JP6593263B2 - Lane marking recognition system - Google Patents

Description

  The present invention relates to a lane marking recognition device for recognizing a lane marking from a captured image of a camera.

  In an apparatus that recognizes a lane marking that divides a road lane from a captured image of a camera mounted on a vehicle, there is a problem in that noise included in the captured image is erroneously recognized as a lane marking. For this reason, various devices have been proposed for suppressing erroneous recognition of noise as a lane marking. For example, the lane mark recognition apparatus described in Patent Document 1 uses the fact that a repair mark of a lane mark, which is one of noises, is close to the lane mark and has a lower brightness than the lane mark. Recognize the repair traces of and prevent misrecognition.

Japanese Patent Laying-Open No. 2015-197829

  As noise included in a photographed image, there is road surface reflection in which light hitting a road surface is reflected and enters a camera lens. Like the lane marking, this road surface reflection has a higher brightness than the road surface on the image, and may be erroneously recognized as a lane marking. However, the road surface reflection does not always appear close to the lane marking, and does not necessarily have lower luminance than the lane marking. Therefore, the apparatus described in Patent Document 1 cannot distinguish between road surface reflections and lane markings.

  This invention is made | formed in view of the said situation, and makes it a main objective to provide the lane marking recognition apparatus which can discriminate | determine road surface reflection.

  The present invention relates to a lane marking recognition device (30) for recognizing a lane marking drawn on a road surface so as to divide a lane from a captured image obtained by capturing the surroundings of the vehicle with a camera (10) mounted on the vehicle (70). And a feature point detection unit (S100), a behavior acquisition unit (S110), a correction unit (S140 to S160), and a reflection determination unit (S140 to S160). The feature point detection unit detects a feature point representing a brighter part than the road surface from the captured image. The behavior acquisition unit acquires the behavior of the vehicle. The correction unit sets the frames of the captured image at a time point away from a preset period as the first frame and the second frame, and acquires the position of the first feature point that is the feature point detected in the first frame. Is corrected to the position in the second frame using the above behavior, and the corrected position is calculated. The reflection determination unit corresponds to a correction position where the second feature point does not exist in the first frame when the second feature point that is the feature point detected in the second frame does not exist in the calculated correction position. The portion represented by the first feature point is determined as the road surface reflection portion.

  According to the present invention, the position of the first feature point detected in the first frame is corrected to the position in the second frame using the behavior of the vehicle, and the corrected position is calculated. When the first feature point is a feature point representing a lane marking, the second feature point is detected at a correction position in which the position of the first feature point is moved by the amount corresponding to the behavior of the vehicle in the second frame. On the other hand, since the positional relationship between the camera and the light source can be considered to be constant in a short period, the road surface reflection portion is detected at the same position between frames. Therefore, when the first feature point is a feature point representing a road surface reflection portion, the second feature point is not detected at the correction position in the second frame. That is, in the first frame, the first feature point corresponding to the correction position where the second feature point is not detected is a feature point representing a road surface reflection portion. Therefore, it is possible to accurately determine the road surface reflection portion.

  In addition, the code | symbol in the parenthesis described in this column and a claim shows the correspondence with the specific means as described in embodiment mentioned later as one aspect, Comprising: The technical scope of this invention is shown. It is not limited.

It is a block diagram which shows schematic structure of a lane marking recognition apparatus. It is explanatory drawing which shows the arrangement position of a camera. It is a figure explaining a white line, repair trace, and road surface reflection. It is a figure explaining the characteristic of the white line which is a broken line. It is a figure explaining the characteristic of the white line which is a solid line. It is a figure explaining the feature of road surface reflection. It is a flowchart which shows the process sequence which outputs the recognition result of a white line. It is a flowchart which shows the process sequence which recognizes a white line. It is a figure explaining the determination of the road surface reflection part in the vehicle rear. It is a figure explaining the determination of the road surface reflective part in front of a vehicle.

Hereinafter, embodiments for carrying out the invention will be described with reference to the drawings.
<Configuration>
The lane marking recognition device according to the present embodiment is a device that is mounted on a vehicle 70 and recognizes a lane marking on a road. As shown in FIG. 1, the lane marking recognition apparatus according to the present embodiment includes an ECU 20, and a camera 10, sensors 17, and a vehicle control device 50 are connected to the ECU 20. The lane marking is a white line or a yellow line drawn on the road surface to divide the road lane. Hereinafter, the lane markings including the lane markings other than white are referred to as white lines.

  The camera 10 includes a front camera 11, a left side camera 12, a right side camera 13, and a rear camera 14. Each of the cameras 11 to 14 is a well-known camera configured using a CCD image sensor or a CMOS image sensor. As shown in FIG. 2, the front camera 11 is installed, for example, on a bumper at the front end of the vehicle so that the road surface in front of the vehicle is within the shooting range. The left side camera 12 is installed, for example, on the left side mirror so that the road surface on the left side of the vehicle falls within the shooting range. The right side camera 13 is installed, for example, on the right side mirror so that the road surface on the right side of the vehicle is in the shooting range. For example, the rear camera 14 is installed in a bumper at the rear end of the vehicle such that the road surface behind the vehicle is within the shooting range. Each of the cameras 11 to 14 repeatedly shoots at a preset time interval, for example, 1/15 (second) interval, and outputs a camera image to the ECU 20.

  The sensors 17 are sensors that measure the behavior of the vehicle 70. Specifically, the sensors 17 are a plurality of sensors including a vehicle speed sensor that measures the vehicle speed of the vehicle 70 and a yaw rate sensor that measures the yaw rate of the vehicle 70.

  The vehicle control device 50 is configured around a known microcomputer including a semiconductor memory such as a CPU, a ROM, a RAM, and a flash memory. The vehicle control device 50 is a control device that controls the steering of the vehicle 70, the brake, the engine, and the like so that the vehicle 70 travels in the lane based on the recognition result of the white line output from the ECU 20.

  The ECU 20 is configured around a known microcomputer including a semiconductor memory such as a CPU, a ROM, a RAM, and a flash memory. Each function realized by the ECU 20 is realized by the CPU executing a program stored in a non-transitional physical recording medium. In this example, the semiconductor memory corresponds to a non-transitional tangible recording medium storing a program. Further, by executing this program, a method corresponding to the program is executed. The number of microcomputers constituting the ECU 20 may be one or plural.

  The ECU 20 includes functions of an input processing unit 21, a synthesis processing unit 22, a recognition processing unit 23, and an output processing unit 24. The recognition processing unit 23 includes functions of a feature point detection unit, a behavior acquisition unit, a region setting unit, a correction unit, a reflection determination unit, and a lane marking recognition unit. The method of realizing these elements is not limited to software, and some or all of the elements may be realized using hardware that combines a logic circuit, an analog circuit, and the like.

<Characteristics of road surface reflection part>
In FIG. 3, the schematic diagram of the bird's-eye view image produced | generated by the synthetic | combination process part 22 is shown. A region indicated by a broken line in the center of the figure is a vehicle region where the vehicle 70 exists. The composition processing unit 22 synthesizes the camera images taken by the four cameras 11 to 14 and converts them into a bird's eye view, and generates a bird's eye view image surrounding the vehicle area. In the bird's-eye view image shown in FIG. 3, there are a broken white line, a solid white line, and a repair trace as characteristic portions of the road surface. In the present embodiment, a feature that is brighter than a road surface such as paint drawn on the road surface is used as a characteristic portion. The repair trace is a trace of repairing a crack on the road surface, for example, a trace of repairing a crack on the road surface of asphalt with tar, or a trace of repairing a crack on the road surface of concrete with asphalt. Cracks on the road surface often occur along the white line under tire pressure, particularly in snowy areas such as North America, and the repaired trace often becomes a straight line along the white line. In the present embodiment, the bird's eye view image corresponds to the captured image.

  Usually, the repaired track is darker than the road surface on the bird's eye view image. However, when light such as sunlight hits the repaired track and is reflected, and the reflected light is incident on the lens of the camera 10, the reflected portion that reflects the light becomes brighter than the road surface on the bird's eye view image. As described above, the feature of being brighter than the road surface on the image is not limited to the reflection portion of the repaired trace, but the entire road surface reflection portion. Therefore, in general, the road surface reflection portion may be erroneously recognized as a white line. Therefore, in the present embodiment, the difference in characteristics between the road surface reflection portion and the white line portion is clarified, and the road surface reflection portion and the white line portion are determined on the bird's eye view image. In addition, as a road surface reflection part other than the reflection part of a repair trace, there exists a reflection part of the wet road surface etc. when light hits and reflects on a wet road surface, for example.

  Next, with reference to FIG. 4 to FIG. 6, characteristics of the broken white line, the solid white line, and the road surface reflection portion will be described. 4 to 6 are schematic diagrams of bird's-eye view images at time points t0, t1, and t2. The time point t1 is a time point when the period ΔT has elapsed from the time point t0, and the time point t2 is a time point when the period ΔT has elapsed from the time point t1. The three frames flow from the front to the rear in the traveling direction of the vehicle 70 with the passage of time. Hereinafter, the frames at time t0 to time t2 are referred to as frame 0 to frame 2, respectively. The period ΔT is a preset period, and is a sufficiently short period in which the positional relationship between the sun, which is the light source, and the vehicle 70, that is, the positional relationship between the sun and the cameras 11 to 14 can be regarded as constant. It may be longer than the shooting time interval of -14. That is, the frame 0 and the frame 1 may be bird's-eye view images respectively generated from continuously captured camera images, or may be bird's-eye view images generated from non-continuous camera images. In the present embodiment, the period ΔT is set to a longer period as the vehicle speed of the vehicle 70 is lower.

  When the feature point representing a brighter part than the road surface represents a broken white line, as shown in FIG. 4, in the frame 1, the feature point is located between the frames in the direction opposite to the traveling direction from the position detected in the frame 0. Is detected at a position where the movement amount of the vehicle 70 is shifted. In the frame 2, the feature point is detected at a position shifted from the position detected in the frame 1 by the movement amount of the vehicle 70. In other words, in the current frame, the feature point is detected at a correction position obtained by correcting ego motion from the position detected in the previous frame.

  Next, when the feature point represents a solid white line, as shown in FIG. 5, the feature point in the current frame is subjected to ego-motion correction at the position detected in the previous frame as in the case of the dashed white line. It is detected at the correction position. However, in the case of the solid white line, the feature points are continuous in the traveling direction, and therefore the feature points are detected at the same position as the position detected in the previous frame in the current frame.

  On the other hand, when the feature point represents a road surface reflection portion, as shown in FIG. 6, the feature point is detected only at the same position between frames in the traveling direction and the width direction of the vehicle 70. That is, in the frame 1, the feature point is not detected at the correction position obtained by correcting the position detected in the frame 0 by ego motion correction. Further, in the frame 1, the feature point is not detected at a correction position obtained by correcting the position detected in the frame 2 by ego motion correction.

  If the positional relationship between the sun and the cameras 11 to 14 is constant, road surface reflection occurs at the same position even if the frames are different. Therefore, feature points representing road surface reflection portions are detected only at the same position between frames. Such a characteristic of the road surface reflection portion is not limited to the portion reflected by the light hitting the repair trace and is a feature of the road surface reflection portion in general.

<Processing>
Next, a processing procedure for recognizing and outputting a white line using the difference in characteristics between the white line and the road surface reflection portion described above will be described with reference to the flowchart of FIG. This processing procedure is repeatedly executed by the ECU 20 at the photographing time intervals of the cameras 11 to 14.

First, in step S10, camera images taken by the cameras 11 to 14 are acquired, and the acquired camera images are sampled and converted into digital signals.
Subsequently, in step S20, the four camera images converted into digital signals are converted into a bird's-eye view viewed from a preset virtual viewpoint, and a bird's-eye view image showing the surroundings of the vehicle 70 is generated.

Subsequently, in step S30, a white line is recognized from the bird's eye view image generated in step S20. Details of the white line recognition process will be described later.
Subsequently, in step S40, the detected white line recognition result is output to the vehicle control device 50 via the in-vehicle network. This process is complete | finished above.

  In the present embodiment, the process of step S10 corresponds to the process executed by the function of the input processing unit 21, and the process of step S20 corresponds to the process executed by the function of the synthesis processing unit 22. Further, the process of step S30 corresponds to the process executed by the function of the recognition processing unit 23, and the process of step S40 corresponds to the process executed by the function of the output processing unit 24.

Next, the white line recognition processing procedure will be described with reference to the flowchart of FIG.
First, in step S100, a feature point representing a brighter part than the road surface is detected from the generated bird's eye view image. In this embodiment, an edge point having a luminance value higher than the surrounding by a threshold value or more is used as a feature point, and a feature point is detected by applying a Sobel filter or the like to the bird's eye view image.

Subsequently, in step S110, the vehicle speed and yaw rate of the vehicle 70 measured by the sensors 17 are acquired.
Subsequently, in step S120, in the bird's-eye view image, three regions are set in front, center, and rear with respect to the traveling direction of the vehicle 70, and the feature points detected in step S100 are classified into the regions. The size of each area is set according to the vehicle speed of the vehicle 70. Specifically, as the vehicle speed is higher, the correction amount in the ego motion correction is larger, so the size of the front region and the rear region is increased and the size of the central region is decreased.

  Subsequently, in step S130, one area is selected from the three areas, and it is determined whether the area is the rear, front, or center. If the selected region is the rear region, in step S140, the road surface reflection portion is determined using the condition A, and the feature point representing the road surface reflection portion is removed as noise from the feature point detected from the bird's eye view image. If the selected region is the front region, in step S150, the road surface reflection portion is determined using Condition B, and the feature point representing the road surface reflection portion is removed as noise from the feature points detected from the bird's eye view image. . Details of the determination of the road surface reflection portion using the condition A and the determination of the road surface reflection portion using the condition B will be described later.

  Furthermore, when the selected area is the central area, the road surface reflection portion is determined using the condition A and the condition B in step S160. That is, the road surface reflection portion is determined using the condition A with the central region regarded as the rear region, and the road surface reflection portion is determined using the condition B with the central region regarded as the front region. Then, from the feature points detected from the bird's eye view image, the feature points representing the road surface reflection portion determined using the condition A and the feature points representing the road surface reflection portion determined using the condition B are removed as noise. To do. The feature point determined using the condition A and the feature point determined using the condition B may be the same.

  Subsequently, in step S170, it is determined whether or not the noise removal processing has been completed for all regions. If the noise removal process has not been completed for all regions, the process returns to step S130. If the noise removal process has been completed for all areas, the process proceeds to step S180.

  Subsequently, in step S180, a straight line is detected by applying Hough transform or the like to the remaining feature points from which noise has been removed among the feature points detected in step S100. The detected straight line is set as a white line candidate.

  Subsequently, in step S190, considering the acquired behavior of the vehicle 70, white line candidates that are likely to be white lines are selected from the white line candidates detected in step S170 on the left and right sides of the vehicle, respectively.

  Subsequently, in step S200, white line parameters are estimated from the white line candidates selected in step S180, and the white line is recognized. The white line parameter is, for example, the white line curvature, the lane width, and the angle formed by the traveling direction of the vehicle 70 and the tangential direction of the white line. This process is complete | finished above.

  In the present embodiment, the process of step S100 corresponds to the process executed by the function of the feature point detection unit, and the process of step S110 corresponds to the process executed by the function of the behavior acquisition unit. Further, the process of step S120 corresponds to the process executed by the function of the area setting unit, and the processes of steps S140 to S160 correspond to the process executed by the functions of the correction unit and the reflection determination unit. And the process of step S180-S200 is corresponded to the process which the function of a lane marking recognition part performs.

<Determination using Condition A or Condition B>
The determination of the road surface reflection portion using the condition A will be described with reference to FIG. In FIG. 9, the frame at the time point t110 is a bird's eye view image frame 110 generated by the processing of the current cycle, and the frame at the time point t100 is a frame 100 of the bird's eye view image generated by the processing before the N cycles. Further, the feature point detected in the frame 110 is defined as a feature point Xa, and the feature point detected in the frame 100 is defined as a feature point Ya. N is an integer equal to or greater than 1, and is a preset value. In the determination using the condition A, the frame 110 corresponds to the first frame, the frame 100 corresponds to the second frame, the feature point Xa corresponds to the first feature point, and the feature point Ya corresponds to the second feature point.

  As described above, unlike the feature points that represent the broken line and the solid white line, the feature points that represent the road surface reflection portion are not detected in the corrected position in the frame 100 where the position detected in the frame 110 is corrected by egomotion. The condition A is a condition of this feature. In the determination using the condition A, the position of the feature point Xa detected in the current frame 110 is corrected by ego motion to calculate a corrected position. In the determination using the condition A, when the feature point Ya does not exist at the correction position in the frame 100, the portion represented by the feature point Xa corresponding to the correction position where the feature point Ya does not exist in the frame 110 is determined as the road surface. It is determined as a reflection part. The feature point Xa corresponding to the correction position is the feature Xa detected at the detection position of the frame 100 before the egomotion correction. Further, ego motion correction is performed using the vehicle speed and yaw rate of the vehicle 70 acquired in step S110.

  In addition, if it is determined that the condition A is to be used and the portion represented by the feature point Xa detected at the same position between frames is a road surface reflection portion, the solid white line and the road surface reflection portion are not distinguished, and the solid white line is the road surface. Since it may be erroneously determined as a reflection part, it is not desirable. However, in addition to the determination using the above condition A, in the frame 110, a portion corresponding to the correction position where the feature point Ya does not exist and represented by the feature point Xa detected at the same position as the feature point Ya is You may determine with a road surface reflection part. In this way, the road surface reflection portion is determined with higher accuracy.

  Next, determination of a road surface reflection portion using condition B will be described with reference to FIG. In FIG. 10, the frame at the time t210 is a bird's eye view image frame 210 generated by the current cycle process, and the frame at the time t200 is a bird's eye view image frame 200 generated by the process N cycles ago. Further, the feature point detected in the frame 200 is defined as a feature point Xb, and the feature point detected in the frame 210 is defined as a feature point Yb. N is an integer equal to or greater than 1, and is a preset value. In the determination using the condition B, the frame 200 corresponds to the first frame, the frame 210 corresponds to the second frame, the feature point Xb corresponds to the first feature point, and the feature point Yb corresponds to the second feature point.

  Even in the determination using the condition B, it is used that the feature point representing the road surface reflection portion is not detected at the corrected position after the egomotion correction. However, when the position of the feature point Yb detected in the front area of the current frame 210 is corrected by ego motion, the correction position in the frame 200 may be outside the frame in front of the vehicle 70. Therefore, in the determination using the condition B, the position of the feature point Xb detected in the past frame 200 is corrected by ego motion, and the correction position is calculated. In the determination using the condition B, when the feature point Yb does not exist at the correction position in the frame 210, the portion represented by the feature point Xb corresponding to the correction position where the feature point Yb does not exist in the frame 200 is determined on the road surface. It is determined as a reflection part. However, in order to remove noise in the processing of this cycle, it is necessary to determine the road surface reflection portion in the current frame 210.

  Therefore, in the determination using the condition B, it is further used that the feature point representing the road surface reflection portion is detected at the same position between frames. Specifically, in the determination using the condition B, the part represented by the feature point Yb detected at the same position as the feature point Xb determined as the road surface reflection part in the frame 200 in the frame 210 is determined as the road surface reflection part. .

<Effect>
According to this embodiment described above, the following effects are obtained.
(1) In the front area of the bird's eye view image, the correction position in the current frame is calculated with the feature point detected in the past frame as the first feature point. If the second feature point does not exist at the calculated correction position in the current frame, the portion represented by the first feature point corresponding to the correction position in which no second feature point exists in the past frame is a road surface reflection portion. Determined. Since the road surface reflection portion is detected at the same position between the frames, the second feature point detected at the same position as the first feature point determined as the road surface reflection portion in the past frame in the current frame is represented. The portion can be determined as a road surface reflection portion. Therefore, in the current frame, it is possible to recognize the lane markings with high accuracy except for the second feature points of the detected second feature points that are determined to be road surface reflection portions.

  (2) In the rear region of the bird's-eye view image, the feature point detected in the current frame is set as the first feature point, and the correction position in the past frame is calculated. In the past frame, when the second feature point does not exist at the calculated correction position, the portion represented by the first feature point corresponding to the correction position where the second feature point does not exist in the current frame is a road surface reflection portion. Can be judged. Therefore, in the current frame, it is possible to recognize the lane markings with high accuracy by excluding the first feature points of the detected first feature points that are determined to be road surface reflection portions.

  (3) In the bird's-eye view image area, the road surface reflection portion can be determined in the same manner as in the front area, and the marking line can be recognized by removing the road surface reflection portion from the detected feature points, and the same as in the rear area. Thus, it is possible to determine the road surface reflection portion and recognize the lane markings by removing the road surface reflection portion from the detected feature points.

  (4) Since the amount of movement of the vehicle 70 between frames changes according to the vehicle speed, the road surface reflection portion can be appropriately determined by setting the size of each region of the bird's eye view image according to the vehicle speed.

  (5) The greater the vehicle speed, the greater the amount of movement of the vehicle 70 between the frames. Therefore, the higher the vehicle speed, the smaller the intermediate area and the larger the front and rear areas. Can be determined.

(Other embodiments)
As mentioned above, although the form for implementing this invention was demonstrated, this invention is not limited to the above-mentioned embodiment, It can implement in various deformation | transformation.

  (A) In the above embodiment, three areas are set for the bird's-eye view image, but only the two areas of the front area and the rear area may be set without setting the intermediate area. In this case, as the vehicle speed increases, the amount of movement of the vehicle 70 between the frames increases. Therefore, if the front area is increased, noise can be appropriately removed.

  (B) The size of the two or three regions set for the bird's eye view image may be a constant size regardless of the vehicle speed of the vehicle 70. In this case, the image may be equally divided into two or three in the traveling direction, or may be divided at a preset ratio.

  (C) A plurality of functions of one constituent element in the above embodiment may be realized by a plurality of constituent elements, or a single function of one constituent element may be realized by a plurality of constituent elements. . Further, a plurality of functions possessed by a plurality of constituent elements may be realized by one constituent element, or one function realized by a plurality of constituent elements may be realized by one constituent element. Moreover, you may abbreviate | omit a part of structure of the said embodiment. In addition, at least a part of the configuration of the above embodiment may be added to or replaced with the configuration of the other embodiment. In addition, all the aspects included in the technical idea specified only by the wording described in the claim are embodiment of this invention.

  (D) In addition to the lane marking recognition device described above, a road surface reflection determination device, a program for causing a computer to function as the lane marking recognition device, a non-transition actual recording medium such as a semiconductor memory in which the program is recorded, and a lane marking The present invention can also be realized in various forms such as a recognition method and a road surface reflection determination method.

  10 ... camera, 20 ... ECU, 70 ... vehicle.

Claims (6)

A lane marking recognition device (30) for recognizing a lane marking drawn on a road surface so as to divide a lane from a captured image obtained by capturing the surroundings of the vehicle with a camera (10) mounted on the vehicle (70). And
A feature point detector (S100) for detecting a feature point representing a brighter part than the road surface from the captured image;
A behavior acquisition unit (S110) for acquiring the behavior of the vehicle;
The vehicle of which the position of the first feature point, which is the feature point detected in the first frame, is obtained by setting the frames of the captured image at a time point away from a preset period as the first frame and the second frame. A correction unit (S140 to S160) that corrects the position in the second frame using the behavior and calculates the correction position;
When the second feature point that is the feature point detected in the second frame does not exist in the calculated correction position, it corresponds to the correction position in which the second feature point does not exist in the first frame. A reflection determination unit (S140 to S160) that determines a portion represented by the first feature point to be a road surface reflection portion ,
In the captured image, an area setting unit (S120) that sets a plurality of areas including a front area and a rear area with respect to the traveling direction of the vehicle;
A lane marking recognition unit that recognizes the lane markings from the feature points of the feature points detected by the feature point detection unit, excluding the feature points of the portion determined to be a road surface reflection portion by the reflection determination unit. (S180 to S200),
In the set front area,
The first frame is a frame at a time before the current time, and the second frame is a current frame,
The reflection determination unit is the same as the first feature point corresponding to the correction position where the second feature point does not exist when the second feature point does not exist at the calculated correction position in the two frames. A portion represented by the second feature point detected at a position is determined as a road surface reflection portion;
The lane marking recognition unit extracts the lane marking from the second feature point except for the second feature point of the detected second feature point that is determined as the road surface reflection portion by the reflection determination unit. Recognizing lane marking recognition device. In the set rear region,
The first frame is a current frame, and the second frame is a frame before the current time,
The lane marking recognition unit recognizes the lane marking from the first feature point excluding the first feature point of a portion of the first feature point determined to be a road surface reflection portion by the reflection determination unit. The lane marking recognition apparatus according to claim 1 . The region setting unit sets a front region, a central region, and a rear region with respect to the traveling direction of the vehicle,
In the central area set by the area setting unit,
The reflection determination unit determines the road surface reflection portion with the central region as the front region, and determines the road surface reflection portion with the central region as the rear region.
The division line recognition unit performs recognition of the partition lines the central area as the area in front of, to recognize the division line of the central area as the rear area, according to claim 2 A lane marking recognition device. The lane marking recognition device according to any one of claims 1 to 3 , wherein the area setting unit sets the size of each area according to the vehicle speed of the vehicle acquired by the behavior acquisition unit. The lane marking recognition device according to claim 1 or 2 , wherein the area setting unit sets the area ahead as the vehicle speed of the vehicle acquired by the behavior acquisition unit increases. The lane marking recognition device according to claim 3 , wherein the region setting unit sets the central region to be smaller as the vehicle speed of the vehicle acquired by the behavior acquisition unit is higher.

Images