JP2600427B2 - Traveling lane detection device for vehicles - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a traveling lane detecting device for a vehicle.

2. Description of the Related Art A conventional inter-vehicle distance traveling lane detecting device disclosed in Japanese Patent Application Laid-Open No. 63-14900 will be described below. FIG. 11 is a block diagram of a conventional apparatus.
1 is a lightness detecting means, 112 is a position detecting means, 113 is an instruction detecting means, 114 is a determining means, and 115 is a warning means. The position detection means 112 detects the position of the lane display on the road from the brightness distribution on the traveling road surface or the search line near the road detected by the lightness detection means 111, and the position of the lane display detected at this time is not in a predetermined state. Regardless of the instruction detection means 113
When the turning instruction from the driver is not detected, the determining unit 114 determines that the state is abnormal, and the warning unit 115
Gives a warning to the driver.

In the above-described conventional vehicle traveling lane detection device, the brightness is detected only on one search line provided on a road surface close to the vehicle, and the brightness is detected immediately before the vehicle. Since only the state of the lane can be detected, it is not possible to make a danger judgment in consideration of the direction of the vehicle to the traveling lane. Further, since only one search line is used, erroneous detection may occur due to noise.

The present invention solves the above-described problems, and provides a vehicle traveling lane detection device that can make a danger determination in consideration of the direction of a vehicle with respect to the traveling lane and that is less likely to be erroneously detected by noise. With the goal.

Means for Solving the Problems In order to solve the above-mentioned problems, the present invention provides a vehicle driving lane detecting device, a video camera mounted on a vehicle and imaging a road ahead of a predetermined distance from the vehicle, and an image captured by the video camera. Edge detecting means for detecting an edge portion of the road image, and edge image binarizing means for binarizing an edge image formed by the edge components detected by the edge detecting means using a predetermined threshold value; A lane display line detecting means for detecting a lane display line from the image binarized by the edge image binarizing means, and knowledge about a road structure used when the lane display line detecting means detects the lane display line is stored. A traveling lane during which the vehicle is traveling using the road structure knowledge storage means and the lane display line detected by the lane display line detecting means. Driving lane detecting means for detecting the turning lane, turning instruction detecting means for judging whether or not the driver intends to make a turn based on the operation state of the turn signal, the state of the driving lane detected by the driving lane detecting means and the turning Danger determining means for determining the degree of danger based on the presence or absence of a turning instruction detected by the instruction detecting means; and providing a warning to the driver of the vehicle when the danger determining means determines that the vehicle is in a dangerous state. It is composed of alarm means.

In the vehicle running lane detecting device of the present invention configured as described above, danger prediction can be performed in consideration of even the positional relationship between the vehicle and the running lane, and it is possible to prevent accidents during drunken driving such as drowsy driving. Therefore, the safety of the vehicle can be significantly improved. Further, since the device is hardly affected by noise, the reliability of the device can be improved.

Embodiment An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram of a vehicle driving lane detecting device according to the present embodiment. In the figure, reference numeral 11 denotes a video camera, which captures a road image a predetermined distance ahead of the vehicle. Reference numeral 12 denotes an edge detection unit that performs a filtering process or the like on a road image to detect an edge component. Reference numeral 13 denotes an edge image binarizing unit, which binarizes the detected edge image with a predetermined threshold. Reference numeral 14 denotes a lane display line detecting unit that detects a lane display line from a binarized edge image using knowledge about a road structure. Reference numeral 15 denotes a road structure knowledge storage unit, which stores knowledge about a road structure used for detecting a traveling lane. Reference numeral 16 denotes a traveling lane detecting means, which detects a lane in which the vehicle is traveling from the detected lane display line.
Reference numeral 17 denotes a turning instruction detecting means for judging the presence or absence of a turning intention by monitoring the operation of the turn signal by the driver. Reference numeral 18 denotes danger determining means for determining the degree of danger of the vehicle based on the detected state of the driving lane and the presence or absence of the driver's intention to turn. Reference numeral 19 denotes an alarm means for notifying the driver that the running state is abnormal by sound, light, or display on a monitor or the like.

Next, the operation of the embodiment of the present invention configured as described above will be described with reference to the control flowchart shown in FIG. First, in step 21, it is checked whether the switch of the device is ON. At this time, the switch is ON
If the driver does not perform any further processing, the driver operates the switch to avoid unnecessary operation of the device when the vehicle is running in the parking lot. Can be. If the switch of the apparatus is ON, the process proceeds to step 22, and a road image ahead of the vehicle by a predetermined distance from the video camera 11 is captured. FIG. 3 shows an example of a road image captured at this time.

Next, proceeding to step 23, the edge detecting means 12 performs an edge detecting process on the road image captured in step 22. For this edge detection processing, an edge detection filter represented by a SOBEL operator can be used. Then, the process proceeds to a step 24, wherein the edge image binarizing means 13 is used.
Is performed. The binarization of the image is performed by dividing the edge image obtained in step 23 by a predetermined threshold value.
This can be done by quantifying. FIG. 4 shows an image obtained by performing edge detection processing on the road image shown in FIG. 3 and binarizing it.

Next, the routine proceeds to step 25, where the lane display line detection processing in the lane display line detection means 14 is performed. In particular, on a highway or the like, a range from several tens meters to about 100 meters in front of the vehicle can be treated as a substantially straight line, and therefore, a straight line detecting means such as Hough transform can be used to detect the lane display line. FIG. 5 is a diagram showing a straight line on the XY plane, and FIG. 6 is a diagram showing a point on the straight line shown in FIG. 5 converted into a curve on the ρ-θ plane by the Hough transform. .. Pn on a straight line L on the XY plane as shown in FIG. 5 is obtained by the following equation: ρ = x · cos θ + y · sin θ Curve C1 ・ C2 on θ plane
.. Converted to Cn and using the fact that the intersection of the curves represents a straight line L on the ρ-θ plane, , A straight line on the XY plane can be detected by detecting points where many curves intersect. That is, each point of the road image binarized in step 24 is transformed into a curve on the ρ-θ plane by Hough transformation, and a straight line in the image is detected by detecting points where many curves intersect. It is possible to select a lane display line from among them. Here, by using the knowledge stored in the road structure knowledge storage means 15 to detect the lane display line, the lane display line can be detected at high speed without error.

FIG. 7 is an example of a captured driving lane. When the video camera 11 is mounted on a vehicle, the mounting position and angle of the video camera 11 are adjusted so that the driving lane a is captured at a position as shown in FIG. Screen can be adjusted as follows. At this time, if the X axis is oriented rightward in the screen horizontal direction and the Y axis is oriented upward in the screen vertical direction with the lower left corner of the image as the origin, the angle θ of the left and right lane display lines with respect to the X axis becomes Without much change,
Since each of them takes a substantially constant value, these angles can be stored in the road structure knowledge storage means 15 as road structure knowledge and used for detecting a lane display line. Therefore, the area can be limited on the ρ-θ plane centering on these values, and each area can be set as a detection area for the right and left lane display lines.

FIG. 8 is an example in which the detection area for the left and right lane display lines is set on the ρ-θ plane shown in FIG. 6, and the processing is accelerated by processing only the portion b indicated by the solid line in the detection area. The detection of unnecessary straight lines can be prevented by detecting one straight line in each detection area. Here, the setting area for the left and right lane display lines can be made variable using information from an image one screen before or knowledge stored in the road structure knowledge storage unit 15.

In this embodiment, the angle of the lane display line in the image is used as the knowledge. However, the three-dimensional shape of the road or the like can be used as the knowledge.

Next, the routine proceeds to step 26, where a traveling lane detecting process by the traveling lane detecting means 16 is performed. In the traveling lane detection process, the lane in which the vehicle is traveling is detected in the road image from the left and right lane display lines detected in step 25 and their intersections.
FIG. 9 shows an example of a traveling lane a detected when the vehicle is traveling normally, and FIG. 10 shows a traveling lane a detected when the vehicle is traveling at an inclination to the traveling lane.
Here is an example.

Next, the routine proceeds to step 27, where the danger determining means 18 monitors the running state. This monitoring of the running state can be performed by examining how much the position of the intersection of the left and right lane display lines in the traveling lane detected in step 26 deviates from the position of the intersection shown in FIG. 9 as a reference point. The risk can be calculated by monitoring the difference between the positions. If it is determined that the position of the intersection of the left and right lane display lines detected here is not more than a predetermined distance from the position of the reference point and that the traveling state is normal, the process proceeds to step 28, and if the alarm is ON. If it
Turns off and returns to step 21.If it is determined that the position of the intersection of the left and right lane display lines is more than a predetermined distance from the reference point and the running state is out of the steady state, step 29
Then, it is checked whether or not there is a turning instruction detected by the turning instruction detecting means 17. At this time, if the turn instruction is correctly issued for the deviation from the reference position of the traveling lane detected in step 27, the process returns to step 21 with the alarm turned off via step 28, and if not, the process returns to step 21. Proceed to 30 to give an alarm to the driver by the alarm means 19.

Effect of the Invention As described above, in the present invention, it is possible to give an alarm when the driver attempts to incline the traveling direction of the vehicle with respect to the road ahead without expressing the driver's intention to turn. This makes it possible to prevent the vehicle from deviating from the road due to falling asleep, looking aside, etc., and thus has the effect of greatly improving the safety of the vehicle.

[Brief description of the drawings]

FIG. 1 is a block diagram of a vehicle driving lane detecting device according to an embodiment of the present invention, FIG. 2 is a flowchart of control in the embodiment, and FIG. 3 is a diagram showing an example of an image taken from a video camera. FIG. 4 is a diagram showing an image obtained by binarizing the image shown in FIG. 3 after edge detection, and FIG. 5 and FIG.
FIG. 7 is a diagram illustrating an example of a road image captured during normal driving, FIG. 8 is a diagram illustrating an area limitation in the Hough conversion, and FIGS. 9 and 10 are diagrams illustrating detected driving. FIG. 11 shows an example of a lane, and FIG. 11 is a block diagram of a conventional vehicle traveling lane detecting device. 11 video camera 12 edge detection means 13 edge image binarization means 14 lane display line detection means
15 ... road structure knowledge storage means, 16 ... travel lane detection means, 17 ... turn instruction detection means, 18 ... danger determination means, 19
…… alarm means.

Claims (1)

(57) [Claims] 1. A video camera mounted on a vehicle and capturing an image of a road ahead of the vehicle by a predetermined distance, edge detection means for detecting an edge portion of a road image captured by the video camera, and detection by the edge detection means. Edge image binarizing means for binarizing an edge image formed by the edge components using a predetermined threshold value, and a lane detecting a lane display line from the image binarized by the edge image binarizing means. Display line detection means, road structure knowledge storage means for storing the standard angle of the lane display line appearing in the image as knowledge about the road structure used when the lane display line detection means detects the lane display line,
Driving lane detecting means for detecting a driving lane in which the vehicle is traveling by using the lane indicating line detected by the lane indicating line detecting means; and determining whether or not the driver intends to make a turn based on the operation state of the turn signal. Turning instruction detecting means, danger judging means for judging the degree of danger from the state of the driving lane detected by the driving lane detecting means and the presence or absence of a turning instruction detected by the turning instruction detecting means; And a warning means for giving a warning to a driver of the vehicle when it is determined that the vehicle is in a dangerous state.