KR20200140769A - System and method for detecting vehicle by using sensor - Google Patents

Abstract

Provided are a system and a method for recognizing a vehicle by using a sensor. The system for recognizing a vehicle by using a sensor comprises: an image sensor for receiving light reflected from a rear reflector of a preceding vehicle in front of a vehicle and generating image information including received intensity of the received light; a pixel detection unit for detecting a pixel area having the received strength within a previously trained boundary range according to the distance from the preceding vehicle in order to detect, from the generated image information, a pixel area having the received strength of the light reflected from the rear reflector of the preceding vehicle; and an autonomous emergency braking (AEB) operation point control unit for calculating the number of detected pixel areas, classifying the vehicle model of the preceding vehicle by using the calculated number of pixel areas, and determining an operating point of the AEB system of the vehicle according to the classified vehicle model of the preceding vehicle. Therefore, classifying performance can be increased, and wrong detections can be minimized.

Description

Vehicle recognition system and method using sensors {SYSTEM AND METHOD FOR DETECTING VEHICLE BY USING SENSOR}

The present invention relates to a vehicle recognition system and method using a sensor, and more particularly, to a system and method for recognizing a preceding vehicle in front using a LIDAR sensor.

Currently, 3D image-based obstacle recognition systems for Autonomous Emergency Breaking (AEB) systems in the automotive field are generally RADAR (RAdio Detection And Ranging) sensors, LIDAR (LIght Detection And Ranging) sensors, stereo cameras, sensor fusion, etc. The same 3D image camera sensor is used.

1 is a diagram showing the types of 3D image camera sensors according to the prior art. As shown in FIG. 1, the 3D image camera sensor can be classified into a passive sensor and an active sensor. A typical passive sensor is a stereo camera 10 in a visible light region, and as an active sensor, a LIDAR sensor 20 in an ultraviolet region and a RADAR sensor 30 in a millimeter wave band are typical. Specifically, the stereo camera 10 acquires 3D information of an obstacle or a preceding vehicle 40 based on two camera images and trigonometry, and the LIDAR sensor 20 and the RADAR sensor 30 are based on TOF (Time Of Flight). Acquires 3D information of an obstacle or a preceding vehicle 40. Among them, the RADAR sensor 30 has a disadvantage of low horizontal resolution and the stereo camera 10 has a weak vulnerability in the night environment, so in the future, the LIDAR sensor 20 that is strong against the day/night environment of the vehicle is used alone. There is a need for a way to develop an obstacle recognition system.

2 is a view showing a mounting position in a vehicle of the LIDAR sensor system according to the prior art. As shown in Fig. 2, the LIDAR sensor system is mounted on the windshield 21 or bumper 22 of the vehicle, and although not shown, it can be mounted on the rear of the vehicle lamp, and a flash or scan type laser beam is used. It is an obstacle recognition system that recognizes the preceding vehicle by using.

3 is a block diagram showing a LIDAR sensor system according to the prior art. As shown in FIG. 3, the LIDAR sensor system 100 includes a transmitter, a receiver, and a processor. The LD (Laser Diode) 134 of the transmitter 130 generates a laser signal through the LD driver 132 and emits a laser signal through the optical system 136. The receiving unit 110 condenses the laser signal returned from the front obstacle or the preceding vehicle 40 to the focal plane array (FPA) 114 through the optical system 112. The condensed signal is transmitted to the processor 120 after improving the signal to noise ratio (SNR) in the preprocessor 116. Here, the FPA 114 is composed of PD (Photo Diode) 115 using light of a single wavelength band, and each PD 115 receives the light reflected from an obstacle or the preceding vehicle 40 to determine the distance and light reflection amount. Calculate The processing unit 120 detects and classifies the obstacle or the preceding vehicle 40 using distance information and light reflection amount information transmitted from the receiving unit 110. As a classification method, surrounding PDs 115 having similar distance values are grouped and detected as one obstacle, and the obstacle is classified using the amount of light reflection from the detected obstacle.

However, in the existing LIDAR sensor system 100 that classifies obstacles using the amount of light reflection from the detected obstacle, it is difficult to maintain the reliability of obstacle classification due to the color, geometric shape of the obstacle, reflected noise due to the surrounding environment, and bad weather. Very difficult. Therefore, in addition to the existing LIDAR sensor system 100, there is a need for a recognition system that improves the classification performance of obstacles and minimizes erroneous detection of obstacles.

Accordingly, it is an object of the present invention to detect and classify the preceding vehicle using the number and spacing characteristics of rear reflectors mounted at the rear of the vehicle in order to solve the above-described problem, thereby improving classification performance and detecting falsehood in addition to the existing sensor system. It is to provide a vehicle recognition system and device that minimizes.

A vehicle recognition system using a sensor according to an aspect of the present invention to achieve the above object receives the reception intensity of light reflected from the preceding vehicle in front of the vehicle, and generates image information including the reception intensity of the received light. In order to detect a pixel region having a reception intensity of light reflected from a rear reflector of the preceding vehicle from the generated image information, a reception intensity within a boundary range previously learned according to a distance from the preceding vehicle A pixel detection unit for detecting a pixel area having a and calculating the number of the detected pixel areas, classifying the vehicle type of the preceding vehicle using the calculated number of pixel areas, and classifying the vehicle type of the preceding vehicle. It includes an AEB operation point control unit that determines an operation point of an automatic emergency braking system (AEB) of the vehicle.

The configuration and operation of the present invention will become more apparent through specific embodiments described later together with the drawings.

According to the present invention, as a vehicle recognition system that detects a preceding vehicle with geometric characteristics of a rear reflector of a preceding vehicle using an image sensor, it is resistant to day/night environments and can obtain improved detection and classification performance than conventional sensor systems. It works.

In addition, there is an effect of reducing the damage of a vehicle accident by adjusting the operating point of an automatic emergency braking system (AEB) using the vehicle type information of the classified preceding vehicle and the current vehicle operation information.

1 is a diagram showing the types of 3D image camera sensors according to the prior art.
2 is a view showing a mounting position in a vehicle of the LIDAR sensor system according to the prior art.
3 is a block diagram showing a LIDAR sensor system according to the prior art.
4 is a block diagram showing a vehicle recognition system according to an embodiment of the present invention.
5 is a graph showing I ₀ (d), which is a boundary value of light reception intensity according to a distance between an own vehicle and a preceding vehicle according to an embodiment of the present invention.
6 is a flowchart illustrating a method of detecting a rear reflector of a preceding vehicle according to an embodiment of the present invention.
7 is a flowchart illustrating a method of calculating an operation point of an automatic emergency braking (AEB) system according to an embodiment of the present invention.
8 is a view showing a two-wheeled vehicle and a four-wheeled vehicle existing in the front according to an embodiment of the present invention.
9 is a diagram showing an operation point of an AEB system according to a vehicle type difference according to an embodiment of the present invention.
10 is a diagram illustrating an exemplary driving situation for explaining the operation of the 3D image-based vehicle recognition system and method according to the present invention.
11 is a diagram for explaining an offset difference, which is one of conditions for vehicle recognition in the situation of FIG. 10.
12 is a diagram illustrating a 3D image of 5*8 pixels as an example of acquired 3D image information.

The present invention provides a vehicle recognition system and method for detecting and classifying a preceding vehicle using the number and spacing characteristics of rear reflectors of the preceding vehicle. In addition, it provides a vehicle recognition system and method for determining an operation point of an automatic emergency braking system (AEB) using the classified information of the vehicle type of the preceding vehicle and the own vehicle.

Advantages and features of the present invention, and a method of achieving them will become apparent with reference to the embodiments described below in detail together with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in a variety of different forms, and only these embodiments make the disclosure of the present invention complete, and are common knowledge in the technical field to which the present invention pertains. The scope of the present invention is determined by the description of the claims, since it is provided only to completely inform the scope of the invention to the person having the present invention.

Meanwhile, terms used in the present specification are for explaining embodiments and are not intended to limit the present invention. In the present specification, the singular form also includes the plural form unless specifically stated in the phrase. As used herein, "comprise" and/or "comprising" refers to the presence of one or more other components, steps, actions and/or elements in the mentioned component, step, operation and/or element. Or does not preclude adding.

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

4 is a block diagram showing a vehicle recognition system according to an embodiment of the present invention, and FIG. 5 is a boundary value of light reception intensity according to a distance between an own vehicle and a preceding vehicle according to an embodiment of the present invention, I ₀ (d ) Is a graph.

A vehicle recognition system according to an exemplary embodiment of the present invention will be described in detail with reference to FIGS. 4 and 5 together.

As shown in FIG. 4, the vehicle recognition system according to an exemplary embodiment of the present invention provides light reflected from a preceding vehicle (specifically, from a rear reflector) among light reflected from an obstacle in front of the vehicle and returned. It detects and classifies the vehicle's automatic emergency braking system (AEB) to adjust the operating point. To this end, the vehicle recognition system includes an image sensor 400 and a processing unit 410.

The image sensor 400 calculates the position of the obstacle by emitting light such as a laser pulse and measuring the time it takes for the light to return after being reflected from the obstacle in front. In addition, the image sensor 400 generates image information representing the calculated position of the obstacle as the light reception intensity (I(x,y,d)) 50 according to the distance. Here, I is a function representing the intensity of light reception, x is the distance between obstacles, and y is the height of the obstacle measured from the ground. And, d means the distance between the obstacle and the own vehicle. An image sensor that generates image information expressed by the light reception intensity (I(x,y,d)) 50 may be a LIDAR sensor.

The processing unit 410 detects light reflected by the rear reflector of the preceding vehicle from the image information generated by the image sensor 400, classifies the vehicle type of the preceding vehicle from the detected light, and classifies the vehicle type of the preceding vehicle from the detected light, and an automatic emergency braking system (AEB) of the host vehicle Adjust the operating point of To this end, the processing unit 410 includes a pixel detection unit 412 and an AEB operation point control unit 414.

The pixel detection unit 412 receives image information expressed as the light reception intensity (I(x,y,d)) 50 from the image sensor 400, so that x, which is the distance between obstacles reflecting light, and Image information expressed as I(d), which is the intensity of light reception according to the distance d between the obstacle and the host vehicle, is generated, excluding the height y. By using the generated image information I(d), the pixel detector 412 identifies the light reflected from the rear reflector of the preceding vehicle among the light reflected from the obstacle.

In addition, the pixel detection unit 412 detects a pixel region having a reception intensity of light reflected from a rear reflector of a preceding vehicle among the received light reception intensity (I(x,y,d)) 50. Receiving intensity of road sign (I _cat (d)), reception intensity of rear reflector (I _ref (d)), and threshold of light reception intensity (I ₀ '(d )). Here, the boundary value of the light reception intensity according to the distance to the host vehicle (I ₀ '(d)) is a value that is set through an experiment or can be updated through continuous learning while driving the vehicle, as shown in FIG. 5, It has a value between the reception intensity of the road marker (I _cat (d)) and the reception intensity of the rear reflector (I _ref (d)). In addition, the reception intensity I(d) of the light reflected from the rear reflector of the preceding vehicle must have a value greater than the threshold value I ₀ '(d) of the light reception intensity according to the distance. Therefore, I ₀ '(d) has a value greater than I _cat (d) and less than I _ref (d), and the reception intensity of light reflected from the rear reflector of the preceding vehicle is greater than I ₀ '(d). Have.

In this way, the pixel detector 412 detects an I(d) value having a value greater than the threshold value I ₀ '(d) of the light reception intensity according to the distance, and transmits the light corresponding to this I(d) to the rear. It is generated as image information of the pixel area P _ref (x',y',d') of light reflected from the reflector.

The AEB operation point control unit 414 classifies the vehicle type of the preceding vehicle and compares it with the operation information 52 of the own vehicle to adjust the operation point of the automatic emergency braking system AEB of the own vehicle. To this end, the AEB operation point control unit 414 includes a classifier 414A and an operation point determination unit 414B.

The classifier 414A calculates the number of pixel areas Pref(x',y',d') of the rear reflector detected by the pixel detection unit 412, and groups adjacent pixel areas among the calculated pixel areas of the rear reflector. . Since the number of grouped pixel areas means the number of rear reflectors of the preceding vehicle, the vehicle type of the preceding vehicle is classified using the calculated number of rear reflectors.

The operation point determination unit 414B uses the vehicle type information of the preceding vehicle classified by the classifier 414A and the operation information 52 of the own vehicle received from the ECU (Electronic Control Unit) 420 of the own vehicle. The operation point of the automatic emergency braking (AEB) system is determined, and the AEB operation signal 54 is transmitted to the AEB system 430 of the host vehicle.

Here, the driving information 52 of the own vehicle received from the ECU 420 includes information such as vehicle type, braking force, speed, steering angle, wiper, and headlight status.

Hereinafter, a method of detecting a rear reflector of a preceding vehicle according to an embodiment of the present invention will be described in more detail with reference to FIG. 6.

6 is a flowchart illustrating a method of detecting a rear reflector of a preceding vehicle according to an embodiment of the present invention.

Unless otherwise noted, the execution subject of FIG. 6 refers to the pixel detection unit 412.

As shown in FIG. 6, the intensity of the light received from the image sensor 400 (I(x,y,d)) 50 is reflected between the obstacles that reflect the light, x, and the height of the obstacle. Except for y, image information expressed as I(d), which is the light reception intensity according to the distance d between the obstacle and the host, is generated (S600).

Then, of the reception intensity I (d) of the light reflected from the obstacle, an I (d) value greater than I ₀ '(d), which is a threshold value of the reception intensity of light according to the distance to the host vehicle, is detected (S610). Here, I ₀ '(d) is the learned boundary value, the learned reception intensity I _cat (d) of the road marker according to the distance to the host vehicle, the learned reception intensity I _ref (d) of the rear reflector, and the fine light This information is obtained through calculation using I _noise (d), the reception strength of the surrounding obstacles reflecting the I ₀ '(d) is greater than I _noise (d), and between I _cat (d) and I _ref (d). Has the value of

Next, the light reflected from the rear reflector of the preceding vehicle is detected using the I(d) value falling within the above-described range, and the pixel area of the light reflected from the rear reflector is P _ref (x', y', d') image information. It is generated as (S620).

Thereafter, the classifier 414A classifies the vehicle type of the preceding vehicle using the generated pixel area (Pref(x',y',d')) information of the rear reflector. The classified vehicle type information is used by the operation point determination unit 414B to calculate the operation point of the automatic emergency braking (AEB) system of the host vehicle.

A method of classifying a vehicle type of a preceding vehicle and a method of calculating an operation point of an automatic emergency braking (AEB) system of an own vehicle according to an embodiment of the present invention will be described in detail with reference to FIG. 7.

7 is a flowchart illustrating a method of calculating an operation point of an automatic emergency braking (AEB) system according to an embodiment of the present invention.

Unless otherwise noted, the performing subject of FIG. 7 refers to the classifier 414A.

As shown in FIG. 7, the pixel area (P _ref (x', y', d')) of the rear reflector detected by the pixel detection unit 412 is a pixel area using only d', which is the distance information between the host vehicle and the preceding vehicle. Image information represented by P _ref (d') is generated (S700).

Hereinafter, the pixel area P _ref (d') is grouped with adjacent pixel areas, and the pixel area P _{ref_GP} (d') is generated as image information represented by the grouped pixel area P _{ref_GP} (d') (S710). Here, one grouped pixel area P _{ref_GP} (d') refers to light reflected from one rear reflector of the preceding vehicle.

The number of grouped pixel areas P _{ref_GP} (d') is calculated, and the vehicle type of the preceding vehicle is first classified using the number of P _{ref_GP} (d'), that is, the number of rear reflectors of the preceding vehicle. In general, a two-wheel vehicle has one rear reflector, and four or more vehicles have two rear reflectors, so the vehicle type of the preceding vehicle is classified as a two-wheel vehicle or a four-wheel vehicle using the calculated number of rear reflectors. .

Here, when the number of P _{ref_GP} (d') is 2 or more (S720), that is, when the vehicle type of the preceding vehicle is classified as a 4-wheel vehicle, the distance between the detected rear reflectors is calculated (S730), and the calculated rear reflection By comparing the period interval with the previously stored reference rear reflector interval, the vehicle type of the preceding vehicle is classified into a small car, a midsize car, and a large car.

Table 1 is an interval reference table showing the interval between the rear reflection periods according to vehicle types.

division
Two wheeled vehicle
(2W) 4-wheel vehicle (4W) Small car Midsize car Large car Rear reflector spacing (m) S ₂ Small(S _4S ) Middle(S _4M ) Large(S _4L ) 0 1.4~1.5 1.5~1.6 1.6~ Braking distance short short middle long Degree of blinding little little middle many

Referring to Table 1, when the number of detected rear reflectors is two and the interval is between 1.5 and 1.6 m, the vehicle type of the preceding vehicle is classified as a mid-size vehicle of a four-wheel vehicle. In addition, when the number of P _{ref_GP} (d') is 3, that is, when the front four-wheel vehicle 58 and the two-wheel vehicle exist at the same time, the detected rear reflection period distances are each calculated, and three rear reflections Among the period intervals, the second largest interval is classified as that of a four-wheel vehicle.

In summary, when the calculated number of pixel areas is one, the vehicle type of the preceding vehicle is classified as a two-wheeled vehicle. In addition, when the calculated number of pixel regions is two and the interval between the pixel regions falls within the reference interval range of the rear reflector of the four wheel vehicle, the vehicle type of the preceding vehicle is classified as a four wheel vehicle. In addition, when the calculated number of pixel regions is three, the pixel region having the second largest interval among the intervals of the pixel regions is classified as a four-wheel vehicle.

The operation point determination unit 414B receives the vehicle type of the preceding vehicle classified by the classifier 414A and the vehicle type information of the host vehicle received from the ECU of the host vehicle, calculates the difference between the vehicle type of the preceding vehicle and the host vehicle (S740), and calculates An operation point of an automatic emergency braking (AEB) system of the host vehicle is determined based on the difference value of the vehicle type (S750).

A method of distinguishing a front two-wheeled vehicle and a four-wheeled vehicle according to an embodiment of the present invention will be described in detail with reference to FIG. 8.

8 is a view showing a two-wheeled vehicle and a four-wheeled vehicle existing in the front according to an embodiment of the present invention. As shown in Fig. 8, when the four-wheel vehicle 58 and the two-wheel vehicle 60 exist at the same time in front, that is, when the number of P _{ref_GP} (d') calculated by the classifier 414A is three, the classifier ( 414A) calculates the detected rear reflection period intervals S1, S2, and S3, respectively, and classifies the second largest interval S _{2nd max} (S2) among the three rear reflection period intervals as those of the four-wheel vehicle 58.

A method of determining an AEB operation point according to a vehicle type difference between an own vehicle and a preceding vehicle according to an embodiment of the present invention will be described in detail with reference to FIG. 9.

9 is a diagram showing an operation point of an AEB system according to a vehicle type difference according to an embodiment of the present invention. As shown in FIG. 9, when the vehicle type of the host vehicle is a small vehicle 62 and the vehicle type of the preceding vehicle is classified as a large vehicle 64, the classifier 414A determines the operation point of the AEB system as 2.

Table 2 is a table showing numerically the operating points of the AEB system according to the vehicle model difference.

division Two wheeled vehicle Small car Midsize car Large car Two wheeled vehicle 0 One 2 3 Small car One 0 One 2 Midsize car 2 One 0 One Large car 3 2 One 0

Here, the operating point of the AEB is a relative value obtained by comparing the size or weight between the host vehicle and the preceding vehicle. As the size or weight of the vehicle increases, the braking force decreases and the degree of obstruction of the driving vision tends to increase.Therefore, the effect of reducing the degree of impact and damage during an accident can be obtained by fluidly adjusting the operating point of the AEB.

The operation of the 3D image-based vehicle recognition system and method according to the present invention will be supplemented with reference to FIGS. 10, 11 and 12.

10 is a view showing an exemplary driving situation for explaining the operation of the 3D image-based vehicle recognition system and method according to the present invention, and FIG. 11 is an offset difference, which is one of conditions for vehicle recognition in the situation of FIG. 10 FIG. 12 is a diagram illustrating a 3D image of 5*8 pixels as an example of acquired 3D image information.

In order to recognize a vehicle according to the 3D image-based vehicle recognition system and method according to the present invention, the following five conditions are assumed.

-Own vehicle and preceding vehicle drive in the same lane

-2 rear reflectors are represented by at least 1 pixel each

-The spacing between the rear reflectors (S ₂ or S _{3 in} Fig. 8) is expressed as 2 pixels or more

-Vehicle classification is possible only when both rear reflectors are detected

-Offset between the host and the preceding vehicle is less than 50%

For example, as shown in FIG. 10, when the 3D image sensor 66 mounted on the own vehicle recognizes light reflected from an obstacle within the vehicle recognition area R2 and within the same driving lane as the own vehicle, the 3D image sensor Reference numeral 66 may obtain the two rear reflectors of the preceding vehicle and the road marker 72 as 3D image information. In addition, as shown in FIG. 2, the rear reflector of the preceding vehicle in front is expressed by the light reception intensity 70 of 1 pixel or more, and the interval between the rear reflection period is expressed by 2 pixels or more, as shown in FIG. Likewise, if the offset 78 between the host vehicle and the preceding vehicle is 50% or less, a condition for vehicle recognition is established. Here, the offset 78 is a condition for vehicle recognition, and is a reference value indicating the degree of the difference between the preceding vehicle and the host vehicle traveling in the same driving lane, and this offset 78 is the viewing angle, resolution, and recognition of the LIDAR sensor. It can be determined according to the distance.

An area in which vehicle recognition is possible based on the rear reflector may be represented by R2 as illustrated, and this R2 may be defined as an area between the minimum distance d _min and the maximum distance d _max . For example, when the obtained 3D image information can be represented as shown in FIG. 12, when the distance d1 between the preceding vehicle 80 and the host vehicle 76 is 20 m, the interval between the rear reflection period separated from the road marker 72 is When it is calculated as 1.56m, the 3D image sensor 66 classifies the preceding vehicle as a medium-sized vehicle according to the reference table for the distance between the rear reflectors according to the vehicle types in Table 1. In addition, the 3D image sensor calculates the vehicle type difference between the host vehicle and the preceding vehicle based on vehicle information such as braking force and speed of the host vehicle, and transmits an AEB operation signal set according to the calculated vehicle type difference to the ECU. The ECU operates the AEB system by receiving the AEB operation signal.

The above description is merely illustrative of the technical idea of the present invention, and those of ordinary skill in the art to which the present invention pertains, various modifications and variations can be made without departing from the essential characteristics of the present invention. Accordingly, the embodiments expressed in the present invention are not intended to limit the technical idea of the present invention, but are intended to be described, and the scope of the present invention is not limited by these embodiments. The scope of protection of the present invention should be construed by the following claims, and all technical ideas within the scope equivalent or equivalent thereto should be construed as being included in the scope of the present invention.

400: image sensor, 410: processing unit, 412: pixel detection unit, 414: AEB operation point control unit, 414A: classifier, 414B: operation point determination unit, 420: ECU, 430: AEB system

Claims (11)

The position of the obstacle is calculated from the light reflected from the obstacle, and the calculated position of the obstacle is expressed as the light reception intensity I(x,y,d) according to the distance (where x is between obstacles). A distance, y is the height of the obstacle from the ground, d is the distance between the obstacle and the host vehicle); And
Receives the image information expressed as I(x,y,d) from the image sensor, and generates image information expressed as the reception intensity I(d) of light reflected from the obstacle by excluding x and y, and the Including a pixel detector for distinguishing the light reflected from the rear reflector of the preceding vehicle among the light reflected from the obstacle,
The pixel detector detects an I(d) value greater than I ₀ '(d), which is a boundary value of the light reception intensity according to a distance from the host vehicle, among the generated image information I(d) (here, I ₀ '( d) is the learned threshold, the learned reception strength I _cat (d) of the road marker according to the distance from the host vehicle, the learned reception strength I _ref (d) of the rear reflector, and the surrounding obstacles reflecting fine light. Receiving intensity I _noise (d) is the light intensity information obtained through calculation using _noise (d)), the light corresponding to the value of I (d) is detected as light reflected from the rear reflector of the preceding vehicle, and reflected from the rear reflector. A vehicle recognition system, characterized in that configured to generate a pixel area of light as image information of P _ref (x',y',d'). The method of claim 1,
A classifier that calculates the number of pixel areas detected by the pixel detection unit and classifies a vehicle type of the preceding vehicle using the calculated number of pixel areas; And
Vehicle recognition system further comprising an AEB operation point determination unit for determining an operation point of the automatic emergency braking (AEB) system of the vehicle according to the classified vehicle type of the preceding vehicle. The method of claim 2, wherein the classifier
The number of pixel areas Pref(x',y',d') of the rear reflector detected by the pixel detection unit is calculated, and adjacent pixel areas among the calculated pixel areas of the rear reflector are grouped to determine the number of grouped pixel areas. A vehicle recognition system configured to classify a vehicle type of a preceding vehicle using information on the number of rear reflectors of the preceding vehicle. The method of claim 2, wherein the AEB operation point determination unit
Using the vehicle type information of the preceding vehicle classified by the classifier and the own vehicle operation information including at least one of vehicle type, braking force, speed, steering angle, wiper, and headlight status received from the ECU of the host vehicle, the automatic emergency braking (AEB) system of the vehicle A vehicle recognition system configured to determine an operation point and transmit an AEB operation signal to the AEB system of the own vehicle. 1) Calculate the position of the obstacle from the light reflected from the obstacle, and generate image information expressing the calculated position of the obstacle as the light reception intensity I(x,y,d) according to the distance (where x is the obstacles A distance between the obstacles, y is the height of the obstacle from the ground, and d is the distance between the obstacle and the host vehicle); And
2) An image that receives the image information expressed as I(x,y,d) from the image sensor, and is expressed as the reception intensity I(d) of the light reflected from the obstacle including d, excluding x and y Generating information, comprising the step of distinguishing the light reflected from the rear reflector of the preceding vehicle among the light reflected from the obstacle,
Step 2),
Among the generated image information I(d), an I(d) value greater than I ₀ '(d), which is the boundary value of the light reception intensity according to the distance to the own vehicle, is detected (here, I ₀ '(d) is a learning As the threshold value, the learned reception intensity I _cat (d) of the road marker according to the distance from the host vehicle, the learned reception intensity I _ref (d) of the rear reflector, and the reception intensity I _noise of the surrounding obstacle reflecting fine light It is the light intensity information obtained through calculation using (d)),
The light corresponding to the I(d) value is detected as light reflected from the rear reflector of the preceding vehicle,
And generating a pixel region of light reflected from the detected rear reflector as image information of P _ref (x',y',d'). The method of claim 5,
3) calculating the number of the pixel areas detected in step 2), and classifying the vehicle type of the preceding vehicle using the calculated number of pixel areas; And
4) The vehicle recognition method further comprising determining an operation point of the automatic emergency braking (AEB) system of the vehicle according to the classified vehicle type of the preceding vehicle. The method of claim 6, wherein step 3)
Calculate the number of pixel areas Pref(x',y',d') of the rear reflector, group adjacent pixel areas among the calculated pixel areas of the rear reflector, and the rear of the preceding vehicle represented by the number of grouped pixel areas A vehicle recognition method comprising the step of classifying a vehicle type of a preceding vehicle using information on the number of reflectors. The method of claim 6, wherein step 4)
The operation point of the automatic emergency braking (AEB) system of the vehicle by using the vehicle type information of the classified preceding vehicle and the own vehicle operation information including at least one of vehicle type, braking force, speed, steering angle, wiper, and headlight status received from the ECU of the host vehicle. And transmitting the AEB operation signal to the AEB system of the host vehicle. A means for emitting light to an obstacle in front of the own vehicle and receiving light from an image sensor that detects light reflected from the obstacle,
Calculate the position of the obstacle, and generate image information expressing the calculated position of the obstacle as the light reception intensity I(x,y,d) according to the distance (where x is the distance between obstacles, y is the ground The height of the obstacle from, d is the distance between the obstacle and the vehicle),
A means for generating image information expressed by the reception intensity I(d) of light reflected from an obstacle by excluding x and y from the I(x,y,d) image information,
In order to distinguish the light reflected from the rear reflector of the preceding vehicle among the light reflected from the obstacle, I ₀ '(d), which is the boundary value of the light reception intensity according to the distance to the host vehicle among the generated image information I(d). Means for detecting a larger I(d) value (where I ₀ '(d) is the learned boundary value, the learned reception strength I _cat (d) of the road marker according to the distance from the host vehicle, and the learning of the rear reflector This is the light intensity information obtained through calculation using the received intensity I _ref (d) and the intensity I _noise (d) of the surrounding obstacles reflecting fine light),
Means for detecting the light corresponding to the I(d) value as light reflected from the rear reflector of the preceding vehicle,
Means for generating a pixel region of light reflected from the detected rear reflector as image information of P _ref (x',y',d'),
Means for calculating the number of the detected pixel areas and classifying the vehicle type of the preceding vehicle using the calculated number of pixel areas, and
And means for determining an automatic emergency braking (AEB) operation point of the vehicle according to the classified vehicle type of the preceding vehicle. The method of claim 9, wherein the means for classifying the vehicle type of the preceding vehicle
A preceding vehicle represented by the number of grouped pixel areas by calculating the number of pixel areas Pref(x',y',d') of the generated rear reflector and grouping adjacent pixel areas among the calculated pixel areas of the rear reflector A vehicle recognition system, configured to classify a vehicle type of a preceding vehicle using information on the number of rear reflectors. The method of claim 9, wherein the means for determining the AEB operation point
Automated vehicle by using the vehicle type information of the preceding vehicle classified by the vehicle type classifying means of the preceding vehicle and the own vehicle operation information including at least one of vehicle type, braking force, speed, steering angle, wiper, and headlamp status received from the ECU of the host vehicle. A vehicle recognition system configured to determine an operation point of an emergency braking (AEB) system and to transmit an AEB operation signal to the AEB system of the host vehicle.

Images