Lane recognition device and method thereof

1. A lane recognition device, comprising:

a camera configured to capture an image in front of a vehicle; and

a controller configured to recognize a line based on a curved point of the line obtained from the image in front of the vehicle,

wherein the line is bent at a certain angle at the curved point without being smoothly bent, or the line is bent at the certain angle and includes discontinuous sections, and

wherein the certain angle includes at least one of an acute angle, an obtuse angle, or a right angle,

wherein the controller is configured to:

detect the line in a setting interval of a minimum distance closest to the vehicle to a maximum distance most distant from the vehicle in the image in front of the vehicle,

approximate the detected line to a line equation,

detect the curved point, when an error between the line and the line equation is greater than a first threshold,

generate one or more line equations based on the curved point of the line, and

recognize the line on a road based on the one or more line equations.

2. The lane recognition device of claim 1 , wherein the controller is further configured to convert the detected line from a camera coordinate system to a road coordinate system.

3. The lane recognition device of claim 1 , wherein the error is formed by the line and the line equation.

4. The lane recognition device of claim 1 , wherein the controller is further configured to:

generate a first error graph between the line and the line equation in a process of performing curve fitting of a line set configuring the line in ascending order of x values,

generate a second error graph between the line and the line equation in a process of performing curve fitting of the line set in descending order of the x values, and

set a point where the first error graph and the second error graph are met as the curved point.

5. The lane recognition device of claim 4 , wherein the controller is further configured to:

generate a first line equation by performing curve fitting of a line set in an interval of the minimum distance to the curved point in the ascending order of the x values, and

generate a second line equation by performing curve fitting of a line set in an interval of the maximum distance to the curved point in the descending order of the x values.

6. The lane recognition device of claim 5 , wherein the controller is configured to recognize the line based on the first line equation and the second line equation.

7. The lane recognition device of claim 1 , further comprising a radar configured to detect a location and a relative speed of a preceding vehicle which is located in front of the vehicle.

8. The lane recognition device of claim 7 , wherein the controller is further configured to:

calculate a driving trajectory based on the location and the relative speed of the preceding vehicle,

perform curve fitting of a set configuring the driving trajectory; and

reduce the first threshold, when a curvature rate of the driving trajectory is greater than a second threshold.

9. The lane recognition device of claim 1 , further comprising a communicator configured to obtain a driving trajectory of the road on which the vehicle is currently traveling from a driving trajectory server.

10. The lane recognition device of claim 9 , wherein the controller is configured to:

perform curve fitting of a set configuring the driving trajectory obtained via the communicator, and

reduce the first threshold, when a curvature rate of the driving trajectory is greater than a second threshold.

11. A lane recognition method, comprising:

capturing, by a camera of a vehicle; an image in front of the vehicle; and

recognizing, a controller of the vehicle, a line based on a curved point of the line obtained from the image in front of the vehicle,

wherein the line is bent at a certain angle at the curved point without being smoothly bent, or the line is bent at the certain angle and includes discontinuous sections, and

wherein the certain angle includes at least one of an acute angle, an obtuse angle, or a right angle,

wherein the recognizing of the line includes:

detecting the line in a setting interval of a minimum distance closest to the vehicle to a maximum distance most distant from the vehicle in the image in front of the vehicle;

approximating the line to a line equation;

detecting the curved point, when an error between the line and the line equation is greater than a first threshold;

generating one or more line equations based on the curved point of the line; and

recognizing the line on a road based on the one or more line equations.

12. The lane recognition method of claim 11 , wherein the detecting of the line includes converting the detected line from a camera coordinate system to a road coordinate system.

13. The lane recognition method of claim 11 , wherein the detecting of the curved point includes:

generating a first error graph between the line and the line equation in a process of performing curve fitting of a line set configuring the detected line in ascending order of x values;

generating a second error graph between the line and the line equation in a process of performing curve fitting of the line set in descending order of the x values; and

setting a point where the first error graph and the second error graph are met as the curved point.

14. The lane recognition method of claim 13 , wherein the generating the one or more line equations includes:

generating a first line equation by performing curve fitting of a line set in an interval of the minimum distance to the curved point in the ascending order of the x values; and

generating a second line equation by performing curve fitting of a line set in an interval of the maximum distance to the curved point in the descending order of the x values.

15. The lane recognition method of claim 11 , wherein the generating the one or more line equations further includes:

calculating a driving trajectory based on a location and a relative speed of a preceding vehicle;

performing curve fitting of a set configuring the driving trajectory; and

reducing the first threshold, when a curvature rate of the driving trajectory is greater than a second threshold.

16. The lane recognition device of claim 11 , wherein the generating the one or more line equations further includes:

obtaining a driving trajectory of the road on which the vehicle is currently traveling from a driving trajectory server;

performing curve fitting of a set configuring the driving trajectory; and

reducing the first threshold, when a curvature rate of the driving trajectory is greater than a second threshold.