KR101448506B1 - Measurement Method and Apparatus for Measuring Curvature of Lane Using Behavior of Preceding Vehicle - Google Patents

Abstract

A method and a device for measuring the radius of curvature of a running lane using the behavior of the preceding vehicle are disclosed.
An object extraction unit for detecting at least one preceding vehicle in the image data received from the imaging device; A feature extraction unit for extracting dynamic characteristic information of the at least one preceding vehicle detected by the object extraction unit; And a curvature radius calculation unit calculating a curvature radius of the driving lane based on the dynamic characteristic information extracted by the feature extraction unit.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to a method and apparatus for measuring a curvature of a traveling lane using a behavior of a preceding vehicle,

The present embodiment relates to a method and an apparatus for measuring the radius of curvature of a running lane using the behavior of the preceding vehicle. More particularly, the present invention relates to a method and an apparatus for measuring a curvature of a running lane that calculates a curvature of a lane using an image of a preceding vehicle collected by a vehicle equipped with a vision camera.

The contents described in this section merely provide background information on the present embodiment and do not constitute the prior art.

Generally, an unmanned vehicle has a system for recognizing the driving environment and performing an automatic driving function. Of these, the driving environment that should be noted is the curve section where accidents occur frequently. This is why it is important to predict the radius of curvature of the vehicle before entering the curve to implement the techniques such as Lane Keeping Assist System (LKAS) or Adaptive Cruise Control (ACC) to be.

According to the prior art, there is a problem in which the curvature of the lane is recognized by recognizing the lane by the vision camera of the vehicle, but it is not applicable in an environment where there is no lane or in which it is difficult to recognize the lane (when the lane is blocked by the preceding vehicle) .

Therefore, there is a demand for a technique of predicting the curvature of a lane by using the surrounding environment other than the lane.

The main object of the present invention is to provide a method and an apparatus for measuring the radius of curvature of a running lane using the behavior of the preceding vehicle by solving the problem that it is difficult to predict the curvature by recognizing the running lane in an environment in which there is no lane or in which it is difficult to recognize the lane .

According to an aspect of the present embodiment, there is provided an image processing apparatus including: an object extraction unit that detects at least one preceding vehicle from image data received from an imaging device; A feature extraction unit for extracting dynamic characteristic information of the at least one preceding vehicle detected by the object extraction unit; And a curvature radius calculation unit calculating a curvature radius of the driving lane based on the dynamic characteristic information extracted by the feature extraction unit.

According to another aspect of the present invention, there is provided a method of calculating a radius of curvature of a driving lane using image information of a preceding vehicle, comprising the steps of: collecting the image information with respect to the preceding vehicle; Detecting an edge of the preceding vehicle in the image information and extracting an edge image of the preceding vehicle; Calculating a steering angle of the preceding vehicle based on an edge image of the preceding vehicle; And calculating the radius of the driving lane based on the steering angle, and outputting the calculation result.

According to another aspect of the present invention, there is provided a method of calculating a radius of curvature of a driving lane using image information of a preceding vehicle, comprising the steps of: collecting the image information with respect to the preceding vehicle; Detecting an edge of the preceding vehicle in the image information and extracting an edge image of the preceding vehicle; Calculating a straight line distance and an azimuth angle with respect to the preceding vehicle based on an edge image of the preceding vehicle; Calculating a position of the preceding vehicle based on the straight line distance and the azimuth angle; And calculating a radius mainly based on the position of the preceding vehicle.

As described above, according to the present embodiment, by analyzing the behavior of the preceding vehicle, it is possible to calculate the radius of curvature of the running lane even on a road without a lane, thereby providing safe driving and accident prevention effects.

1 is a side view of a vehicle equipped with a curvature radius calculation device including a vision camera according to an embodiment of the present invention.
2 is a configuration diagram of a curvature radius calculation apparatus according to an embodiment of the present invention.
3 is a diagram illustrating a method of calculating a steering angle based on a vehicle-width distance of a preceding vehicle in a steering angle calculating unit according to an embodiment of the present invention, and a method of calculating a curvature of a radius calculating unit.
4 is a diagram illustrating a method of calculating a steering angle based on a rear wheel input image of a preceding vehicle in a steering angle calculating unit according to an embodiment of the present invention.
5 is a diagram illustrating a method of calculating a radius mainly based on a straight line distance and an azimuth of a preceding vehicle and a subject vehicle in a straight line distance and azimuth extracting unit, a leading vehicle position calculating unit, and a mainly radius calculating unit according to an embodiment of the present invention Fig.
6 is a diagram illustrating a method of extracting straight line distances and azimuth angles of the straight line distance and azimuth extracting unit according to an embodiment of the present invention.
7 is a flowchart illustrating a method of calculating a radius of curvature of a curvature radius calculation apparatus according to an embodiment of the present invention.

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

It should be noted that, in adding reference numerals to the constituent elements of the drawings, the same constituent elements are denoted by the same reference symbols as possible even if they are shown in different drawings. In the following description of the present invention, a detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present invention rather unclear.

In describing the components of the present invention, terms such as first, second, A, B, (a), and (b) may be used. These terms are intended to distinguish the constituent elements from other constituent elements, and the terms do not limit the nature, order or order of the constituent elements. When a component is described as being "connected", "coupled", or "connected" to another component, the component may be directly connected to or connected to the other component, It should be understood that an element may be "connected," "coupled," or "connected."

1 is a side view of a vehicle equipped with a curvature radius calculation device including a vision camera according to an embodiment of the present invention.

The curvature radius calculation device receives an image of the preceding vehicle 130 through a vision camera 110 attached to the vehicle 120 and calculates a radius of curvature of the road based on the behavior of the preceding vehicle 130 do.

The vision camera 110 is a concept that includes an imaging device that can convert a forward image into an electrical signal, as well as receive a forward reflected signal such as an ultrasound wave, an infrared ray, an ultraviolet reflection signal, and the like, The vision camera 110 may be mounted inside or outside the child vehicle 120 and may be installed in a room mirror. Since the curvature radius calculation device is installed in the child vehicle 120, there is no particular restriction on the position where the curvature is calculated by receiving the image, so that the curvature is provided to the vehicle by receiving the image information remotely .

The embodiment of the curvature radius calculation apparatus described in this specification is different in that the curvature of the road is calculated based on the behavior of the preceding vehicle but the operation of the curvature radius calculation apparatus differs according to the numerical value for quantifying the behavior of the preceding vehicle in the image . Variables related to the behavior described in the present specification are a change in the width direction of the preceding vehicle, a change in the shape of the tire, a straight line distance from the preceding vehicle, and an azimuth angle. The turning radius calculated using each variable is calculated independently. In the present embodiment, a method and an apparatus for calculating the turning radius based on the average value of the turning radius calculated for each variable will be described, but may be implemented as independent embodiments for each variable.

The term " steering of the preceding vehicle " used herein means a state in which the traveling direction of the preceding vehicle turns on the basis of the traveling direction of the vehicle, and "steering angle" Respectively. The term "azimuth angle" refers to the angle at which the preceding vehicle appears on the input image based on the traveling direction of the subject vehicle. "Straight distance" means a straight line distance between the host vehicle and the preceding vehicle.

The curvature radius calculation device detects an envelope of the image received from the vision camera 110 and extracts an edge image. The curvature radius calculator divides an object from an edge image and determines whether or not the vehicle is steered based on the presence or absence of lateral acceleration of the azimuth angle or the amount of the vertical component of the edge image. Further, the radius-of-curvature calculator determines whether or not the lateral acceleration exists in the same lane based on whether or not the sign of the lateral acceleration changes.

In the case of a straight ahead vehicle, only the back side of the preceding vehicle is visible, so that the vertical component of the edge image is detected for the rear left and right of the vehicle. However, since the side face of the vehicle is detected at the time of turning, do. Therefore, for accurate edge detection, the curvature radius calculator tracks the edge histogram of the vertical component of the preceding vehicle.

The curvature radius calculation device calculates the steering angle of the preceding vehicle based on the shape of the wheel and the width of the preceding vehicle among the images of the preceding vehicle. The radius of curvature calculator calculates the radius of curvature based on the change of the steering angle and the distance traveled by the preceding vehicle.

The radius of curvature calculator calculates the radius of curvature by specifying the position of the preceding vehicle on the basis of the azimuth and straight line distances of three or more preceding vehicles and assigning them to the general solution of the circle equation.

The radius of curvature calculator can calculate the average value of the radius of curvature calculated on the basis of the shape of the wheel, the width of the preceding vehicle, and the azimuth angle and the straight line distance of the preceding vehicle.

2 is a configuration diagram of a curvature radius calculation apparatus according to an embodiment of the present invention.

The object extracting unit 210 extracts an edge image by detecting an envelope of an image received from an imaging device such as the vision camera 110. [ The object extracting unit 210 may identify an object from an edge image and assign an ID to each preceding vehicle when the object is determined to be a vehicle. In the case of the preceding vehicle traveling straight ahead, since only the back side is visible, the vertical component of the edge image is detected corresponding to the width of the vehicle, but when the vehicle is turned left or right, the side image of the vehicle is detected. Therefore, A plurality of vertical components are detected. Accordingly, the object extraction unit 210 stores and tracks the edge histogram of the vertical component of the preceding vehicle for edge detection of the vehicle width.

The feature extraction unit 220 extracts the dynamic characteristic information of the preceding vehicle based on the extracted object image of the preceding vehicle. In this case, the dynamic characteristic refers to a characteristic relating to the movement of the preceding vehicle, and may be information on the width of the preceding vehicle, the ratio of the height and width of the wheel, the azimuth angle, and the straight line distance. The ratio of the height to the width of the wheel can be calculated by extracting the edge of the wheel on the image and comparing the length. The feature extraction unit 220 can calculate the straight line distance through the y coordinate of the preceding vehicle when the image is represented by the orthogonal coordinate system. The feature extraction unit 220 can calculate the azimuth angle through the x-coordinate of the preceding vehicle when the image is expressed in the orthogonal coordinate system. The feature extraction unit 220 can calculate the input width at the same distance through the width of the edge and the straight line distance in the image.

The curvature radius calculation unit 230 calculates the curvature radius of the driving lane using the dynamic characteristic information of the preceding vehicle. The curvature radius calculation unit 230 includes a steering angle calculation unit 240, a first curvature calculation unit 250, a preceding vehicle position calculation unit 260, a second curvature calculation unit 270, and a final curvature calculation unit 280 . And may further include a filtering unit (not shown).

The radius of curvature calculator 230 calculates the steering angles of the first and second timings using the ratios of the widths and heights of the wheels respectively input at the first and second timings, The curvature can be calculated by calculating the moving distance of the preceding vehicle using the time between the two points of view and the speed of the preceding vehicle. Here, the first viewpoint and the second viewpoint refer to two different viewpoints for calculating the steering angle. The radius of curvature calculation unit 230 calculates the radius of curvature of the preceding vehicle by using the width of the preceding vehicle measured at the straight line and the width of the preceding vehicle at the same distance as the distance measured at the straight line, The curvature can be calculated by calculating the steering angle of the vehicle and calculating the moving distance of the preceding vehicle using the time between the first and second points of time and the speed of the preceding vehicle. Also, the curvature radius calculation unit 230 may calculate the curvature of the preceding vehicle by calculating the curvature radius calculation unit 230 using the azimuth angle and the straight line distance of the preceding vehicle. Meanwhile, the radius of curvature calculator 230 may output the average of the curvatures calculated by one or more methods as the final curvature.

The filtering unit (not shown) is positioned between the feature extracting unit 220 and the radius of curvature calculator 230. Before the radius of curvature calculator 230 receives the dynamic characteristic, It is possible to determine whether the steering wheel is steered based on whether the vertical component of the curvature radius is increased or not, and to transmit the start signal of the radius of curvature calculation to the radius of curvature calculator 230. It is also possible to determine whether to stop the operation of the object of the corresponding ID or remove the data of the object based on whether the sign of the lateral acceleration changes or not. If the sign of the lateral acceleration changes, it can be judged that the turning direction of the vehicle has changed, so that it is removed from the curvature calculation of the same lane. If the dynamic characteristic has a value larger than a preset threshold value, it is determined that the object is not a curved path, and a signal is sent to the curvature radius calculation unit 230 to stop the operation of the object of the corresponding ID or to remove the object data.

The steering angle calculation unit 240 calculates the steering angle using at least one of the vehicle width and the linear distance in the linear principal of the preceding vehicle extracted from the feature extraction unit 220 or the ratio of the height and width of the wheel. A concrete calculation method of the steering angle will be described later with reference to Figs.

The first curvature calculation unit 250 may calculate the lane radius using the change of the steering angle of the preceding vehicle with respect to time or may calculate the lane radius using the straight line distance and the azimuth angle of the preceding vehicle. The method of calculating the radius of the vehicle using the change of the steering angle of the preceding vehicle will be described first. The difference between the steering angle of the preceding vehicle and the steering angle of the preceding vehicle at the second point of time and the difference between the steering angle of the preceding vehicle and the first point The curvature is calculated using the moving distance l of the preceding vehicle.

The preceding vehicle position calculation unit 260 receives the straight line distance and the azimuth angle of the preceding vehicle from the feature extraction unit 220 and outputs the rectilinear distance and the azimuth angle of the preceding vehicle as the y- The position of the preceding vehicle is calculated as the coordinate value. It is possible to transmit the calculated information to the second curvature calculation unit 260 only when the number of preceding vehicles is three or more. A specific method of calculating the curvature using the position of the preceding vehicle will be described later with reference to Fig.

If there are three points on the coordinate plane, the circle is determined. The second curvature calculation unit 270 substitutes the coordinates of the three points from the preceding vehicle position calculation unit 260 in the general solution of the original equation to calculate the lane radius. Therefore, the curvature is also determined. A specific method will be described later with reference to Fig.

The final curvature calculator 280 may output a curvature value of an average value of at least one of a curvature value calculated by the first curvature calculator 250 and a curvature value calculated by the second curvature calculator 270. [

3 is a diagram illustrating a method of calculating a steering angle and a curvature based on a vehicle-width distance of a preceding vehicle in a steering angle calculating unit and a first curvature calculating unit according to an embodiment of the present invention.

The preceding vehicle indicated by V1 is a diagram showing the position and the steering angle of the preceding vehicle at the first time point, and the preceding vehicle indicated by V2 is the position and steering angle of the preceding vehicle at the second time point.

The steering angle calculation unit 240 calculates the steering angle using the width of the preceding vehicle and the width of the preceding vehicle measured in accordance with the steering and the straight distance with the preceding vehicle when the linear distance is moved. At this time, the ratio between the actual width of the preceding vehicle and the width measured at the time of steering is used. However, since the distance between the vehicle and the preceding vehicle may vary depending on the viewpoint, and the magnitude of the observed width varies when the distances are different, it is necessary to calculate the magnitude of the width to be observed at the same distance. In order to calculate the width of the preceding vehicle to be observed, when the ratio between the straight line distance from the preceding vehicle and the straight line distance at the time of steering is accumulated, The size can be calculated by converting it to correspond to the size of the image of the preceding vehicle at the time of straight forward.

When the width W of the preceding vehicle and the width W 'of the preceding vehicle at the time of steering are obtained in this way, the steering angle calculation unit 240 calculates the steering angle? ₁ at the first time as shown in Equation (1).

Where W 'is the width of the preceding vehicle at the time of steering, W is the width of the preceding vehicle at the time of going straight, and θ ₁ is the steering angle at the first time.

In the same way, the steering angle calculation section 240 can obtain the steering angle? ₂ at the second time point.

The steering angle calculation unit 240 transmits the respective steering angles calculated at the first and second points of time to the first curvature calculation unit 250.

First curvature calculating unit 250 calculates the radius of the car using the travel distance l to the second point on the steering angle θ _2, the first point of the steering angle θ ₁ and the second point of the first point. At this time, the radius R of the driving lane is derived as shown in Equation (2).

Here, dθ is moved from a first point in time with respect to the circle corresponding to the curvature to a second point, a center angle, l is the moving distance of the preceding vehicle at a first point to a second point in time, R is the radius of the circle corresponding to the curvature, θ ₁ Denotes a steering angle at a first time point, and? ₂ denotes a steering angle at a second time point.

The travel distance of the preceding vehicle can be calculated using the time between the first and second points of time and the speed of the preceding vehicle and the speed of the preceding vehicle can be approximated to the speed of the subject vehicle.

4 is a diagram illustrating a method of calculating a steering angle based on a rear wheel input image of a preceding vehicle in a steering angle calculating unit according to an embodiment of the present invention.

The driver manipulates the desired steering angle using the front wheel, but the steering angle of the vehicle body is determined by the angle of the rear wheel not involved in the steering. Therefore, in this embodiment, a method of calculating the curvature based on the steering angle of the rear wheel will be described. However, various methods such as a method of calculating the direct curvature using the angle difference between the rear wheel and the front wheel will be possible.

As the steering angle changes, the ratio between the height and width of the wheel changes. Using this, the steering angle calculation unit 240 can calculate the steering angle. Because the wheels of a car are all round, the height and the actual width are the same length. Therefore, the height of the wheel (= the actual width of the wheel) when the width of the wheel corresponds to W 'during steering is equivalent to W in FIG. If there is a difference, the rear side of the vehicle body and the wheel are perpendicular to each other, so that the phase difference is 90 degrees and the ratio of the height and the width is constant regardless of the straight distance (because the height and width of the wheel at the same position are compared) There is no need for correction according to the present invention. Therefore, when calculating the steering angle using the shape of the wheel, information on the straight line distance is unnecessary. 3, when the steering angle is calculated based on the shape of the wheel, the steering angle calculation unit 240 calculates the steering angle with the ratio of the width W of the wheel to the height h.

When the widths of the wheels at the first time point, the second time point and the third time point are respectively W ₁ , W ₂ and W ₃ and the height of the wheel measured at each time point is h, the steering angle calculation section 240 calculates the steering angle θ ₃ , ? ₄ ,? ₅ can be calculated using the following equation (3).

Where W ₁ is the width of the observation wheel in the first point in time, W ₂ is the width of the observation wheel in the second time point, W ₃ is the width of the observation wheel in the third point, θ ₃ is the steering angle, θ of the first point, ₄ is the steering angle at the second time point,? ₅ is the steering angle at the third time point, and h is the height of the wheel observed at each time point.

The difference between the equation (1) and the equation (1) is 90 degrees, because the direction of the vehicle width and the direction of the wheel are arranged vertically in one vehicle. The steering angle of the two or more viewpoints thus calculated is transmitted to the first curvature calculation unit 250. It is possible to simultaneously transmit the steering angles of two or more preceding vehicles to the first curvature calculation unit 250 according to the ID of the preceding vehicle displayed in one frame.

The first curvature calculation unit 250 calculates the curvature using Equation (2). In this case, θ ₁ denotes a steering angle at a first point in time, and θ ₂ denotes a steering angle at a second point in time. The method of calculating the curvature based on this has already been described. However, θ ₁ and θ ₂ are not necessarily the steering angles of the first and second points. When the steering angle of two or more preceding vehicles is input in one frame from the steering angle calculation unit 240, the distance between the preceding vehicles may be calculated and directly calculated based on the coordinates received from the preceding vehicle position calculation unit 260.

5 is a diagram illustrating a method of calculating a lane radius based on a straight line distance and an azimuth angle of a preceding vehicle and a child vehicle in a preceding vehicle position calculating unit and a second curvature calculating unit according to an embodiment of the present invention.

Origin and Points If three coordinates are specified, you can define a circle that passes through three points. The preceding vehicle position calculation unit 260 receives the linear distance and the azimuth angle from the feature extraction unit 220 to the three or more preceding vehicles to which the ID is assigned. Based on the extracted straight line distances and azimuths, the preceding vehicle position calculation unit 260 specifies the coordinates of each of the preceding vehicles in an orthogonal coordinate system having the y direction of the vehicle running direction and the position of the vehicle as its origin. At this time, the preceding vehicle A, B, as the distance from _{C l 1, l 2, l} 3 , and when referred to as the azimuth angle of the preceding vehicle A, B, C θ _6, θ _7, θ ₈ the preceding vehicle A, B, C (4), and the preceding vehicle position calculation unit 260 calculates the coordinates of the preceding vehicle position using Equation (4).

A is the coordinate of the preceding vehicle A, B is the coordinate of the preceding vehicle B, C is the coordinate of the preceding vehicle C, r3 is the distance to the preceding vehicle A, r2 is the distance to the preceding vehicle B, The distance,? ₈ is the azimuth angle of the preceding vehicle A,? ₇ is the azimuth angle of the preceding vehicle B, and? ₆ is the azimuth angle of the preceding vehicle C.

The equation of a circle having a radius of R with the center (a, b) in the orthogonal coordinate system as shown in Equation (5).

Where a is the x coordinate of the center of the circle, b is the y coordinate of the center of the circle, and R is the radius of the circle.

The radius calculation unit 240 mainly substitutes the coordinates of the preceding vehicles calculated by the preceding vehicle position calculation unit 260 using Equation 4 into the formula of Equation 5 to calculate the original equations . The radius R and the curvature are calculated using this.

Here, although a method of calculating using the orthogonal coordinate system has been described, a method of directly calculating the straight line in the polar coordinate system by omitting the leading vehicle position calculation unit 260 by applying the polar coordinate system can be easily considered.

FIG. 6 is a diagram illustrating a method of confirming dynamic characteristics in an input image according to an exemplary embodiment of the present invention. Referring to FIG.

6A is an exemplary view showing an angle formed by a point at which the vision camera of the subject vehicle 610 contacts the ground when the image including the preceding vehicle is input to the height line 660 of the vision camera. The point at which the first preceding vehicle 620 abuts the ground and the point 640 equidistant from the vision camera and the straight line are received at the same angle ₁₀ and displayed on the image at the same height. A point 650 at which the second preceding vehicle 630 also contacts the ground and equidistant from the vision camera and the straight line distance is received at the same angle ₁₁ and is displayed at the same height on the image. To illustrate the height line 660, the preceding vehicle is illustrated as a vehicle of the same height.

Figure (b) is a screen in which the reception image shown in Figure (a) actually appears. Since the vehicle height is the same, the preceding vehicle appears at the height of the height line 660, but the point at which the preceding vehicle touches the ground approaches the height line 660 as the distance increases. Therefore, the feature extracting unit 220 can measure the straight line distance between the preceding vehicle and the child vehicle by measuring the distance between the height line indicated in the image and the point contacting the ground of the preceding vehicle. In the same way, the azimuth angle can be calculated using the deviation of the preceding vehicle from the center line dividing the image vertically. It is also possible to calculate the ratio of the width of each vehicle to the height and width of the wheel based on the edge image.

7 is a flowchart illustrating a method of calculating a radius of curvature of a curvature radius calculation apparatus according to an embodiment of the present invention.

Vision cameras are used to recognize the preceding vehicle. At this time, the edge image can be extracted based on the envelope of the image to identify the object, and IDs can be assigned to the identified objects (S710).

The presence of transverse acceleration in the image confirms that the recognized preceding vehicle changes direction. At this time, the histogram of the vertical edge can be stored (S710).

If the direction is changed, the steering angle of the preceding vehicle is calculated using the change in the width of the preceding vehicle. The steering angle can be calculated by using the ratio of the measured widths using the image based on the calculated widths of the edge histograms. The steering angle is measured at two or more intervals (S730).

Calculate the radius of curvature lane based on the steering angle. Since the steering angle is the moving center angle and the moving distance is the moving distance of the arc, the radius of the curvature difference is calculated using the difference between the traveling distance and the steering angle during the time difference (S740).

The steering angle is calculated using the tire shape of the preceding vehicle. The ratio of the height and width of the tire shown in the image is calculated by a trigonometric function to calculate the steering angle (S750).

Calculate the radius of curvature lane based on the steering angle. The curvature of the lane can be calculated through the steering angle and the travel distance calculated with the tires. The method is the same as in S740.

It is checked whether the number of object recognition of the preceding vehicle whose direction is changed in the sub lane is 3 or more (S770).

3 or more, the position of the preceding vehicle is calculated based on the straight line distance and the azimuth of the preceding vehicle. The straight line distance and the azimuth angle are determined based on the position of the preceding vehicle with respect to the height line or center axis shown in the image. The position of the preceding vehicle is indicated as coordinates in an orthogonal coordinate system having the subject vehicle as the origin and the traveling direction of the subject vehicle as the y-axis (S780).

Calculate the radius of curvature difference based on the position of the preceding vehicle. The center and radius of the circle are calculated by substituting the coordinates of the preceding vehicle into the equation of the circle in the rectangular coordinate system (S790).

The average value of the calculated radius of curvature is calculated and output (S800).

7, it is described that steps S710 to S800 are sequentially executed. However, this is merely an illustrative description of the technical idea of the embodiment of the present invention, and it is not intended to limit the scope of the present invention to the general knowledge in the technical field to which the embodiment of the present invention belongs. Those skilled in the art will appreciate that various modifications and adaptations may be made to those skilled in the art without departing from the essential characteristics of one embodiment of the present invention by changing the order described in Figure 7 or by executing one or more of steps S710 through S800 in parallel 7, it is not limited to the time-series order. Also, although S710 and S720 are common, the radius of the road is calculated in steps S730 and S740, S750 and S760, and the radius of the road is calculated with S770 to S790 alone, so that an alternative relationship can be obtained. Therefore, they can be performed independently of each other. However, two or three of these methods can be used at the same time.

The foregoing description is merely illustrative of the technical idea of the present embodiment, and various modifications and changes may be made to those skilled in the art without departing from the essential characteristics of the embodiments. Therefore, the present embodiments are to be construed as illustrative rather than restrictive, and the scope of the technical idea of the present embodiment is not limited by these embodiments. The scope of protection of the present embodiment should be construed according to the following claims, and all technical ideas within the scope of equivalents thereof should be construed as being included in the scope of the present invention.

110: vision camera 120:
130: a preceding vehicle 210: a vision camera
220: leading vehicle image extracting unit 230: steering angle calculating unit
240: Mainly radius calculation unit 250: Linear distance and azimuth extraction unit
260: preceding vehicle position calculation unit 610:
620: first preceding vehicle 630: second preceding vehicle
640: Equidistance line up to the first preceding vehicle 650: Equidistant line up to the second preceding vehicle
660: Height line

Claims (16)

An object extraction unit for detecting at least one preceding vehicle in the image data received from the imaging device;
A feature extraction unit for extracting dynamic characteristic information of the at least one preceding vehicle detected by the object extraction unit; And
A curvature radius calculation unit for calculating a curvature radius of the driving lane based on the dynamic characteristic information extracted by the feature extraction unit,
, ≪ / RTI &
The radius of curvature calculation unit may calculate the radius of curvature of each of the plurality of preceding vehicles based on the vehicle width variation of the preceding vehicle at a plurality of points of time, the ratio of the height and width of the wheels of the preceding vehicle at a plurality of points of time, And calculating the radius of curvature by using at least one of positions on an orthogonal coordinate system of the preceding vehicle at a time point. The method according to claim 1,
Wherein the dynamic characteristic information includes at least one of a vehicle width of the at least one preceding vehicle, a height of the preceding vehicle, a wheel image change of the preceding vehicle, an azimuth angle with the preceding vehicle, and a straight line distance with the preceding vehicle Wherein the curvature radius calculation device is characterized by: The method according to claim 1,
The feature extraction unit may extract,
When the sign of the lateral acceleration of the preceding vehicle changes, when the vehicle width of the preceding vehicle and the height of the preceding vehicle suddenly change more than a predetermined degree, and when the wheel image change of the preceding vehicle suddenly changes at a predetermined degree or more Further comprising a filtering unit which excludes the preceding vehicle from the calculation object when any one or more of the preceding vehicles is included in the calculation. The method according to claim 1,
Wherein the radius of curvature calculation unit comprises:
Further comprising a steering angle calculation section for calculating a steering angle based on at least any one of a vehicle width change rate of the preceding vehicle and a ratio of a height and a width of the wheel of the preceding vehicle. 5. The method of claim 4,
The steering angle calculation unit calculates,
The vehicle width measured after the preceding vehicle is measured on the basis of the reference vehicle width of the preceding vehicle observed when the preceding vehicle is in a straight line and the straight line distance between the preceding vehicle and the preceding vehicle, And calculating a steering angle. 5. The method of claim 4,
Wherein the radius of curvature calculation unit comprises:
A steering angle of the first viewpoint and a steering angle of the second viewpoint are received and the difference between the steering angle of the first viewpoint and the steering angle of the second viewpoint is calculated, and the time difference between the first viewpoint and the second viewpoint Further comprising a first curvature radius calculation unit for calculating a radius of curvature by calculating a movement distance of the preceding vehicle calculated based on the time and the speed of the preceding vehicle. delete The method according to claim 1,
Wherein the radius of curvature calculation unit comprises:
And a second curvature radius calculation unit for calculating a radius of curvature based on the coordinates when the coordinates of the preceding vehicle are calculated as three or more. The method according to claim 1,
The straight line distance from the preceding vehicle,
Height line with the preceding vehicle displayed on the image and the distance between the point of contact with the ground of the preceding vehicle. The method according to claim 1,
Wherein the radius of curvature calculation unit calculates a radius of curvature calculated by using a ratio of a height and a width of a wheel of the preceding vehicle at a plurality of points of time, Further comprising a final curvature radius calculation unit for calculating an average value obtained by averaging two or more curvature radiuses among the radius curvatures calculated using the position of the preceding vehicle at the time point of the previous curvature radius, as the calculated value of the final radius of curvature. A method of calculating a curvature radius of a driving lane using image information of a preceding vehicle,
Collecting the image information with respect to the preceding vehicle;
Detecting an edge of the preceding vehicle in the image information and extracting an edge image of the preceding vehicle;
Calculating a steering angle of the preceding vehicle based on an edge image of the preceding vehicle; And
Calculating and outputting the radius of the driving lane based on the change in the steering angle of the preceding vehicle with respect to time
And calculating a radius of curvature of the curved surface. 12. The method of claim 11,
The method for calculating the steering angle includes:
And calculating a steering angle based on a change in the width of the edge image. 12. The method of claim 11,
The method for calculating the steering angle includes:
Calculating a steering angle based on a ratio of a width and a height of the wheel portion to a wheel portion of the edge image. A method of calculating a curvature radius of a driving lane using image information of a preceding vehicle,
Collecting the image information with respect to the preceding vehicle;
Detecting an edge of the preceding vehicle in the image information and extracting an edge image of the preceding vehicle;
Calculating a straight line distance and an azimuth angle with respect to the preceding vehicle based on an edge image of the preceding vehicle;
Calculating a position of the preceding vehicle based on the straight line distance and the azimuth angle; And
Calculating a radius of curvature of the running lane based on at least three positions of the preceding vehicle with respect to time;
And calculating a radius of curvature of the curved surface. 15. The method of claim 14,
Calculating a straight line distance and an azimuth from the preceding vehicle,
Calculating the straight line distance and the azimuth angle based on a position on the screen of the edge image of the preceding vehicle. 15. The method of claim 14,
Calculating a position of the preceding vehicle,
Calculating a coordinate value of the preceding vehicle in an orthogonal coordinate system with the y axis of the traveling direction of the vehicle and the origin of the vehicle as the origin.

Images