KR102326184B1 - Calibration apparatus and method of rear view camera - Google Patents

Abstract

The present invention discloses an apparatus and method for calibrating a rear camera. A rear camera calibration apparatus of the present invention includes: a vehicle state input unit for receiving a vehicle state input from a vehicle control device; a viewpoint conversion lookup table for storing a lookup table for converting a viewpoint of a rear image; And when the vehicle state is input from the vehicle state input unit and the behavior condition for calibration is satisfied, the rear image is input from the rear camera, the viewpoint is converted to the top view through the viewpoint conversion lookup table, and the pitch angle is corrected based on the angle between the left and right lanes. , a control unit that converts the viewpoint to the front view to correct the roll angle based on the difference between the left and right lane angles, and corrects the yaw angle based on the vertical angle of the lane.

Description

Reversing camera calibration device and method {CALIBRATION APPARATUS AND METHOD OF REAR VIEW CAMERA}

The present invention relates to an apparatus and method for calibrating a rear camera, and more particularly, to an apparatus and method for calibrating a rear camera for performing image distortion correction by analyzing an image of a rear camera while driving in a straight line at a constant speed of a vehicle.

In general, an Around View Monitoring (AVM) system provides a system that provides a single image with surrounding images acquired from front, rear, left and right cameras of a vehicle so that the driver can recognize the surrounding situation. say the system The AVM system tends to be popularized as a driving support system that assists the driver by providing convenience, such as providing the driver with an image of a blind spot of a vehicle and a rear image when parking.

In order to fuse images obtained from a plurality of cameras mounted on a vehicle into one image, internal parameters (light center and focal length, etc.) and external parameters (installation position and installation angle, etc.) of the camera must be provided, and the internal parameters are Since it is invariant, the value obtained once through calibration can be continuously used, but external parameters can be easily changed by external shocks, so continuous correction is required.

Among the external parameters, the installation position does not change significantly due to the installation jig even if the camera is replaced. do.

Conventionally, in a calibration field in which a specific calibration pattern is drawn on the ground, a skilled worker performs the calibration of the camera using separate equipment. This method has the advantage of obtaining precise results by accurately knowing the relative position of a specific marker on the calibration pattern in advance, but it is mainly accompanied by a spatial constraint that requires a space having an area of more than a certain area, such as a factory where vehicles are shipped, Since the time and cost of the operator are consumed in that the operation is performed by a skilled worker using separate equipment, a system for improving the convenience of the calibration operation is required.

Background art of the present invention is disclosed in Korean Patent Publication No. 10-2011-0055412 (published on May 25, 2011, calibration device).

The present invention has been devised to improve the above problems, and an object of the present invention according to one aspect is to analyze the image of the rear camera during constant speed straight driving of a vehicle and perform image distortion correction by analyzing the image of the rear camera, and to provide a way

A rear camera calibration apparatus according to an aspect of the present invention includes: a vehicle state input unit for receiving a vehicle state input from a vehicle control device; a viewpoint conversion lookup table for storing a lookup table for converting a viewpoint of a rear image; And when the vehicle state is input from the vehicle state input unit and the behavior condition for calibration is satisfied, the rear image is input from the rear camera, the viewpoint is converted to the top view through the viewpoint conversion lookup table, and the pitch angle is corrected based on the angle between the left and right lanes. , a control unit that converts the viewpoint to the front view to correct the roll angle based on the difference between the left and right lane angles, and corrects the yaw angle based on the vertical angle of the lane.

In the present invention, based on the vehicle speed and the yaw angle or steering angle, the control unit satisfies the behavior condition when the vehicle speed is greater than the set speed, the acceleration is less than the set acceleration, and the yaw or steering angle is less than the set angle is maintained for a set time based on the vehicle speed and the yaw angle or steering angle. It is characterized by judging by

In the present invention, the viewpoint conversion lookup table includes a front view lookup table for converting a viewpoint from a rear image to a front view and a top view lookup table for converting a viewpoint from a rear image to a top view.

In the present invention, the control unit includes: a viewpoint converting unit for converting a viewpoint to a top view or converting a viewpoint to a front view through a viewpoint transformation lookup table for the rear image; a lane detection unit for detecting a lane in the top view image or the front view image in which the viewpoint is converted through the viewpoint converting unit; a pitch angle correcting unit that determines whether there are parallel lanes based on the angle between the left and right lanes detected from the top view image by the lane detection unit and updates the pitch angle of the viewpoint conversion lookup table; a roll angle correcting unit that updates the roll angle of the viewpoint conversion lookup table based on the difference between the left and right lane angles based on the vanishing point of the left and right lanes detected by the lane detection unit from the front view image; and a yaw angle correction unit for updating the yaw angle of the viewpoint conversion lookup table based on the vertical angle at which the lane detected from the top view image by the lane detection unit is deviated from the vertical line.

In the present invention, the lane detection unit is characterized in that the effective lane is detected by removing the erroneously detected lane by analyzing statistical characteristics of the position and direction of the detected lane.

A method of calibrating a rear camera according to another aspect of the present invention includes the steps of: a control unit receiving a vehicle state from a vehicle state input unit and determining whether a behavior condition of the vehicle for calibration is present; receiving, by the controller, a rear image from the rear camera when the vehicle behavior condition is satisfied; detecting a lane by the controller converting a viewpoint into a top-view image based on a viewpoint transformation lookup table for the input rear image; correcting the pitch angle by determining, by the controller, whether the pitch angle is distorted based on the angle between the left and right lanes detected from the top view image; detecting, by the controller, whether the pitch angle is distorted, and if normal, converting a viewpoint into a front view image based on a viewpoint transformation lookup table for the rear image to detect a lane; correcting the roll angle by determining, by the controller, whether the roll angle is distorted based on the difference between the left and right lane angles detected from the front view image; and correcting the yaw angle by determining whether the roll angle is distorted by the control unit and determining whether the yaw angle is distorted based on the vertical angle of the lane if it is normal.

In the present invention, the step of determining the behavior condition includes, by the controller, based on the vehicle speed input from the vehicle state and the yaw or steering angle, the vehicle speed is greater than or equal to the set speed, the acceleration is less than or equal to the set acceleration, and the yaw or steering angle is less than or equal to the set angle. It is characterized in that it is determined whether the set time is maintained.

In the detecting of the lane in the present invention, the control unit analyzes the statistical characteristics of the position and direction of the detected lane and removes the erroneously detected lane to detect an effective lane.

In the present invention, the step of correcting the pitch angle by determining whether the pitch angle is distorted is that the control unit determines whether there are parallel lanes based on the angle between the left and right lanes and reflects the pitch angle for parallelism to update the viewpoint transformation lookup table. characterized.

In the present invention, in the step of correcting the roll angle by determining whether the roll angle is distorted, the control unit determines whether the difference between the left and right lane angles is equal to or less than the set difference value based on the vanishing point of the left and right lanes, and reflects the difference to the roll angle to change the viewpoint. It is characterized by updating the table.

In the present invention, in the step of correcting the yaw angle by determining whether the yaw angle is distorted, the control unit calculates the vertical angle of the deviated lane based on the vertical line, determines whether the angle is less than or equal to the set angle, and reflects the vertical angle of the lane to the yaw angle. is characterized in that it is updated.

The apparatus and method for calibrating a rear camera according to an aspect of the present invention perform image distortion correction by analyzing the image of the rear camera during constant speed straight driving of a vehicle, so that even if an external parameter is changed, it can be automatically corrected, so calibration is performed when reversing. In addition to reducing the load for the camera, it can be corrected even if the rear camera is incorrectly installed during the process, thereby improving convenience.

1 is a block diagram illustrating an apparatus for calibrating a rear camera according to an embodiment of the present invention.
2 is a view for explaining a viewpoint-converted image in the calibration apparatus of the rear camera according to an embodiment of the present invention.
3 is a view showing a rear camera installed in a vehicle in the apparatus for calibrating the rear camera according to an embodiment of the present invention.
4 is a view showing a viewpoint-converted image of the rear image in the apparatus for calibrating the rear camera according to an embodiment of the present invention.
5 is an exemplary view showing a pitch angle correction state in the apparatus for calibrating the rear camera according to an embodiment of the present invention.
6 is an exemplary view showing a roll angle correction state in the calibration apparatus of the rear camera according to an embodiment of the present invention.
7 is an exemplary diagram illustrating a yaw angle correction state in the apparatus for calibrating a rear camera according to an embodiment of the present invention.
8 is a flowchart illustrating a method of calibrating a rear camera according to an embodiment of the present invention.

Hereinafter, an apparatus and method for calibrating a rear camera according to the present invention will be described with reference to the accompanying drawings. In this process, the thickness of the lines or the size of the components shown in the drawings may be exaggerated for clarity and convenience of explanation. In addition, the terms to be described later are terms defined in consideration of functions in the present invention, which may vary according to the intention or custom of the user or operator. Therefore, definitions of these terms should be made based on the content throughout this specification.

1 is a block diagram illustrating an apparatus for calibrating a rear camera according to an embodiment of the present invention. , FIG. 3 is a view showing a rear camera installed in a vehicle in the apparatus for calibrating the rear camera according to an embodiment of the present invention, and FIG. 4 is a viewpoint for the rear image in the apparatus for calibrating the rear camera according to an embodiment of the present invention It is a view showing a converted image, and FIG. 5 is an exemplary view showing a pitch angle correction state in a calibration apparatus of a rear camera according to an embodiment of the present invention, and FIG. 6 is a rear view camera calibration according to an embodiment of the present invention. It is an exemplary view showing the roll angle correction state in the apparatus, and FIG. 7 is an exemplary view showing the yaw angle correction state in the rear camera calibration apparatus according to an embodiment of the present invention.

As shown in FIG. 1 , the apparatus for calibrating a rear view camera according to an embodiment of the present invention includes a vehicle state input unit 20 , a viewpoint conversion lookup table 40 , and a control unit 30 .

The vehicle state input unit 20 receives the vehicle state from a vehicle control device (not shown) and provides it to the controller 30 so that the controller 30 can determine a behavior condition for automatic calibration.

Here, the vehicle state input unit 20 may be connected to the vehicle control device through CAN communication to receive a vehicle speed, a yaw angle, or a steering angle.

The viewpoint conversion lookup table 40 converts the viewpoint to a top view viewpoint for the rear image as shown in (A) of FIG. 2 to generate a top view image as shown in (B) of FIG. and a front view lookup table for generating a front view image as shown in (c) of FIG. 2 by converting to a front view viewpoint.

The initial rear camera 10 is shipped with a pitch angle, a roll angle, and a yaw angle initially set when the vehicle is shipped as shown in FIG. 3 . However, when a change in the external environment of the vehicle occurs due to a change in tire pressure, trunk load, passenger boarding, camera misalignment, etc., the initially set image is distorted and may cause errors in various applications.

Accordingly, the control unit 30 receives the vehicle state from the vehicle state input unit 20 and, when the behavior condition for calibration is satisfied, receives the rear image from the rear camera 10 and views it as a top view through the viewpoint conversion lookup table 40 . to correct the pitch angle based on the angle between the left and right lanes, convert the viewpoint to the front view to correct the roll angle based on the difference between the left and right lane angles, and correct the yaw angle based on the vertical angle of the lane.

Here, the control unit 30 determines the constant speed straight driving state as a behavior condition for automatic calibration. Based on the vehicle speed and the yaw or steering angle from the vehicle state, the vehicle speed is greater than or equal to the set speed, and the acceleration is less than or equal to the set acceleration, yaw or When the state in which the steering angle is less than the set angle is maintained for the set time, it may be determined as a constant speed straight driving state.

As shown in FIG. 4 , the rear image input from the rear camera 10 when the vehicle is driving is in the acceleration state and the deceleration state, the vehicle is tilted forward or backward, and the lanes are not seen in parallel. Therefore, in the present embodiment, the constant speed straight running state is determined as the behavior condition.

Also, the control unit 30 may include a viewpoint converting unit 31 , a lane detecting unit 32 , a pitch angle correcting unit 33 , a roll angle correcting unit 34 , and a yaw angle correcting unit 35 .

As shown in FIG. 2 , the viewpoint converting unit 31 converts the viewpoint into a top view image through the viewpoint transformation lookup table 40 for the rear image, or converts the viewpoint into a front view image.

The lane detection unit 32 detects a lane in the top view image or the front view image in which the viewpoint is converted through the viewpoint conversion unit 31 .

Here, the lane detection unit 32 may analyze the statistical characteristics of the position and direction of the detected lane and remove the erroneously detected lane to detect an effective lane.

For example, when a lane is detected by analyzing an image, it is necessary to remove the erroneously detected lane because there is inevitably a erroneously detected lane in the detected lane. If the vehicle is driven along the lane, the change in the position and direction of the detected lane is limited, and if most of the detected lane is the lane of the driving lane, the characteristics of the position and direction of the lane form a cluster. By analyzing the statistical characteristics of the direction, it is possible to remove an out-of-cluster error, that is, an erroneously detected lane.

The pitch angle correction unit 33 determines whether there are parallel lanes based on the angle between the left and right lanes detected from the top view image by the lane detection unit 32 and updates and corrects the pitch angle of the viewpoint conversion lookup table 40 .

For example, as shown in FIG. 5 , when detecting a lane from the top view image, (A) is a distorted state because the pitch angle is shifted by +3 degrees, and (B) is a distorted state because the pitch angle is shifted by -3 degrees, and the left and right lanes are distorted. can't be parallel Accordingly, the pitch angle corrector 33 may update the viewpoint conversion lookup table 40 by applying the angle between the left and right lanes to the pitch angle to achieve parallelism as shown in (c).

The roll angle corrector 34 updates and corrects the roll angle of the viewpoint conversion lookup table 40 based on the difference between the left and right lane angles detected from the front view image by the lane detection unit 32 .

For example, the roll angle corrector 34 may calculate the difference between the left and right lane angles by drawing a vertical line based on the vanishing point where the lanes meet each other as shown in FIG. 6 . At this time, (A) is a distorted state due to a +5 degree roll angle, and (B) is a distorted state due to a -5 degree roll angle. Accordingly, the roll angle correcting unit 34 may update the viewpoint conversion lookup table 40 by reflecting 5 degrees, which is the difference between the left and right lane angles, to the roll angle as shown in (c).

The yaw angle correction unit 35 updates and corrects the yaw angle of the viewpoint conversion lookup table 40 based on the wrong vertical angle of the lane detected by the lane detection unit 32 based on the vertical line from the top view image.

For example, as shown in FIG. 7 , the yaw angle correcting unit 35 may calculate a yaw angle that is a vertical angle of a lane that is misaligned with respect to a vertical line. In (A), the yaw angle is distorted by +5 degrees, and in (B), the yaw angle is distorted by -5 degrees, and the vertical angle with respect to the vertical line of the lane deviates from the set angle of 0 degrees. Accordingly, the yaw angle correction unit 35 may update the viewpoint transformation lookup table 40 by reflecting the vertical angle of the deviated lane to the yaw angle as shown in (c).

As such, the control unit 30 automatically corrects the rear image of the rear camera 10 when the vehicle is driving at a constant speed while driving and outputs it to the image recognition application and the parking assistance display through the output unit 50 .

As described above, according to the apparatus for calibrating the rear camera according to the embodiment of the present invention, the image of the rear camera is analyzed and the image of the rear camera is analyzed while the vehicle is traveling straight at a constant speed, so that even if an external parameter is changed, automatic correction is possible. In this way, it is possible to reduce the load for performing calibration when reversing, and to improve convenience because it is possible to correct even if the rear camera is incorrectly installed during the process.

8 is a flowchart illustrating a method of calibrating a rear camera according to an embodiment of the present invention.

As shown in FIG. 8 , in the method of calibrating the rear view camera according to an embodiment of the present invention, first, the controller 30 receives the vehicle state of the vehicle while driving from the vehicle state input unit 20 ( S10 ).

The control unit 30 receiving the vehicle state input in step S10 determines whether the vehicle is in a constant speed straight driving state, which is a behavior condition of the vehicle for automatic calibration (S12).

Here, the control unit 30 determines the constant speed straight driving state as a behavior condition for automatic calibration. Based on the vehicle speed and the yaw or steering angle from the vehicle state, the vehicle speed is greater than or equal to the set speed, and the acceleration is less than or equal to the set acceleration, yaw or When a state in which the steering angle is equal to or less than the set angle is maintained for the set time, it may be determined as a constant speed straight driving state.

In step S12, if the constant speed straight driving state is not performed, as shown in FIG. 4 , the lane may be distorted due to the acceleration state and the deceleration state, and thus the calibration is not performed.

On the other hand, if the constant speed straight driving state in step S12, the control unit 30 receives the rear image from the rear camera 10 (S14).

After receiving the rear image in step S14, the controller 30 converts the rear image into a top view image by using the viewpoint conversion lookup table 40 as shown in (B) of FIG. 2 (S16).

After converting the viewpoint to the top view image in step S16, the controller 30 detects a lane from the top view image (S18).

Here, when detecting a lane, the controller 30 may analyze the statistical characteristics of the position and direction of the detected lane and remove the erroneously detected lane to detect an effective lane.

For example, when a lane is detected by analyzing an image, it is necessary to remove the erroneously detected lane because there is inevitably a erroneously detected lane in the detected lane. If the vehicle is driven along the lane, the change in the position and direction of the detected lane is limited, and if most of the detected lane is the lane of the driving lane, the characteristics of the position and direction of the lane form a cluster. By analyzing the statistical characteristics of the direction, it is possible to remove an out-of-cluster error, that is, an erroneously detected lane.

After detecting the lane in step S18, the controller 30 calculates the angle between the left and right lanes detected from the top view image (S20).

After calculating the angle between the left and right lanes in step S20, the controller 30 compares the angle between the left and right lanes to determine whether the lanes form parallel lanes (S22).

If parallel lanes are not formed in step S22, the viewpoint conversion lookup table 40 may be updated by correcting the pitch angle with the angle between the left and right lanes (S24).

For example, as shown in FIG. 5 , when detecting a lane from the top view image, (A) is a distorted state because the pitch angle is shifted by +3 degrees, and (B) is a distorted state because the pitch angle is shifted by -3 degrees, and the left and right lanes are distorted. can't be parallel Accordingly, the control unit 30 may update the viewpoint transformation lookup table 40 by applying the angle between the left and right lanes to the pitch angle to achieve parallelism as shown in (c).

In step S22, when it is determined whether the lane detected from the top view image forms a parallel lane and forms a parallel lane, the control unit 30 converts the rear image to the viewpoint transformation lookup table 40 as shown in (c) of FIG. As shown, the viewpoint is converted into a front view image (S26).

After the viewpoint is converted into the front view image in step S26, the controller 30 detects a lane from the front view image (S28).

Here again, when detecting a lane, the controller 30 may analyze the statistical characteristics of the position and direction of the detected lane and remove the erroneously detected lane to detect an effective lane.

After detecting the lane in step S28, the controller 30 calculates a difference value between the angles of the left and right lanes based on the vanishing point of the left and right lanes detected from the front view image (S30).

After calculating the difference between the left and right lane angles in step S30 , the controller 30 compares the difference between the left and right lane angles to determine whether the difference is equal to or less than a set difference value ( S32 ).

In the present embodiment, it may be determined whether the difference between the left and right lane angles is 0. That is, it may be determined whether the left and right lane angles are the same.

When the left and right lane angles are not the same in step S32, the viewpoint conversion lookup table 40 may be updated by reflecting the difference between the left and right lane angles in the roll angle (S34).

For example, as shown in FIG. 6 , the controller 30 may calculate a difference value between left and right lane angles by drawing a vertical line based on a vanishing point where lanes meet each other. At this time, (A) is a distorted state because the roll angle is twisted by +5 degrees, and (B) is a distorted state because the roll angle is twisted by -5 degrees. Accordingly, as shown in (c), the control unit 30 may update the viewpoint conversion lookup table 40 by reflecting 5 degrees, which is the difference between the left and right lane angles, to the roll angle.

In step S32, if the difference between the left and right lane angles is 0, that is, if it is normal by determining whether the roll angle is distorted, the control unit 30 calculates the vertical angle of the lane that is distorted based on the vertical line in the top view image (S36).

After calculating the vertical angle of the lane in step S36, the controller 30 determines whether the vertical angle of the lane is equal to or less than a set angle (S38).

In this embodiment, it may be determined whether the vertical angle is 0. That is, it can be determined whether the state is vertical.

In step S38 , it is determined whether the vertical angle of the lane is equal to or less than the set angle, and when it is equal to or less, the controller 30 may update the viewpoint conversion lookup table 40 by reflecting the vertical angle of the lane to the yaw angle ( S40 ).

For example, as shown in FIG. 7 , the controller 30 may calculate a yaw angle that is a vertical angle of a lane that is twisted with respect to a vertical line. In (A), the yaw angle is distorted by +5 degrees, and in (B), the yaw angle is distorted by -5 degrees, and the vertical angle with respect to the vertical line of the lane deviates from the set angle of 0 degrees. Accordingly, the controller 30 may update the viewpoint transformation lookup table 40 by reflecting the vertical angle of the deviated lane to the yaw angle as shown in (c).

As described above, according to the calibration method of the rear camera according to the embodiment of the present invention, the image of the rear camera is analyzed and the image of the rear camera is analyzed while the vehicle is traveling straight at a constant speed, so that even if an external parameter is changed, automatic correction is possible. In this way, it is possible to reduce the load for performing calibration when reversing, and to improve convenience because it is possible to correct even if the rear camera is incorrectly installed during the process.

Although the present invention has been described with reference to the embodiment shown in the drawings, this is merely exemplary, and it is understood that various modifications and equivalent other embodiments are possible by those of ordinary skill in the art. will understand

Therefore, the true technical protection scope of the present invention should be defined by the following claims.

10: rear camera 20: vehicle state input unit
30: control unit 31: viewpoint conversion unit
32: lane detection unit 33: pitch angle correction unit
34: roll angle correction unit 35: yaw angle correction unit
40: viewpoint transformation lookup table 50: output unit

Claims (11)

a vehicle state input unit receiving a vehicle state input from the vehicle control device;
a viewpoint conversion lookup table for storing a lookup table for converting a viewpoint of a rear image; and
When the vehicle state is input from the vehicle state input unit and the behavior condition for calibration is satisfied, the rear image is received from the rear camera and the viewpoint is converted into a top view through the viewpoint conversion lookup table, and the pitch angle is based on the angle between the left and right lanes. and a control unit that corrects the angle of the roll based on the difference between the left and right lane angles by changing the viewpoint to the front view, and corrects the yaw angle based on the vertical angle of the lane.
The control unit may include: a viewpoint converting unit for converting a viewpoint to the top view or converting a viewpoint to the front view through the viewpoint transformation lookup table for the rear image;
a lane detection unit configured to detect a lane in the top view image or the front view image in which the viewpoint is converted through the viewpoint converting unit; and
A roll angle correction unit for updating the roll angle of the viewpoint conversion lookup table based on the difference between the left and right lane angles based on the vanishing point of the left and right lanes detected from the front view image by the lane detection unit; calibration device.
The set time according to claim 1, wherein the control unit is based on the vehicle speed and the yaw angle or the steering angle from the vehicle state, wherein the vehicle speed is greater than or equal to the set speed, the acceleration is less than or equal to the set acceleration, and the yaw or the steering angle is equal to or less than the set angle. When maintained, the calibration device of the rear camera, characterized in that it is determined as a satisfactory state of the behavior condition.
The method of claim 1, wherein the viewpoint conversion lookup table comprises a front view lookup table for converting the viewpoint from the rear image to the front view and a top view lookup table for converting the viewpoint from the rear image to the top view. A calibration device for the rear camera.
According to claim 1, wherein the control unit,
a pitch angle correcting unit configured to update a pitch angle of the viewpoint conversion lookup table by determining whether there are parallel lanes based on the angle between the left and right lanes detected by the lane detection unit from the top view image; and
A yaw angle correction unit for updating the yaw angle of the viewpoint conversion lookup table based on a vertical angle at which the lane detected from the top view image by the lane detection unit is distorted with respect to the vertical line;
The apparatus of claim 1 , wherein the lane detection unit detects an effective lane by removing the erroneously detected lane by analyzing statistical characteristics of the position and direction of the detected lane.
determining, by the control unit, whether the vehicle state is a behavior condition for calibration by receiving the vehicle state from the vehicle state input unit;
receiving, by the controller, a rear image from a rear camera when the behavior condition of the vehicle is satisfied;
detecting a lane by converting a viewpoint into a top-view image based on a viewpoint transformation lookup table for the rear image input by the controller;
correcting the pitch angle by determining, by the controller, whether the pitch angle is distorted based on the angle between the left and right lanes detected from the top view image;
detecting, by the control unit, whether the pitch angle is distorted, and converting a viewpoint to a front view image based on the viewpoint transformation lookup table for the rear image when normal, and detecting a lane; and
Compensating the roll angle by determining whether the roll angle is distorted based on the difference between the left and right lane angles detected from the front view image, by the control unit;
In the step of correcting the roll angle by determining whether the roll angle is distorted, the control unit determines whether the difference between the left and right lane angles is equal to or less than a set difference value based on the vanishing point of the left and right lanes, and reflects the difference to the roll angle. A calibration method of a rear camera, characterized in that updating a viewpoint conversion lookup table.
The yaw angle of claim 6 , wherein the determining of the behavior condition comprises: the controller based on the vehicle speed input from the vehicle state and the yaw or steering angle, wherein the vehicle speed is greater than or equal to a set speed, and the acceleration is less than or equal to the set acceleration, and the yaw angle Alternatively, the calibration method of the rear camera, characterized in that it is determined whether the state in which the steering angle is equal to or less than a set angle is maintained for a set time.
The calibration of the rear camera according to claim 6, wherein, in the detecting of the lane, the control unit detects an effective lane by removing the erroneously detected lane by analyzing statistical characteristics of the position and direction of the detected lane. Way.
The method of claim 6, wherein the correcting of the pitch angle by determining whether the pitch angle is distorted comprises: the controller determines whether a parallel lane is based on an angle between left and right lanes and reflects the pitch angle for parallelism A calibration method of a rear camera, characterized in that updating a viewpoint conversion lookup table.
delete determining, by the control unit, whether the vehicle state is a behavior condition for calibration by receiving the vehicle state from the vehicle state input unit;
receiving, by the controller, a rear image from a rear camera when the behavior condition of the vehicle is satisfied;
detecting a lane by converting a viewpoint into a top-view image based on a viewpoint transformation lookup table for the rear image input by the controller;
correcting the pitch angle by determining, by the controller, whether the pitch angle is distorted based on the angle between the left and right lanes detected from the top view image;
detecting, by the control unit, whether the pitch angle is distorted, and converting a viewpoint to a front view image based on the viewpoint transformation lookup table for the rear image when normal, and detecting a lane;
correcting the roll angle by determining, by the controller, whether the roll angle is distorted based on the difference between the left and right lane angles detected from the front view image; and
Compensating the yaw angle by determining whether the roll angle is distorted by the control unit and determining whether the yaw angle is distorted based on the vertical angle of the lane if normal
In the step of correcting the yaw angle by determining whether the yaw angle is distorted, the control unit calculates the vertical angle of the lane that is distorted based on the vertical line, determines whether it is less than or equal to a set angle, and reflects the vertical angle of the lane to the yaw angle to reflect the time point A method for calibrating a rear view camera, comprising updating a conversion lookup table.

Images