KR101326502B1 - Device and method for keeping one's lane - Google Patents

Abstract

The present invention more clearly detects lanes (= white lines) of both blind spots (blind corners) while the vehicle is being driven, and displays the detected lanes on a blind corner monitor so that the driver can determine whether the vehicle is traveling in both lanes or lanes. The present invention relates to a lane departure detecting apparatus and method for visually recognizing whether the vehicle travels away from the vehicle.

To this end, the present invention is a camera cover which is installed on the front grille of the vehicle protruding, camera windows formed on both sides of the front end of the camera cover, a prism mounted toward the camera window from the inside of the camera cover, the camera A blind corner camera composed of a lens mounted behind the prism inside the cover, and an imaging device mounted behind the lens inside the camera cover; An image processing unit connected to the imaging device so as to be capable of signal exchange; A blind corner monitor mounted on the front side of a driver's seat in a cabin while being connected to the image processor so as to exchange signals; It provides a lane departure detection apparatus and method comprising a.

Lane Departure, Detector, Camera, Blind Corner, Image Processing Unit

Description

Device and method for keeping one's lane

1 is a schematic diagram showing a lane departure detecting apparatus and an image capturing unit according to the present invention;

2 is a perspective view showing the appearance of the lane departure detecting apparatus and the image pickup unit according to the present invention;

3 and 4 are graphs illustrating a lane departure detection method according to the present invention;

5 and 6 are schematic views illustrating the operation relationship of the lane departure detection apparatus according to the present invention;

7 is a schematic diagram illustrating a lane departure warning apparatus or lane keep system that recognizes a lane as a prior art;

8 is a schematic diagram illustrating a brand corner monitor system as another prior art;

9 is a schematic diagram illustrating an embodiment of an image pickup apparatus for detecting lane departure according to the present invention;

FIG. 10 is a flowchart for explaining a lane departure detection operation by the imaging device of FIG. 9;

FIG. 11 is a schematic view for explaining an image pickup apparatus for photographing a lane on a road surface at a straight distance (closest distance) from a side part of a vehicle, as another conventional technique;

Fig. 12 is a schematic diagram illustrating another image pickup apparatus for photographing a lane on a road surface at a straight distance (closest distance) from a side portion of a vehicle, as another conventional technique.

FIG. 13 is a flowchart illustrating a lane departure detection operation performed by the scratch apparatus of FIG. 11 or FIG. 12.

Fig. 14 is a diagram for explaining an image transmission method according to the NTSC standard.

<Explanation of symbols for the main parts of the drawings>

10: camera cover 12: camera window

14 prism 16: lens

18: imaging device 20: image processing unit

22: blind corner monitor 100: blind corner camera

The present invention relates to a lane departure detecting apparatus and method, and more particularly, to detect lanes (= white lines) of both blind spots (blind corners) while driving a vehicle, and display the detected lanes on a blind corner monitor, and lane departure The present invention relates to a lane departure detecting apparatus and method for allowing a driver to visually and audibly recognize whether a vehicle is driving in both lanes or driving away from the lane.

As a prior art, a lane departure warning device or lane keep system that recognizes a lane by image processing an image captured by a front camera or a rear view camera has been proposed (see FIG. 7).

As another conventional technique, a brand corner monitor system has been proposed that assists the identification of vehicles, bicycles, and pedestrians by installing a camera in the center of the front grill of the vehicle and approaching an intersection or T-shaped intersection where the front is not visible ( 8).

Common problems with these prior arts are as follows.

1) Image recognition processing techniques for lane detection generally include binarization → edge extraction of contours or curves of images → hough transformation, lane extraction by artificial intelligence, and vehicle upside. Although there is a lane extraction method by a viewpoint transformation process and the like, either method has a disadvantage in that a large amount of calculation is required.

2) When driving at night, a halation phenomenon occurs in the camera of the vehicle caused by the headlights of the opposite vehicle, which makes it difficult to detect the lane.

3) When driving at night, the rear view camera has difficulty in clearly photographing the lane because there is no separate lighting means for illuminating the lane.

4) In rainy weather, water droplets may be deposited on the camera, which may cause a decrease in lane recognition of the camera.

As another conventional technology, as shown in FIG. 11, in order to detect a lane on a road surface adjacent to a side surface of a vehicle, a prism and an image pickup device are mounted on the front side of the vehicle horizontally with each other along the front and rear directions, and the image on the right side of the vehicle. And left-side images are captured by the image pickup device through a prism.

However, in the lane detection apparatus shown in FIG. 11, a lane immediately adjacent to the side of the vehicle (directly) may exceed an angle of view, and in this way, a lane existing on the road surface of the vehicle directly near the vehicle may be detected. It may be difficult to photograph.

As another conventional technique, in order to detect a lane on a road surface adjacent to a side portion of a vehicle as shown in FIG. 12, the prism and the image pickup device are mounted to the front side of the vehicle at an angle to each other along the front and rear directions, at the right side. There is a method in which the room image and the left room image are captured by the image pickup device through a prism.

However, in the lane detecting apparatus shown in FIG. 12, the prism and the image pickup device are inclined at a predetermined angle to each other along the front and rear directions, so that it is possible to photograph the lane existing on the road surface of the vehicle directly, but the lane to be photographed Since it is not horizontal, there is a disadvantage that image processing optimization is difficult.

In addition, as shown in the flowchart of FIG. 13, the lane detecting apparatus illustrated in FIG. 11 or 12 stores all screens in an image input area → an image storage area → a linear part extraction operation → a control in a white line position information. Since the operation flow is made, there is a disadvantage in that a storage area for temporarily storing the entire image is required because the transmission order of the image information and the direction of the lane for performing the processing are different.

The present invention has been made in view of the above, and a blind corner camera capable of photographing the left and right directions of the vehicle, that is, the blind corner area, using a prism or the like is installed on the front grille of the vehicle, and the photographed left and right lanes Displays lanes horizontally through blind corner monitors in the cabin and alerts on lane departures, enabling lane drivers to visually and audibly recognize whether the vehicle is traveling in both lanes or out of lanes It is an object of the present invention to provide an apparatus and method.

Lane departure detection apparatus of the present invention for achieving the above object: a camera cover is installed on the front grille portion of the vehicle protruding, a camera window formed on both sides of the front end of the camera cover, the camera inside the camera cover A blind corner camera comprising a prism mounted toward a window, a lens mounted behind the prism inside the camera cover, and an imaging device mounted behind the lens inside the camera cover; An image processing unit connected to the imaging device so as to be capable of signal exchange; A blind corner monitor mounted on the front side of a driver's seat in a cabin while being connected to the image processor so as to exchange signals; It is configured to include, characterized in that the lanes in both blind corner areas of the vehicle is detected by the blind corner camera to be displayed horizontally on the blind corner monitor.

In a preferred embodiment, the angle of the camera window relative to the longitudinal direction of the camera cover is formed to form an acute angle with the longitudinal direction of the camera cover so that rain water drops fall from the camera window while driving the vehicle.

Another lane departure detecting apparatus of the present invention for achieving the above object is:

A polarizing prism mounted horizontally in front of the vehicle; An image pickup device mounted directly at the rear of the polarizing prism with the upper end facing forward and the lower end tilted at an angle toward the rear; An image processing unit connected to the imaging device so as to be capable of signal exchange; A blind corner monitor mounted on the front side of a driver's seat in a cabin while being connected to the image processor so as to exchange signals; It is configured to include, it characterized in that the lanes of the straight in both blind corner areas of the vehicle is detected by the blind corner camera to be displayed horizontally on the blind corner monitor.

A lane departure detecting method of the present invention for achieving the above object comprises the steps of: inputting an image for both blind corner areas of the vehicle to the imaging device through the prism and lens of the blind corner camera; Image processing by the image processing unit the image signal obtained from the image capturing device taking the blind corner region, extracting the lane position on the road surface, and calculating the distance between the lane position and the vehicle; The lane position of the left and right calculated from the image processing unit is characterized in that it comprises the step of acoustically warning while displaying a horizontal line on the blind corner monitor.

In a preferred embodiment, the calculating step of the image processing unit includes the steps of: calculating the density distribution by integrating the image signal obtained from the imaging device of the blind corner camera in the horizontal direction; The maximum concentration in the concentration distribution is characterized by including the step of extracting the lane position from the indicated peak position.

In a more preferred embodiment, the calculating step of the image processing unit further comprises the step of determining the distance between the vehicle and the lane according to the size of the lane width photographed in order to increase the reliability of the lane position.

More specifically, the determining of the distance between the vehicle and the lane of the image processing unit may include determining that the distance between the lane and the vehicle is close when the photographed lane width is large and that the distance between the lane and the vehicle is far when the width of the photographed lane is narrow. It is characterized by determining whether the vehicle is in the lane or outside the lane.

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

The present invention detects the lane (white line) position and the distance between the vehicle and the lane in both areas (blind corner area) of the vehicle, while the halation phenomenon of the camera according to the headlights of the opposite vehicle during the night driving The main point is that the lanes can be clearly photographed while being prevented, and in rainy weather, water droplets on the camera can be easily removed, and the vehicle can be easily detected whether the vehicle is in the lane or the lane has been separated.

1 is a schematic diagram illustrating a lane departure detecting apparatus and an image capturing unit according to the present invention, and FIG. 2 is a perspective view showing appearance of a lane departure detecting apparatus and an image capturing unit according to the present invention.

The blind corner camera 100 of the lane departure detecting apparatus according to the present invention is installed at the front grill part of the vehicle.

More specifically, the front end of the case-shaped camera cover 10 constituting the skeleton of the blind corner camera 100 is to be in a state protruding from the front grill of the vehicle.

In particular, the camera window 12 is formed on both sides of the front end of the camera cover 10, the angle of the camera window 12 is formed to form an acute angle with respect to the front and rear longitudinal direction of the camera cover 10.

That is, the pair of camera windows 12 are formed to be inclined from the inner side to the outer side from the front side to the rear side.

Therefore, even when water droplets such as rainwater are buried in the camera window 12 during rain, the droplets can be easily removed by the strength of the wind blowing from the front while driving.

Inside the camera cover 10, the prism 14, the lens 16, and the imaging device 18 are sequentially arranged and fixedly mounted from the front in the vehicle body front-rear direction.

More specifically, the prism 14 is disposed inside the camera cover 10 toward the camera window 12, the lens 16 is disposed behind the prism 14, and the rear of the lens 16. An imaging device 18 for photographing an actual image is arranged in the image.

On the other hand, the image processing unit 20 is connected to the image pickup device 18 so that the signal can be exchanged. The image processing unit 20 receives an image signal photographed by the image pickup device 18, and thus the lane width and the lane and the vehicle. It is to calculate the distance between them.

In addition, the image processing unit 20 is connected to the blind corner monitor 22 mounted on the front of the driver's seat in the vehicle interior, and the blind corner monitor 22 is laned according to the operation result signal of the image processing unit 20. Will be displayed on the horizontal line.

Hereinafter, the lane departure detection method by the lane departure detection apparatus of the present invention including the above configuration will be described sequentially.

First, photographing of the blind corner camera 100 for both areas of the vehicle, that is, the blind corner area, while driving is started.

That is, the imaging device 18 captures an image of the left and right blind corner areas of the vehicle coming through the prism 14 and the lens 16 in real time, and the image signal is transmitted to the image processing unit 20. do.

Next, the image processing unit 20 performs image processing on the image signal received from the imaging device 18, extracts the lane position and width on the road surface, and calculates the distance between the lane position and the vehicle.

In the method of extracting the lane position of the image processing unit 20, as shown in the attached graph of FIG. 3, the image signal obtained from the imaging device 18 is integrated in the horizontal direction to obtain a density distribution, and among these density distributions. The maximum concentration extracts the lane position from the peak position indicated, and calculates the width of the peak position to determine the lane width.

In addition, the image processing unit 20 determines the distance between the vehicle and the lane according to the size of the lane width photographed in order to increase the reliability of the lane position, as shown in the graph of FIG. If the distance between the vehicle and the lane is far, and if the photographed lane width is large, the distance between the vehicle and the lane is close using the inverse relationship.

That is, based on the perspective, close objects are photographed in a large size, and distant objects are photographed in a small size. When the photographed lane width is small, the distance between the vehicle and the lane is maintained. Larger indicates that the vehicle is in close proximity to the lane.

Accordingly, the image processing unit 20 determines that the distance between the lane and the vehicle is close when the photographed lane width is large, and determines that the distance between the lane and the vehicle is far when the width of the photographed lane is narrow, while the vehicle is in the lane or Determine if you are outside the lane.

Next, the left and right lanes are displayed on the blind corner monitor 22 according to the operation result signal received from the image processing unit 20, and the left and right lanes are displayed in horizontal form with different widths. .

For example, as shown in FIG. 5, when the distance between the right lane and the vehicle is far, the width of the lane on the right is narrowly displayed, and when the distance between the lane on the left and the vehicle is close, the width of the lane on the left is displayed wide.

In addition, based on the lower end of the blind corner monitor 22, if the distance between the right lane and the vehicle is far, the right lane is moved upward and displayed, and the left lane is lower (bottom side of the blind corner monitor). And are displayed.

As a result, lanes on the left and right sides are displayed on the blind corner monitor 22 while moving up and down with different widths according to the distance between each lane and the vehicle, so that the driver can easily recognize the position and departure information of the lane. Will be.

Of course, if the distance between the left and right lanes and the vehicle is the same, the left and right lanes are displayed while maintaining the same height.

As another embodiment, the lanes displayed on the blind corner monitor 22 may be displayed with different luminance. For example, when the left lane is kept far from the vehicle, the lane is displayed at low luminance, and the right lane is displayed with the vehicle. If it is maintained at a short distance, it is displayed at a high luminance, so that the driver can easily recognize lane position and departure information.

On the other hand, even if the headlights of the opposite vehicle approaching from the opposite lane toward the camera of the lane departure detection apparatus according to the present invention at night driving, as shown in Figure 6 the camera of the present invention to shoot both blind corner areas of the vehicle Therefore, it is not influenced by much, and eventually, the halation of the camera can be prevented.

Here, another embodiment of the lane departure detecting apparatus of the present invention will be described.

9 is a schematic view illustrating another embodiment of the image pickup apparatus for detecting lane departure according to the present invention, and FIG. 10 is a flowchart illustrating a lane departure detection operation performed by the image pickup apparatus of FIG. 9.

As shown in FIG. 9, a polarizing freeze and an image pickup device are mounted to the front of the vehicle (eg, the front grille) so as to photograph a straight lane from the side of the vehicle, and the polarizing prism is horizontal along the front and rear directions. The image pickup device is mounted inclined at an angle.

That is, the image pickup device is mounted to be inclined at an angle immediately after the top of the polarizer prism and the bottom thereof toward the rear of the polarizing prism, so that it is easy to photograph the lanes straight from the side of the vehicle.

Accordingly, the captured image signal is processed by the image processing unit to extract the lane position and width on the road surface, and calculate the distance between the lane position and the vehicle, and is displayed horizontally on the blind corner monitor.

More specifically, the image processing unit accumulates the image signal in the horizontal direction to obtain a vertical density distribution, extracts the lane position from the peak position of the density distribution, and specifies the vehicle position, that is, the horizontal synchronization signal By integrating at every half cycle, the horizontal density distribution in the left and right images can be simultaneously created, and the horizontal scanning method that conforms to the NTSC method, which is a moving image transmission standard, does not require a storage area for storing all screens. The circuit configuration can be made, and manufacturing cost can be reduced.

Here, another embodiment of the lane departure detecting apparatus according to the present invention will be described.

11 is a schematic view for explaining another embodiment of the image pickup apparatus for detecting lane departure according to the present invention.

As shown in FIG. 11, a flat light prism and an image pickup device are mounted to the front of the vehicle (eg, the front grill) so as to photograph a straight lane from the side of the vehicle, and the polarization prism and the image pickup device are horizontally aligned along the front and rear directions. Device.

In this method, as shown in the photographed image shown in Fig. 11, the lane is photographed in the horizontal direction with respect to the image, but the lane existing in the immediate vicinity of the vehicle cannot be photographed.

12 is a schematic diagram illustrating another embodiment of the image pickup apparatus for detecting lane departure detection according to the present invention, and another embodiment of the embodiment shown in FIG. 13.

As shown in FIG. 12, the polarizing prism and the image pickup device are mounted in front of the vehicle (eg, the front grill) so that the lanes of the straight line can be photographed from the side of the vehicle, and the polarizer prism is mounted parallel to the image pickup device. Thus, the image pickup device and the polarizing prism are mounted to be inclined at a predetermined angle.

In this method, as shown in the captured image shown in Fig. 11, since the lane is photographed in an inclined direction with respect to the image, as shown in the operation flowchart shown in Fig. 13, the lane position is detected unless all screens are stored in the image storage area. Can not.

As described above, according to the lane departure detecting apparatus and method according to the present invention,

1) By installing a blind corner camera capable of photographing blind corner areas on the front grille of the vehicle and displaying the photographed left and right lanes horizontally through a blind corner monitor in the cabin, the driver can drive the vehicle in both lanes. It is easy to visually recognize whether the vehicle is traveling or driving away from the lane.

2) When driving at night, halation of the camera due to the headlights of the opposite vehicle can be prevented and lane detection can be clearly performed without a separate lighting means.

3) In case of rain, even if water droplets get on the camera window, there is an advantage that water droplets are eliminated due to the running wind in front.

Claims (7)

A camera cover protruding from the front grille of the vehicle, camera windows formed on both sides of the front end of the camera cover, a prism mounted toward the camera window from the inside of the camera cover, and mounted behind the prism inside the camera cover. A blind corner camera composed of a lens and an imaging device mounted behind the lens inside the camera cover; Connected to the image pickup device so as to be capable of signal exchange, and calculating the density distribution by integrating the image signal obtained from the image pickup device of the blind corner camera in the horizontal direction and extracting the lane position from the peak position representing the maximum density among the density distributions. An image processing unit; A blind corner monitor mounted on the front side of a driver's seat in a cabin while being connected to the image processor so as to exchange signals; The lane departure detection apparatus of claim 2, wherein lanes in both blind corner areas of the vehicle are detected by the blind corner camera and displayed horizontally on the blind corner monitor. The lane departure detecting apparatus according to claim 1, wherein an angle of the camera window with respect to the longitudinal direction of the camera cover is formed to form an acute angle with the longitudinal direction of the camera cover such that rain water drops fall from the camera window while driving the vehicle. delete An image of both blind corner regions of the vehicle is input to the imaging device through the prism and the lens of the blind corner camera; Image processing by the image processing unit the image signal obtained from the image capturing device taking the blind corner region, extracting the lane position on the road surface, and calculating the lane position and width and the distance between the lane and the vehicle;   Displaying left and right lane positions calculated from the image processing unit as horizontal lines on a blind corner monitor; Lt; / RTI &gt; The calculating step of the image processing unit is: Obtaining a density distribution by integrating the image signal obtained from the imaging device of the blind corner camera in a horizontal direction; Extracting lane positions from peak positions representing the maximum concentrations in the concentration distribution; Lane departure detection method comprising a. delete The method of claim 4, wherein the calculating step of the image processing unit: Determining a distance between the vehicle and the lane according to the size of the lane width photographed to increase the reliability of the lane position; Lane departure detection method further comprising. The method according to claim 4 or 6, The determining of the distance between the vehicle and the lane of the image processor may include: If the lane width is large, the distance between the lane and the vehicle is close. If the width of the lane is narrow, the distance between the lane and the vehicle is determined to be far, and the lane departure is characterized in that it is determined whether the vehicle is in or outside the lane. Detection method.

Images