 200 (sec)

(6) Speed automatic correction value

A speed automatic correction value for correcting the speed of the vehicle by the image processing by using the vehicle width Li by the image processing and the vehicle lubrication L2 using the piezo sensor 200 is obtained as follows.

Speed automatic correction value = Li  Of L2

L2  = ( D1  + D2 ) Of tan (θ)- L3

   = V x (( T12  - T0 ) Of tan (θ)- L3

V: vehicle speed by image processing

T12: Time when the left wheel of the first axis of the vehicle stepped on the piezo sensor 200 (sec)

T0: time when the vehicle reaches the entrance A of the image detection area 11 (sec)

θ: angle between the piezo sensor 200 and the vertical line crossing the lane

L3: Right margin of the vehicle (m)

(7) Traffic volume per unit time per vehicle lane

If good images are not learned on musical instruments (e.g., heavy fog, heavy rain, etc.), image processing is impossible or very inaccurate. Therefore, by dividing the cumulative axis per unit time by lane by the traffic volume per unit time by lane to obtain the axis correction coefficient for each lane, if the detection of the traffic volume using the digital camera 100 is impossible on the instrument detected by the piezo sensor 200 The amount of traffic per unit time for each lane is calculated by dividing the cumulative number of axes per unit time for each lane.

(8) Instrument speed

In order to calculate the vehicle speed at the time of the instrument, before the vehicle by using the above formula (L2 = V x ((T12-T0) / tan (θ)-L3) for the vehicle lubrication L2) The average value of the vehicle lease L2 (average lease for each lane) is calculated and stored.

If the vehicle's speed cannot be calculated by the constant image processing of the instrument, replace the vehicle with an average vehicle by storing the L2 in the following equation for the vehicle entrepreneur (L2), and replace V with the vehicle's constant vehicle speed. Create the following relationship and use it to calculate the vehicle speed.

tan (θ) = D2 / L2

D2 = L2 x tan (θ)

V x (T12-T11) = L2 x tan (θ)

V = [L2 x tan (θ)] / (T12-T11)

On the instrument, L2 cannot be calculated because the speed of the vehicle cannot be calculated through image processing. Therefore, V, which is obtained by substituting the average entrepreneurs for each lane stored in L2, is estimated as the speed of the vehicle. The relation is as follows.

Musical instrument  Vehicle speed = [ tan (θ) x lane Average Ethics ] / ( T12  - T11 )

As described above, when the traffic parameters related to the speed of the vehicle, the vehicle lease, the number of axles, the wheelbase (distance), the vehicle length, etc. are calculated using the algorithm built into the operation unit 300, the vehicle types are classified by using this information. Traffic volume by vehicle type can also be calculated.

In Korea, regardless of the type of road, the car is divided into 12 types of classification system, and information about the vehicle length, axle number, wheelbase, etc., which is a variable for classifying the car in advance, is inputted, and the digital camera 100 and the piezo sensor ( Compared to the traffic parameter calculated using the signal detected from the 200, it is possible to easily classify the vehicle.

The calculation unit 300 also includes a built-in algorithm for classifying a vehicle, and in the case of a two-axis vehicle having two vehicle axes, a vehicle type classification using a vehicle length (full length), a wheelbase (between wheelbase), and a wheelbase ratio (1-4 types) The algorithm is shown in FIG.

As described above, the present invention has been described with reference to specific embodiments of the present invention, but the protection scope of the present invention is not necessarily limited to these embodiments, and various design changes and notifications are made within the scope of not changing the technical gist of the present invention. In the case of addition or deletion of technology, and simple numerical limitations, it is obvious that the scope of the present invention is included.

1 illustrates a problem of a conventional image detector, which shows that a shadow of a vehicle passing through a surrounding lane can be recognized as a separate vehicle.

2 is a perspective view showing the overall configuration of the present invention.

3 is a plan view showing factors for calculating the image detection area 11, the buried state of the piezo sensor 200, the left and right lanes, and the traffic parameters in the individual lanes.

4 is a flowchart of an algorithm for classifying vehicle types.

<Explanation of symbols for the main parts of the drawings>

100: digital camera

200: Piezo sensor

300: calculation part

400: road sign

11: Image detection area

A: Entrance

B: exit

Claims (11)

A digital camera 100 installed at an upper portion of the lane and acquiring an image of the vehicle moving on the lane; A piezo sensor 200 which detects a load acting on a wheel of a vehicle, which is buried on a road in the image detection area 11 through which the digital camera 100 acquires an image of the vehicle; And, Vehicle passing through the image detection area 11 by using the digital image transmitted from the digital camera 100 and the wheel load signal transmitted from the piezo sensor 200 determines whether the vehicle passes through the image processing Computing unit 300 for calculating the speed of the; It is configured to include, The piezo sensor 200 is not installed transversely perpendicular to the direction of the lane, but is installed at an inlet (A) side edge of the image detection area 11 so as to traverse diagonally by a predetermined angle (θ). Visual wondering measurement device. In claim 1, The piezo sensor 200 detects a load acting on each wheel of the vehicle, The calculating unit (300) is a visual wonder measuring device, characterized in that to calculate the weight of the vehicle by summing the load for each wheel transmitted from the piezo sensor (200). In claim 2, A road light sign 400 which receives one or more of data on the speed of the vehicle passing through and the weight of the vehicle passing from the calculating unit 300 and provides visual information to the driver of the vehicle; The visual wandering measuring apparatus further comprises. A digital camera 100 installed at an upper portion of the lane and acquiring an image of the vehicle moving on the lane; A piezo sensor 200 which detects the axle of the vehicle passing through the digital camera 100 embedded in the lane in the image detection area 11 for acquiring an image of the vehicle; And determining whether the vehicle passes through the image detection area 11 by using the digital image transmitted from the digital camera 100 and the axle detection signal transmitted from the piezo sensor 200, and passing through the image processing. Computing unit 300 for calculating the speed of the vehicle to be configured, and relates to a traffic parameter calculation method using the wandering measurement device, When the vehicle enters the image detection area 11 where image processing is performed by using the digital camera 100 and the calculation unit 300, the time T0 when the vehicle reaches the entrance A of the image detection area 11. ), And detects the time T1 at which the vehicle reaches the exit B of the image detection region 11, V = D / (T0-T1) V: vehicle speed by image processing D: distance (m) between the entrance (A) and the exit (B) of the image detection area (11) T0: time when the vehicle reaches the entrance A of the image detection area 11 (sec) T1: time (sec) when the vehicle reaches the exit B of the image detection area 11 A method for calculating a traffic parameter using a visual wandering measuring device, characterized in that the speed (V) of the vehicle by image processing is calculated using a relational expression. In claim 4, L3  = D1  Of tan (θ)    = V x ( T0  - T11 ) Of tan (θ) L3: Right margin of the vehicle (m) θ: angle between the piezo sensor 200 and the vertical line crossing the lane V: vehicle speed by image processing T0: time when the vehicle reaches the entrance A of the image detection area 11 (sec) T11: time when the right wheel of the first axis of the vehicle stepped on the piezo sensor 200 (sec)       That is, the signal generation time (sec) of the first axis sensor A method for calculating a traffic parameter using a visual wandering measuring device, characterized in that the distance (L1) between the right wheel and the right lane of the vehicle is calculated using a relational expression. In claim 5, L2  = ( D1  + D2 ) Of tan (θ)- L3    = V x (( T12  - T0 ) Of tan (θ)- L3 L2: Vehicle Arbitration (m) V: vehicle speed by image processing T12: Time when the left wheel of the first axis of the vehicle stepped on the piezo sensor 200 (sec) T0: time when the vehicle reaches the entrance A of the image detection area 11 (sec) θ: angle between the piezo sensor 200 and the vertical line crossing the lane A vehicle parameter calculation method using a visual wandering measurement device, characterized in that for calculating the vehicle lease (L2) using a relational expression. In claim 6, L1  = L-( L2  + L3  ) L1: Left margin of vehicle (m) L: Distance between right and left lanes (m) L2: Vehicle Arbitration (m) L3: Right margin of the vehicle (m) A method for calculating a traffic parameter using a visual wandering measuring device, characterized in that the left margin of the vehicle (L1) is calculated using a relational expression. The method according to any one of claims 4 to 7, Ds = V x 1/2 x [(T11-T21) + (T12-T22)] Ds: Wheelbase (m) V: vehicle speed by image processing T11: time when the right wheel of the first axis of the vehicle stepped on the piezo sensor 200 (sec) T12: Time when the left wheel of the first axis of the vehicle stepped on the piezo sensor 200 (sec) T21: Time when the right wheel of the 2nd axis of the vehicle stepped on the piezo sensor 200 (sec) T22: time when the left wheel of the second axis of the vehicle stepped on the piezo sensor 200 (sec) A method for calculating a traffic parameter using a visual wandering measuring device, characterized in that for calculating the wheelbase (Ds) which is the distance between the wheels of the vehicle using a relational equation. The method according to any one of claims 4 to 7, Speed automatic correction value  = Li  Of L2 L2  = ( D1  + D2 ) Of tan (θ)- L3    = V x (( T12  - T0 ) Of tan (θ)- L3 V: vehicle speed by image processing T12: Time when the left wheel of the first axis of the vehicle stepped on the piezo sensor 200 (sec) T0: time when the vehicle reaches the entrance A of the image detection area 11 (sec) θ: angle between the piezo sensor 200 and the vertical line crossing the lane L3: Right margin of the vehicle (m) A method for calculating a traffic parameter using a visual wandering measuring device, characterized in that the speed automatic correction value is calculated using a relational equation. The method according to any one of claims 4 to 7, The axis correction coefficient for each lane is obtained by dividing the cumulative number of axes per unit time by lane by the traffic volume per unit time by lane. When it is impossible to detect the traffic volume using the digital camera 100 on a musical instrument, the unit is divided by the axial correction coefficient for each lane which has previously calculated the cumulative axis per unit time for each lane detected by the piezo sensor 200. A traffic parameter calculation method using an visual wandering measuring device, characterized in that the traffic volume per hour is estimated. In claim 6, In order to calculate the speed of the vehicle at all times, the vehicle lease (L2) by car in advance by using the equation L2 = V x ((T12-T0) / tan (θ)-L3 for the vehicle entrepreneur (L2) Calculate and store the average value (average berth by lane) for If the vehicle's speed (V) cannot be calculated by the instrumental constant image processing, replace the vehicle with the average vehicle by the pre-stored L2 in the equation below for the vehicle entrepreneur (L2). In place of tan (θ) = D2 / L2 D2 = L2 x tan (θ) V x (T12-T11) = L2 x tan (θ) V = [L2 x tan (θ)] / (T12-T11) Musical instrument  Vehicle speed = [ tan (θ) x lane Average Ethics ] / ( T12  - T11 ) A method for calculating a traffic parameter using a visual wandering measuring device, comprising: inducing a relational expression and calculating the vehicle speed at all times using the same.

Images