KR101278237B1 - Method and apparatus for recognizing vehicles - Google Patents

Abstract

A vehicle recognition method and apparatus are provided. In a vehicle recognition method according to an embodiment of the present invention, a vehicle recognition method for recognizing an opposing vehicle by analyzing an image acquired from a camera, the method comprising: (a) a region of interest at a position defined in the acquired image; Setting); (b) detecting a portion having a specific brightness or more in the ROI and setting it as a target light source; obtaining a motion vector representing a temporal motion of the target light source; And (d) recognizing whether the target light source is a counter vehicle based on the motion vector.

Description

Vehicle recognition method and apparatus {Method and apparatus for recognizing vehicles}

The present invention relates to a vehicle recognition method and apparatus, and more particularly, to a vehicle recognition method and apparatus for recognizing a vehicle against a driving vehicle by analyzing an image acquired by a camera.

In general, a camera module for acquiring a front image includes a lens and an image sensor. Here, the lens collects the light reflected from the subject, and the image sensor detects the light collected by the lens and converts the light into an electrical image signal. The image sensor may be largely composed of an image tube and a solid state image sensor, and examples of the solid state image sensor may include a charge coupled device (CCD) and a metal oxide semiconductor (MOS).

In addition, the vehicle has a headlamp that secures the driver's vision at night by irradiating light in the same direction as the driving direction, and various technologies for acquiring a front image of the vehicle using the camera module and recognizing the vehicle against the vehicle have been developed. It is becoming.

However, the conventional counter vehicle recognition technology uses a pixel pair in consideration of two headlights of a counter vehicle, and has a limitation in recognizing a counter vehicle in which headlights such as motorcycles are not paired.

Therefore, regardless of whether the front vehicle is detected according to the pixel value or the pixel size or the headlight of the opposing vehicle is the pixel pair, the front object is obtained at night and the image processing for the image is performed. Sensing technology is required.

In addition, when the vehicle turns on the headlamp at night, the driver of the preceding vehicle or the opposite vehicle positioned in front of the driving direction may cause glare and obstruct the view, thereby increasing the possibility of a vehicle accident. In order to solve the problem, it is necessary to obtain a road environment information through the image of the driving road in front of the vehicle and to solve the problem by rotating the headlamp according to the road environment information. There is a need for a system that automatically controls the headlamps of a vehicle, such as an Adaptive Front-Light System (AFLS) that can secure the driver's vision by controlling the pattern.

The present invention is to solve the above problems, to provide a vehicle recognition method for acquiring an image to determine the light source of a specific region as a vehicle candidate and recognizes the light source that the motion vector of the light source passes in a specific direction or a specific region as the vehicle. It is.

Another object of the present invention is to provide an apparatus for recognizing a light source as a counter vehicle by acquiring an image, determining a light source in a specific region as a counter vehicle candidate, and comparing the light source with the database of movement patterns of the counter vehicle.

In addition, the vehicle can recognize the opposite vehicle to determine its position, and thereby adjust the light blocking area in the headlamp of the driving vehicle or implement the optimal light irradiation pattern by rotating the headlamp to maximize the driver's field of view. To provide vehicle recognition technology.

Problems to be solved by the present invention are not limited to the above-mentioned problems, and other problems not mentioned will be clearly understood by those skilled in the art from the following description.

A vehicle recognition method according to an embodiment of the present invention for achieving the above object, in the vehicle recognition method for recognizing the opposite vehicle by analyzing the image obtained from the camera, (a) at a position defined in the acquired image Establishing a region of interest; (b) detecting a portion having a specific brightness or more in the ROI and setting it as a target light source; obtaining a motion vector representing a temporal motion of the target light source; And (d) recognizing whether the target light source is a counter vehicle based on the motion vector.

According to another aspect of the present invention, there is provided a vehicle recognition apparatus comprising: a storage unit configured to store a database of movement patterns of a counter vehicle; An image acquisition unit for acquiring an image in front of the vehicle; A region of interest setting unit configured to set a region of interest in the acquired image; A light source setting unit configured to detect light in the ROI and set it as a target light source; A light source tracking unit which tracks the target light source using a motion vector; And a vehicle determination unit comparing the movement pattern of the stored counter vehicle with the movement pattern of the target light source to determine whether the target light source is the counter vehicle.

Other specific details of the invention are included in the detailed description and drawings.

According to the present invention, since the counter vehicle can be recognized without using the pixel pair of the acquired image in front of the vehicle, a vehicle having a single light source such as a motorcycle or a vehicle whose one head lamp is broken can be recognized as the counter vehicle. have.

1 is a flow chart of a vehicle recognition method according to an embodiment of the present invention.
2 is a diagram illustrating the operation of a vehicle against a vehicle ahead.
3 is a diagram illustrating a region of interest (ROI) set in an image window.
4 is a diagram illustrating a movement pattern of a light source of a vehicle against an image window.
5 is a flowchart illustrating a vehicle recognition method according to another embodiment of the present invention.
6 is a block diagram of a vehicle recognition apparatus according to an embodiment of the present invention.
7A is a diagram illustrating a movement pattern of a light source of a vehicle against a straight road.
7B is a diagram illustrating a movement pattern of a light source of a vehicle against a curved road.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS The advantages and features of the present invention and the manner of achieving them will become apparent with reference to the embodiments described in detail below with reference to the accompanying drawings. The present invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art. Is provided to fully convey the scope of the invention to those skilled in the art, and the invention is only defined by the scope of the claims. Like reference numerals refer to like elements throughout.

Unless otherwise defined, all terms (including technical and scientific terms) used in the present specification may be used in a sense that can be commonly understood by those skilled in the art. Also, commonly used predefined terms are not ideally or excessively interpreted unless explicitly defined otherwise.

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

1 is a flow chart of a vehicle recognition method according to an embodiment of the present invention, Figure 2 is a view showing the operation of the vehicle against the front.

In a vehicle recognition method of recognizing an opposing vehicle by analyzing an image acquired from a camera, a region of interest is set at a position defined in the acquired image (S110), and has a specific brightness or more in the region of interest. A part is detected and set as a target light source (S120), a motion vector representing a temporal motion of the target light source is obtained (S130), and the target light source is determined to be recognized based on the motion vector to recognize it. S140).

In particular, when obtaining a motion vector representing the temporal movement of the target light source, the method further includes storing the position value of the target light source every frame. When determining whether the target light source is the counter vehicle based on the motion vector, the target light source is recognized as the counter vehicle when the connection line connecting the stored position values passes a predetermined reference line.

The driving vehicle 10 acquires an image of the front subject, that is, the counter vehicle 11 through an image acquisition device such as a camera module. Here, a charge coupled device (CCD) module or a complementary metal oxide semiconductor (CMOS) module may be used as the imaging device. The obtained image may be converted into a compressed format in which the image is compressed to facilitate data transmission. Image data in the form of a compressed format may have various formats such as Moving Picture Experts Group (MPEG) -1 or MPEG-4.

The opposing vehicle 11 moves in a direction opposite to the driving vehicle 10, and the vehicle passes on the right side. Therefore, the movement of the light source of the opposing vehicle 11 shows a pattern from the upper left to the lower left with respect to the driving vehicle 10.

3 is a diagram illustrating a region of interest (ROI) set in a front image.

An image acquired through a camera module or the like may be expressed in a two-dimensional plane, which is defined as an image window 100. That is, the image window 100 serves as a window in which the acquired image is displayed. And, as in the case of Figure 3, the horizontal axis can be represented by the X-axis, the vertical axis can be represented by the Y-axis.

The region of interest is set at a predetermined defined position in the image obtained from the camera. When setting a region of interest (ROI) 120 in the acquired image (S110), the region of interest 120 is set to a rectangular shape including vanishing lines. Therefore, the ROI 120 is set inside the image window 100 in which an image is displayed. Here, the vanishing line is a line in which the light source 110 of the counter vehicle 11 disappears.

The light source 110 of the opposing vehicle 11 will be visible for the first time near the vanishing line in the ROI 120, and then the light source 110 will be shown in sequence just below the vanishing line. In addition, it is preferable to obtain the positions of at least three consecutive light sources 110 in accordance with the speed of the opposing vehicle 11. This is to derive the motion vectors MV 130 of the light source 110 of the vehicle 11.

4 is a diagram illustrating a movement pattern of a light source of a vehicle against an image window.

When a portion having a specific brightness or more in the ROI 120 is detected and set as the target light source 110, it is set as the target light source 110 when the light is within a predetermined reference luminance range. That is, light of a portion having a specific brightness or more in the region of interest should be within a predetermined reference luminance range.

When the vehicle 10 is operated at night, not only the vehicle 11 but also the light from the street lamp or the building may be detected in the image. Therefore, in order to recognize only the light from the headlight of the opposing vehicle 11 as the target light source 110, it is necessary to set only the light within the predetermined reference luminance range among the light detected in the region of interest 120 as the target light source 110. have. In this case, a predetermined reference luminance range may be determined using the brightness of the lamp set in the vehicle.

After the portion having a specific brightness or more detected in the region of interest 120 is set as the target light source 110 (S120), the target light source 110 is obtained using the motion vectors 130, MV, and Motion Vector. In this case, the target light source may be tracked only when the size of the motion vector 130 is greater than or equal to a predetermined threshold value. The light set as the target light source 110 in the region of interest 120 becomes a strong candidate for the counter vehicle 11, and the target light source 110 can be tracked by the motion vector 130. That is, it is possible to recognize whether the target light source 110 is a counter vehicle based on the motion vector, and the tracking is performed using the motion vector.

In the image window 100, the target light source 110 in the ROI 120 moves from the upper left to the lower left, and since the image is generally composed of 30 frames per second, the target light source 110 also moves a certain position. The display can be displayed in the image window 100. That is, the motion vector 130 value may be displayed on the image window 100 by displaying the movement of the position of the target light source 110 obtained for each frame with an arrow.

Thus, as shown in FIG. 4, the target light source 110 can be displayed as a motion vector 130 starting at a starting point near the vanishing line and moving along an arrow. In FIG. 4, two motion vectors 130 are shown, indicated by three light source positions and connected by three arrows.

In order to obtain the motion vector 130, a block matching algorithm (BMA) is generally used. BMA is a method of dividing an image into blocks of a predetermined size and expressing all pixels in the block as one motion vector 130. In order to find the motion vector 130, the blocks of the previous frame are moved by one pixel to find the most similar block to the current block. In BMA, motion estimation is performed under the premise that the deformation due to movement between adjacent frames in time is not so large that only all movement can be approximated by the movement of an object. Therefore, the motion vector 130 is represented by one x and y coordinate for each block.

The motion vector 130 may be obtained by using Equation 1 below using Mean Absolute Distance (MAD), which is one of BMA measurement methods.

Here, C (k, l) represents a pixel of the block, and R (i + k, j + l) represents a reference window pixel. The motion vector 130 may find and obtain a pixel having the smallest MAD value obtained from Equation 1 in a reference window.

When the target light source 110 is tracked using the motion vector 130 (S130), the target light source 110 may be tracked only when the size of the motion vector 130 is greater than or equal to a predetermined threshold value. have. The motion vector 130 is obtained from Equation 1, and the magnitude of the motion vector 130 is simply obtained.

Since the target light source 110 of the ROI 120 is generated by the counter vehicle 11, the size of the motion vector 130 is also affected by the speed of the counter vehicle 11. Therefore, by using this, only when the size of the motion vector 130 is greater than or equal to a predetermined threshold, it is determined as a candidate of the vehicle 11, and only the target light source 110 whose size of the motion vector 130 is greater than or equal to the predetermined threshold is tracked. You can do that.

Then, the motion vector 130 is obtained and the target light source 110, which is a candidate for the opposing vehicle 11, is tracked using the motion vector 130 (S130), and the position value is stored every frame. In this case, the position value may be displayed as an X-Y coordinate value with the image or the image window as the X-Y plane. As shown in FIGS. 3 and 4, an image displayed on the image window 100 may be displayed on a two-dimensional plane of the X-Y axis. Thus, the positional information of the target light source 110 may be stored as an X-Y coordinate value in an image in which a predetermined frame is captured every second.

In FIG. 4, it can be seen that the position values of the last frame of the motion vector 130 of the left target light source 110 and the immediately preceding frame of the last frame are represented by (a1, b1) and (a, b), respectively. When the position values of (a1, b1) and (a, b) are connected, the magnitude and direction of the motion vector 130 in the last frame and the frame immediately before the last frame can be known.

When the connection line connecting the stored position values passes a predetermined reference line 140, the target light source may be determined as a counter vehicle. That is, it is possible to determine whether the target light source 110 is the counter vehicle 11 based on the motion vector 130. The connection line connecting the stored position values is the motion vector 130 displayed on the image window 100, and the motion vector 130 passing through the predetermined reference line 140 is opposed to and approaches the driving vehicle 10. It can be seen as.

Here, the reference line 140 is set at the lower left half of the ROI 120. This is because, in the case of right traffic, the opposing vehicle 11 moves from the upper left of the driving vehicle 10 to the lower left. Of course, in the case of the left passage, the reference line 140 may be set at the right bottom of the region of interest 120.

In the case of Figure 4, the reference line 140 is a line connecting (x1, y1) and (x2, y2). The motion vector 130 is displayed from (a, b) to (a1, b1) in the motion vector 130 shown on the left side, and the connection line between the stored position values (a, b) and (a1, b1) (that is, Since the display line of the motion vector) passes through the reference line 140, the target light source 130 passing through (a, b) and (a1, b1) on the left side is recognized as the counter vehicle 11.

Expressed by the equation, when b1> y1> b and a1 <= x2, the target light source 110 passing through (a, b) and (a1, b1) becomes the opposite vehicle 11.

Similarly, the motion vector 130 displayed in the center passes through the reference line 140 indicated by the ruled line and thus is recognized as the counter vehicle 11. However, since the far right motion vector 130 does not cross the reference line 140, it is not recognized as the counter vehicle 11.

5 is a flowchart illustrating a vehicle recognition method according to another embodiment of the present invention.

The determining of whether the target light source 110 of the driving vehicle 10 is the counter vehicle 11 is as described above. That is, in a vehicle recognition method of recognizing an opposing vehicle by analyzing an image acquired from a camera, a region of interest is set at a position defined in the acquired image (S510), and a specific brightness is abnormal in the region of interest. Detecting the portion having the set to the target light source (S520), obtains a motion vector (Motion Vector) indicating the temporal movement of the target light source (S530), and determines whether the target light source is a counter vehicle based on the motion vector ( S540).

When it is determined that the target light source 110 is the counter vehicle 11, the light irradiation pattern of the headlamp of the driving vehicle 10 is controlled according to the position of the counter vehicle 11 (Yes), and the target light source 110 is controlled. If it is not determined that the vehicle 11 is opposed, the operation of controlling the headlamp does not occur (No) (S550).

Here, according to the position of the vehicle 11, the head lamp is rotated or a part of the light irradiated from the head lamp is blocked to form a dark zone, and the head lamp is rotated to control the light irradiation pattern of the head lamp. . Here, dark zone refers to the area | region where light is not irradiated.

The light irradiated from the headlamp of the driving vehicle 10 based on the counter vehicle 11, in particular, an upward light, causes glare to the driver of the counter vehicle 11, thereby obstructing the field of view and thereby causing a momentary vehicle accident. Since the possibility can be increased, it is necessary to adjust the light irradiation pattern of the head lamp of the driving vehicle 10.

Therefore, when the driving vehicle 10 turns on the driving light, the driving lamp 10 controls the driving lights of the headlamp without the driver's operation and controls the driving lights to prevent the driver's view of the opposing vehicle 11 from interfering with the driving vehicle 10 driver. The field of view is to ensure maximum safety to ensure safe driving.

In general, there are two headlamps on the front of the vehicle. The headlamp can be rotated to change the direction of the light source to which light is irradiated. In addition, it is possible to block a part of the light irradiated from the headlamps according to the position of the counter vehicle 11. Of course, it can also be configured to block not all of the light irradiated from the headlamps.

As a result, when the target light source 110 is determined as the counter vehicle 11, the counter vehicle 11 is affected by light emitted from the left head lamp of the driving vehicle 10, and thus, rotates the left head lamp. It is possible to prevent the glare, etc. of the driver of the opposing vehicle 11 by forming a dark zone. In addition, by blocking a part of the light irradiated from the headlamp to form a dark zone it is possible to prevent the glare of the driver of the opposing vehicle (11).

6 is a block diagram of a vehicle recognition apparatus according to an embodiment of the present invention.

The vehicle recognition apparatus 200 may include a storage unit 210 in which database movement patterns of a counter vehicle are stored in a database, an image acquisition unit 220 for acquiring an image in front of the vehicle, and a region of interest in the acquired image. ), A region of interest setting unit 230 for setting a region of light, a light source setting unit 240 for detecting light in the region of interest, and setting it as a target light source, and tracking the target light source using a motion vector. A light source tracking unit 250 and a vehicle determination unit 260 that determines whether the target light source is the counter vehicle by comparing the stored movement pattern of the opposing vehicle and the movement pattern of the target light source. In addition, a light source unit 270 to which light is irradiated and a controller 280 to control the light irradiation pattern of the light source unit 270 according to a light irradiation pattern formed by blocking a part of the light generated from the light source of the driving vehicle 10. And the like may be further included.

In the image acquired by the image acquisition unit 220, a region of interest is set by the region of interest setting unit 230, from which the light source setting unit 240 detects light in the region of interest, and thus the target light source. The target light source is tracked using a motion vector through the light source tracking unit 250. In addition, the vehicle determination unit 260 compares the movement pattern of the counter vehicle with the database stored in the storage unit 210 and determines the target light source as the counter vehicle. Then, when the vehicle determining unit determines the target light source as the counter vehicle, the vehicle determination unit guides the driver to the driver using sound or text.

In order to inform the driver of the driving vehicle of the target light source, the display module may be visually displayed to the driver or an alarm module may be provided.

Here, the display module allows the driver to see the captured image as it is. The display module displays a cathode ray tube (CRT), a liquid crystal display (LCD), a light-emitting diode (LED), and an organic light emitting diode (OLED) that can display an input image signal. A module having an image display means such as a light-emitting diode (PDP) or a plasma display panel (PDP) serves to display the received image information.

In addition, the alarm module generates a warning event to the driver, and the warning event may include a voice warning and automatic braking of different volumes depending on the degree of danger.

FIG. 7A is a diagram showing a movement pattern of a light source of a vehicle against a straight road, and FIG. 7B is a diagram showing a movement pattern of a light source of a vehicle against a curved road.

The movement patterns of the vehicle, which are stored in a database in the storage unit 210, refer to a predetermined pattern according to the driving of the vehicle.

In FIG. 7A, it can be seen that the movement pattern of the opposing vehicle follows the movement of the target light source 110, and moves straight downward with a predetermined slope in the X-Y plane in the case of a straight road. In addition, in FIG. 7B, the movement pattern of the opposing vehicle 11 depends on the movement of the target light source 110, and in the case of a curved road, the movement pattern of the opposing vehicle moves downward with a predetermined curvature in the XY plane. Able to know.

Therefore, by comparing the movement pattern of the counter vehicle 11 previously stored in the storage unit 210 with the actual movement pattern of the target light source 110, it may be determined whether the counter vehicle 11. It may be determined based on a predetermined criterion that the inclination of the motion vector is within a predetermined reference range and the inclination of the motion pattern of the opposing vehicle 11 stored in advance.

As described in the embodiment of the vehicle recognition method, the storage unit 210 stores the position value of the target light source 110 every frame, and the vehicle determination unit 260 connects the stored position value. The target light source 110 may be recognized as the counter vehicle 11 when passing the predetermined reference line.

Here, the position value of the target light source 110, the image acquisition unit 220 for acquiring an image of the front of the vehicle, and the region of interest setting unit 230 for setting a region of interest in the obtained image. And a light source setting unit 240 that detects light in the ROI and sets the target light source 110 as a target light source 110, and a light source tracking unit which tracks the target light source 110 using a motion vector. Obtained according to the inter-frame movement of the motion vector through 250.

When it is determined that the target light source 110 is the counter vehicle 11, the light irradiation pattern of the driving vehicle 10 is controlled according to the position of the counter vehicle 11, and the target light source 110 controls the counter vehicle 11. If not determined, the operation of controlling the light irradiation pattern of the driving vehicle 10 does not occur.

As described above, the vehicle recognition apparatus 200 includes a light source unit 270 to which light is irradiated according to a light irradiation pattern formed by blocking a part of light generated from the light source of the driving vehicle 10 and the light of the light source unit 270. The controller 280 further controls the irradiation pattern.

For example, when the target light source 110 is determined as the counter vehicle 11 by the vehicle determination unit 260, the control unit 280 rotates the light source unit 270 according to the position of the counter vehicle 11 to light the light source unit. The light irradiation pattern of the light source unit 270 is controlled by controlling the light irradiation pattern of 270 or blocking a part of the light irradiated from the light source unit 270.

Here, the controller 280 blocks the rotation of the light source unit 270 or a part of the light irradiated from the light source unit 270 to form a dark zone (not shown), and at least one of both ends of the dark zone is opposed to the vehicle ( The light source unit 270 is rotated to approach 11). That is, the control unit 280 rotates the light source unit 270 according to the position of the counter vehicle 11 to control the light irradiation pattern of the light source unit 270 or to block a part of the light irradiated from the light source unit 270 to light source unit 270. The light irradiation pattern of the control panel may be formed to form a dark zone, and the light source unit 270 is rotated to bring the vehicle 11 close to the dark zone to prevent glare of the driver of the opposite vehicle 11.

As another example, when the target light source 110 is determined as the counter vehicle 11 by the vehicle determination unit 260, the controller 280 may select at least one of the first and second light source units provided in the light source unit 270. Rotating to form a dark zone, or to block a portion of the light irradiated from at least one light source unit 270 of the first and second light source unit to form a dark zone. In addition, the controller 280 rotates at least one light source unit 270 of the first and second light source units such that at least one of both ends of the dark zone closes to the counter vehicle 11, such as glare of the driver of the counter vehicle 11. Will be prevented.

Here, since the vehicle is provided with a left head lamp and a right head lamp, there will be two light source units 270, which are defined as a first light source unit and a second light source unit.

That is, the controller 280 controls the light irradiation pattern of the light source unit 270 by applying at least one of rotation of the light source unit 270 and adjustment of light blocking of the light source unit 27 according to the position of the counter vehicle 11. A first light source unit which forms a dark zone according to the position of the vehicle 11 and rotates the light source unit 270 or configures the light source unit 270 so that one of both ends of the dark zone closes to the counter vehicle 11; At least one of the second light source units may be rotated to secure a view of the driver of the driving vehicle 10 and the opposing vehicle 11.

Therefore, the counter vehicle 11 is distinguished and determined from all the target light sources 110 appearing in front of the driving vehicle 10, and the dark zone is formed from the driving vehicle 10 in the determined counter vehicle 11. It is possible to secure the view of the driver of the driving vehicle 10 while not obstructing the view of the driver of the opposing vehicle 11. In addition, it is possible to provide the driver with more accurate forward recognition so that the driver can drive safely.

While the present invention has been described in connection with what is presently considered to be practical exemplary embodiments, it is to be understood that the invention is not limited to the disclosed embodiments, but, on the contrary, You will understand. It is therefore to be understood that the above-described embodiments are illustrative in all aspects and not restrictive.

100: image window 110: target light source
120: region of interest (ROI) 130: motion vector (MV)
140: reference line 210: storage unit
220: image acquisition unit 230: ROI setting unit
240: light source setting unit 250: light source tracking unit
260: vehicle determination unit 270: light source unit
280: control unit

Claims (22)

In a vehicle recognition method for recognizing an opposing vehicle by analyzing an image obtained from a camera,
(a) setting a region of interest at a location defined in the acquired image;
(b) detecting a portion having a specific brightness or more in the ROI and setting it as a target light source;
(c) obtaining a motion vector representing a temporal movement of the target light source and storing a position value of the target light source every frame; And
and (d) recognizing whether the target light source is a counter vehicle based on the motion vector. delete The method of claim 1,
The step (d)
And recognizing the target light source as a counter vehicle when the connection line connecting the stored position values passes a predetermined reference line. The method of claim 1,
The region of interest is a rectangular shape including a vanishing line. The method of claim 1,
The vehicle recognition method of setting the target light source when the light is a predetermined reference luminance range. The method of claim 1,
And detecting the target light source only when the magnitude of the motion vector is greater than or equal to a predetermined threshold. The method of claim 1,
And the position value is displayed as an XY coordinate value using the image as the XY plane. The method of claim 3,
And the reference line is set at a lower left half of the ROI. The method of claim 1,
And (e) controlling the light irradiation pattern of the headlamp of the driving vehicle according to the position of the counter vehicle when the target light source is determined as the counter vehicle. The method of claim 9,
In the step (e), the head lamp is rotated according to the position of the opposing vehicle or a part of the light radiated from the head lamp is blocked to form a dark zone, and the head lamp is rotated so that the head lamp is rotated. Vehicle recognition method for controlling the light irradiation pattern. A storage unit storing database movement patterns of the opposing vehicle;
An image acquisition unit for acquiring an image in front of the vehicle;
A region of interest setting unit configured to set a region of interest in the acquired image;
A light source setting unit configured to detect light in the ROI and set it as a target light source;
A light source tracking unit which tracks the target light source using a motion vector; And
And a vehicle determination unit comparing the movement pattern of the stored counter vehicle with the movement pattern of the target light source to determine whether the target light source is the counter vehicle. 12. The method of claim 11,
And the movement pattern of the opposing vehicle moves downward left and down with a predetermined slope in the XY plane in the case of a straight road. 12. The method of claim 11,
And the movement pattern of the opposing vehicle moves downward and leftward with a predetermined curvature in the XY plane in the case of a curved road. 12. The method of claim 11,
The storage unit stores the position value of the target light source every frame. The method of claim 14,
And the vehicle determination unit recognizes the target light source as a counter vehicle when a connection line connecting the stored position values passes a predetermined reference line. 12. The method of claim 11,
And a control unit for controlling the light irradiation pattern to which light is irradiated according to the light irradiation pattern formed by blocking a part of the light generated from the light source and the light irradiation pattern. 17. The method of claim 16,
When the target light source is determined to be a counter vehicle by the vehicle determination unit, the control unit rotates the light source unit or blocks a part of light emitted from the light source unit according to the position of the counter vehicle, thereby irradiating the light of the light source unit. Vehicle recognition device to control the pattern. 18. The method of claim 17,
The controller is configured to block the rotation of the light source or a portion of the light irradiated from the light source to form a dark zone, the vehicle recognition device for rotating the light source unit so that at least one of both ends of the dark zone close to the counter vehicle. 17. The method of claim 16,
The light source unit includes a first light source unit and a second light source unit. 20. The method of claim 19,
And the controller is configured to rotate at least one of the first and second light source units to form a dark zone when the target light source is determined to be the vehicle against the vehicle determination unit. 20. The method of claim 19,
And the controller is configured to block a portion of light emitted from at least one light source unit of the first and second light source units when the target light source is determined to be a vehicle against the vehicle, thereby forming a dark zone. 22. The method of claim 21,
And the control unit rotates at least one light source unit of the first and second light source units such that at least one of both ends of the dark zone is close to the counter vehicle.

Images