JP2004139338A - Lane marker recognition device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To detect a variety of lane markers with a reduced computational complexity. <P>SOLUTION: Based on that an amount of reflection is different when a laser beam irradiates each of a white line, Botts Dots and cat's eye, either one of the following methods is properly selected which are: a method of performing a lane marker recognition processing according to intensity changes of images picked up by a CCD camera 1 on the basis of an amount of reflected/received light of the laser beam emitted from a laser radar 2; and a lane marker recognition method by pattern-matching between a picked-up image and a predetermined template. <P>COPYRIGHT: (C)2004,JPO

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to an apparatus for recognizing a lane marker which is a dividing line such as a white line on a road.
[0002]
[Prior art]
2. Description of the Related Art There is known a lane recognition device that captures an image of the front of a vehicle using an on-vehicle camera and detects a lane marker indicating a lane segment of a road based on a change in luminance of the obtained image (see Patent Document 1). There is also known a lane recognition device that detects a lane marker by performing pattern matching between an image captured by a vehicle-mounted camera and a template (see Patent Document 2).
[0003]
[Patent Document 1]
JP-A-7-302346
[Patent Document 2]
JP-A-6-119593
[0004]
[Problems to be solved by the invention]
However, the former lane recognition device cannot detect a lane marker that hardly shows a change in luminance in an image, such as Boots Dots used in North America. Also, in the latter lane recognition device, it is possible to detect the Bottom Dots by matching the template with the Bottom Dots shape. However, the white line includes various types such as a solid line, a dashed line, and a double line. It is necessary to store the pattern in the memory, and there is a problem that the capacity of the memory is increased.
[0005]
An object of the present invention is to appropriately use a method of recognizing a lane marker based on a change in brightness of a captured image and a method of recognizing a lane marker by pattern matching between a captured image and a template to appropriately recognize lane markers of various types. A recognition device is provided.
[0006]
[Means for Solving the Problems]
The lane marker recognizing device according to the present invention includes: an image capturing unit that captures an image in front of a vehicle; a first lane marker recognizing unit that performs a lane marker recognizing process based on a luminance change of a captured image; and a pattern matching between the captured image and a predetermined template. Second lane marker recognizing means for performing lane marker recognizing processing, signal transmitting means for transmitting an optical signal forward of the vehicle, signal receiving means for receiving a reflected signal of the optical signal transmitted from the signal transmitting means, The above object is achieved by providing a selecting means for selecting one of the first and second lane marker recognizing means based on a signal reception amount.
[0007]
【The invention's effect】
According to the lane marker recognition device of the present invention, the lane marker recognition processing is performed by appropriately selecting a recognition method based on a luminance change and a recognition method based on pattern matching based on the reception amount of the reflected signal of the optical signal transmitted from the signal transmission unit. Is performed, various types of lane markers can be reliably detected with a small amount of calculation.
[0008]
BEST MODE FOR CARRYING OUT THE INVENTION
FIG. 1 is a diagram showing a configuration of an embodiment of a lane marker recognition device according to the present invention. The lane marker recognition device according to one embodiment includes a CCD camera 1, a laser radar 2, a lane marker recognition processing device 3, an automatic steering control unit 4, a steering actuator 5, and a vehicle speed sensor 6. The CCD camera 1 is attached to the front of the vehicle, captures an image of the front of the vehicle, and outputs a video signal to the lane marker recognition processing device 3. The laser radar 2 transmits a laser beam to the front of the vehicle, and detects an obstacle or the like existing in front of the vehicle by receiving the reflected light. The received light amount information of the reflected light is output to the lane marker recognition processing device 3.
[0009]
The lane marker recognition processing device 3 includes a CPU, a ROM, a RAM (not shown), and the like. The lane marker type is determined based on the video signal transmitted from the CCD camera 1 and the reflected light reception amount information transmitted from the laser radar 2. And an optimum lane marker recognition processing method is selected as described later. The lane marker recognition processing device 3 calculates the lateral position of the own vehicle with respect to the recognized lane marker and outputs the calculated position to the automatic steering control unit 4. The automatic steering control unit 4 calculates a steering wheel steering amount based on the lateral position information of the own vehicle sent from the lane marker recognition processing device 3 and controls the steering actuator 5. The vehicle speed sensor 6 transmits a vehicle speed signal corresponding to the speed of the vehicle to the laser radar 2, the lane marker recognition processing device 3, and the automatic steering control unit 4.
[0010]
FIGS. 2A to 2F are diagrams showing types of lane markers generally used. FIG. 2A shows a lane marker composed of only the white line 7, which is the most common in Japan. Lane markers composed only of white lines are common in North America and Europe and are used worldwide. FIG. 2B shows a lane marker composed of only the bottoms dots 8. Botts Dots 8 is a lane marker mainly used in North America, and has a disc made of ceramics having a diameter of about 100 mm embedded in a road surface.
[0011]
FIG. 2C shows a lane marker including only a reflector called a cat's eye 9. The cat's eye 9 has the property of reflecting incident light in the same direction. As shown in FIG. 2C, a lane marker including only the cat's eye 9 is used in Japan on a curved road other than an expressway, and in North America is used not only on a curved road but also on a straight road. .
[0012]
FIG. 2D shows a lane marker composed of the white line 7 and the cat's eye 9, and FIG. 2E shows a lane marker composed of the Botts Dots 8 and the cat's eye 9. FIG. 2F shows a lane marker composed of a white line 7 and Botts Dots 8.
[0013]
The lane marker recognition device according to the present invention appropriately uses a method of recognizing a lane marker using a change in luminance in an image captured using the CCD camera 1 and a method of recognizing a lane marker by pattern matching. Here, the method using the luminance change and the method using the pattern matching are already used methods, and will be briefly described below. First, a method using a luminance change will be described with reference to FIGS.
[0014]
FIG. 3 is a flowchart of an embodiment showing a processing procedure for recognizing a lane marker using a luminance change. FIGS. 4A to 4E are views for explaining the processing according to the flowchart shown in FIG. Hereinafter, description will be made in order from step S100 of the flowchart shown in FIG.
[0015]
In step S100, each parameter used for the lane marker recognition processing is initialized, and the process proceeds to step S101. In step S101, an image in front of the vehicle is acquired using the CCD camera 1. The acquired image is shown in FIG. As shown in FIG. 4A, the captured image usually displays a guardrail 10 and the like in addition to the white line 7 on the road surface. Generally, the road surface is often paved with asphalt, and in this case, the road surface is black. Therefore, since there is a luminance difference between the white line and the road surface, the white line is recognized using the luminance difference.
[0016]
In the next step S102, a luminance change image is created by extracting a portion having a large luminance change from the image acquired in step S101. FIG. 4B shows a luminance change image created based on the image shown in FIG. After creating the luminance change image, the process proceeds to step S103. In step S103, a window is set in a portion where it is assumed that a white line exists in the brightness change image created in step S102. FIG. 4C shows a diagram in which the window 11 is set for the luminance change image shown in FIG. The window 11 indicates an area where the lane marker recognition processing is performed. By setting the window 11 in this manner, it is not necessary to perform the processing on the entire captured image, and thus the processing time can be reduced. it can.
[0017]
After setting the window 11, the process proceeds to step S104. In step S104, one lane marker candidate point is extracted from each window 11 set in step S103. FIG. 4D shows a diagram in which the lane marker candidate points 12 are extracted. There are various methods for extracting the lane marker candidate points 12. Here, an intersection between the upper side of the window 11 and the line whose luminance changes greatly is extracted as the lane marker candidate points 12.
[0018]
In step S105 subsequent to step S104, lane markers are detected by connecting the lane marker candidate points 12 for each window 11 extracted in step S104. FIG. 4E shows a lane marker 13 recognized by connecting the lane marker candidate points 12 shown in FIG.
[0019]
Next, a method of recognizing lane markers by pattern matching will be described with reference to FIGS. FIG. 5 is a flowchart of an embodiment showing a processing procedure for recognizing a lane marker by pattern matching. FIGS. 6A to 6E are views for explaining the processing according to the flowchart shown in FIG. Hereinafter, description will be made in order from step S200 of the flowchart shown in FIG.
[0020]
In step S200, each parameter used for the lane marker recognition processing is initialized, and the process proceeds to step S201. In step S201, an image in front of the vehicle is acquired using the CCD camera 1. The acquired image is shown in FIG. FIG. 6A shows a plurality of Botts Dots 8. In the next step S202, a window is set as an area where pattern matching processing is performed on the image acquired in step S201. FIG. 6B shows a diagram in which the window 14 is set for the image shown in FIG. After setting the window 14, the process proceeds to step S203.
[0021]
In step S203, the correlation between the area in the window 14 set in step S202 and the previously prepared Boots Dots 8 template is determined. In step S204 subsequent to step S203, an area having a high correlation is extracted as Boots Dots 8 based on the correlation with the Bots Dots 8 template obtained in step S203. FIG. 6C shows the Bottom Dots 8 extracted in the window 14 shown in FIG. 6B. When the Bottom Dots 8 is extracted, the process proceeds to step S205.
[0022]
In step S205, a lane marker candidate point is extracted from the Bottom Dots 8 extracted in step S204. FIG. 6D shows the extracted lane marker candidate points 15. There are various methods for extracting the lane marker candidate points 15, but in this case, a certain point on the center of the screen is extracted as the lane marker candidate points 15 in the extracted Bottom Dots 8. In step S206 following step S205, a lane marker is detected by connecting the plurality of lane marker candidate points 15 extracted in step S205. FIG. 6E shows a lane marker 16 recognized by connecting the lane marker candidate points 15 shown in FIG. 6D.
[0023]
As described above, since the contrast between the Bottom Dots and the asphalt road surface is small, it is difficult to detect the Bottom Dots by a method using a luminance change. When detecting a white line by pattern matching, it is necessary to prepare various templates such as a broken line and a double line. In this case, even if all existing white line pattern templates are prepared, there is a possibility that a new pattern of white lines will appear. Therefore, it can be said that the pattern matching process is not very suitable as a method for detecting a white line.
[0024]
Therefore, in order to accurately recognize both the white line and the Bottom Dots, it is necessary to appropriately select a method using the luminance change and a method using the pattern matching processing. The lane marker recognition device according to the present invention determines the type of the lane marker using the laser radar 2, and appropriately selects a method using a luminance change and a method using a pattern matching process based on the determined type of the lane marker. .
[0025]
Before describing the lane marker recognition processing performed by the lane marker recognition device according to the present invention, when laser light is emitted from the laser radar 2 with respect to the lane markers shown in FIGS. The amount of light received when the light is reflected by will be described with reference to FIGS. 7A to 7F, the horizontal axis represents the distance from the laser radar 2, and the vertical axis represents the amount of reflected laser light received.
[0026]
FIG. 7A shows the result for the lane marker composed of only the white line shown in FIG. The white line reflects light more easily than asphalt, and has a depth unlike the disc-shaped Bottom Dots, so that reflected light can be obtained from the entire area irradiated with the laser light. Accordingly, the reflected light reception amount shows a waveform that peaks at a certain position and then gradually decreases.
[0027]
FIG. 7B shows the result for the lane marker composed of only the dots Dots shown in FIG. 2B. Since there is not much difference between the amount of reflected light received from the Bottom Dots and the amount of reflected light received from the asphalt, no peak appears in the amount of reflected light received as shown in FIG.
[0028]
FIG. 7C shows the result for the lane marker composed only of the cat's eye shown in FIG. 2C. Since the cat's eye has the property of reflecting light in the direction of incidence, very strong reflected light can be obtained from the cat's eye. Also, as shown in FIG. 2 (c), the cat's eye is usually rectangular and has no depth like a white line. Therefore, the amount of reflected light received has a very large peak, and has a steep waveform as shown in FIG.
[0029]
FIG. 7D shows the result for the lane marker composed of the white line and the cat's eye shown in FIG. 2D. In this case, it has reflection characteristics having both white line reflection characteristics and cat's eye reflection characteristics. That is, the reflected and received light amount has a very large peak, and then shows a waveform that gradually decreases.
[0030]
FIG. 7E shows the result for the lane marker composed of the bottoms dots and the cat's eye shown in FIG. 2E. In this case, it has reflection characteristics having both the reflection characteristics of Bottom Dots and the reflection characteristics of cat's eye. As described above, since the amount of reflected light received from the Bottom Dots is small, the characteristic of the amount of reflected light received is similar to the characteristic of the amount of reflected light received by the lane marker composed of only the cat's eye (FIG. 7C).
[0031]
FIG. 7F shows the result for the lane marker composed of the white line and the Bottom Dots shown in FIG. 2F. In this case, there is a reflection characteristic having both a reflection characteristic of the white line and a reflection characteristic of the Botts Dots, but the reflection characteristic of the white line becomes dominant since the amount of reflected light received from the Botts Dots is small. Therefore, the reflected light receiving amount characteristic shows a waveform similar to the reflected light receiving amount characteristic for the lane marker composed only of the white line shown in FIG.
[0032]
In the lane marker recognition device according to the present embodiment, when the lane marker includes a white line, that is, in the case of the lane markers shown in FIGS. To perform lane marker recognition processing. Referring to FIG. 7, in the following cases (1) and (2), it is determined that the lane marker includes a white line.
(1) When the peak of the reflected waveform is between the first threshold value Th1 and the second threshold value Th2
(2) When the peak of the reflected waveform exceeds the second threshold value Th2 and the waveform is asymmetric
[0033]
The case (1) corresponds to the case of the lane markers shown in FIGS. 2 (a) and 2 (f). Since the amount of reflected light received from the white line is smaller than the amount of reflected light received from the cat's eye, FIGS. 2A and 2F are based on the peak value of the amount of reflected light received and the two thresholds Th1 and Th2. ) Are detected. The case (2) corresponds to the case of the lane marker shown in FIG. When the cat's eye is included in the lane marker, the peak value of the reflected light reception amount becomes larger than the second threshold value Th2. However, when the white line is included, the waveform of the reflected light reception amount is not symmetric unlike the case shown in FIGS. 7C and 7E, so that the lane marker shown in FIG. c) and lane markers shown in (e) can be distinguished.
[0034]
The reflected light received by the laser radar 2 includes, in addition to the reflected light from the lane marker, reflected light from vehicles around the own vehicle, a signboard on the roadside, and the like. Therefore, it is necessary to distinguish the reflected light from the lane marker from the reflected light from other than the lane marker. This distinguishing method will be described with reference to FIGS.
[0035]
FIG. 8A is a diagram showing how the laser beam is emitted from the laser radar 2 when viewed from the side of the vehicle 20, and FIG. 8B is a diagram showing how the laser beam is emitted from the laser radar 2 from the top surface of the vehicle 20. FIG. In order to obtain the reflected light from the lane marker, the laser light emitted from the laser radar 2 needs to reach the road surface. As shown in FIG. 8A, it is assumed that the laser radar 2 is mounted at a height of Hs from the ground, and the optical axis L1 of the emitted laser light is horizontal to the road surface. Assuming that the angle between the optical axis L1 of the laser light and the lower end of the laser light spreading in the vertical direction is θv, the distance Dg at which the lower end of the laser light reaches the road surface can be expressed by the following equation (1).
Dg = Hs / tan θv (1)
[0036]
Therefore, the type determination of the lane marker needs to be longer than the distance Dg calculated by the equation (1). For example, when Hs = 0.35 (m) and the angle θv = 1.5 (deg), the distance Dg is 13.4 (m).
[0037]
The laser radar 2 scans in the left-right direction of the vehicle to secure a wide detection area. If the horizontal distance between the laser radar 2 and the lane marker is W, the horizontal angle between the laser optical axis L1 and the vehicle body axis L2 is θh, and the horizontal beam width of the laser light is θw, the distance at which the laser light hits the lane marker is The lower limit Dn and the upper limit Df can be expressed by the following equations (2) and (3), respectively.
Dn = W / tan (θh + θw / 2) (2)
Df = W / tan (θh−θw / 2) (3)
[0038]
For example, when W = 1.8 (m), the angle θh = 5 (deg), and θw = 1 (deg), Dn = 18.7 (m) and Df = 22.9 (m). Here, the lower limit value Dn of the distance at which the laser light hits the lane marker needs to be larger than the distance Dg calculated by Expression (1). Therefore, taking into account the distance Dg, the scan angle θh of the laser beam is determined based on the lateral distance W between the laser radar 2 and the lane marker, and the distances Dn and Df are calculated from the equations (2) and (3). calculate. In this case, when the scan angle of the laser beam is θh, the type of the lane marker is determined based on the waveform of the reflected light reception amount that appears between the distance Dn and the distance Df calculated from Expressions (2) and (3). . That is, the waveform of the amount of reflected light received at a position less than the distance Dn and the waveform of the amount of reflected light received at a position greater than the distance Df can be ignored.
[0039]
FIG. 9 is a flowchart illustrating a procedure of a lane marker recognition process performed by the lane marker recognition device according to the present embodiment. This processing is performed by the lane marker recognition processing device 3. In step S300, parameters used for the lane marker recognition processing are initialized, and the process proceeds to step S301. At the start of the lane marker recognition process, the distance Wn between the laser radar 2 and the lane marker in the horizontal direction is not known, so that the distances Dn and Df required to detect the reflected light from the lane marker cannot be calculated. Therefore, in step S301, a lane marker recognition process using a change in luminance is performed.
[0040]
In step S302 subsequent to step S301, it is determined whether the lane marker has been successfully recognized by the lane marker recognition process performed in step S301. If it is determined that the lane marker has been recognized, the process proceeds to step S306. If it is determined that the lane marker has not been recognized, the process proceeds to step S304. In step S304, recognition processing by pattern matching is performed. In the next step S305, it is determined whether or not the lane marker has been successfully recognized by the recognition processing based on the pattern matching performed in step S304. If it is determined that the lane marker has been recognized, the process proceeds to step S306. If it is determined that the lane marker has not been recognized, the process returns to step S300 to perform the lane marker recognition process again.
[0041]
In step S306, a horizontal distance W between the laser radar 2 and the lane marker is calculated based on the result of the recognition processing performed in step S301 or S304. After calculating the distance W, the process proceeds to step S307. In step S307, based on the distance W calculated in step S306, the scan angle θh and the distances Dn and Df required to evaluate the reflected waveform from the lane marker are determined. After determining the scan angle θh and the distances Dn and Df, the process proceeds to step S308.
[0042]
In step S308, the waveform of the amount of reflected light received at the scan angle θh is obtained, and the flow advances to step S309. In step S309, it is determined whether the peak value of the waveform of the reflected light reception amount at the position of the distance Dn to Df is equal to or less than the first threshold value Th1 among the reflected light reception amounts acquired in step S308. If it is determined that the peak value of the waveform of the reflected light reception amount is equal to or less than the first threshold value Th1, the process proceeds to step S310. If it is determined that the peak value is greater than the first threshold value Th1, the process proceeds to step S311. In step S311, it is determined whether the peak value of the waveform of the reflected light reception amount is equal to or greater than the second threshold Th2. If it is determined that the peak value of the reflected light reception amount waveform is equal to or greater than the second threshold value Th2, the process proceeds to step S312. If it is determined that the peak value is smaller than the second threshold value Th2, the process proceeds to step S313.
[0043]
In step S312, it is determined whether the waveform of the reflected light reception amount acquired in step S308 is symmetric. If it is determined that the waveform of the reflected light reception amount is symmetric, the process proceeds to step S310; otherwise, the process proceeds to step S313. In step S313, in the case of the above condition (1), that is, when the peak value of the reflected waveform is between the first threshold value Th1 and the second threshold value Th2, or in the condition (2) In other words, since the peak value of the reflected waveform exceeds the second threshold value Th2 and the waveform is asymmetric, the lane marker recognition process using the luminance change is performed. After performing the lane marker recognition processing, the process proceeds to step S314.
[0044]
On the other hand, in step S310, lane marker recognition processing by pattern matching is performed. After performing the lane marker recognition processing, the process proceeds to step S314. In step S314, it is determined whether or not a lane marker has been recognized by the lane marker recognition process performed in step S310 or step S313. If it is determined that the lane marker has been recognized, the process returns to step S306, and the processing from step S306 described above is performed. If it is determined that the lane marker has not been recognized, the process returns to step S300.
[0045]
The contents of the lane marker recognition process performed by the lane marker recognition device according to the present embodiment will be summarized. First, a lane marker recognition process using a luminance change or a lane marker recognition process by pattern matching is performed. When the lane marker is recognized, a horizontal distance W between the laser radar 2 and the lane marker is calculated, and the scan angle θh and the distance Dn are calculated. , Df (Steps S300 to S307). Thereafter, a waveform of the amount of reflected light received at the determined scan angle θh is obtained. When the peak value of the waveform of the reflected light reception amount at the position of the distance Dn to Df is equal to or less than the first threshold value Th1 and smaller than the second threshold value Th2, or the peak of the waveform of the reflected light reception amount If the value is equal to or greater than the second threshold value Th2 and the waveform of the reflected light reception amount is asymmetric, the lane marker recognition process using the luminance change is performed; otherwise, the lane marker recognition by pattern matching is performed. The processing is performed (steps S309 to S313).
[0046]
According to the lane marker recognition apparatus of the present embodiment, a method of performing lane marker recognition processing based on a change in luminance of a captured image based on the amount of reflected light of laser light emitted from laser radar 2 and a method of performing lane marker recognition processing by pattern matching Since the lane marker recognition process is performed by appropriately selecting the method and the method, various types of lane markers can be reliably detected. In this case, a method using both the recognition processing method based on the luminance change and the recognition processing method based on the pattern matching is used. For example, when the lane marker recognition processing is performed based on the luminance change of the image and the lane marker cannot be detected, the lane marker based on the pattern matching is used. A method of performing a recognition process is also conceivable. However, in such a method, the amount of processing calculation of the lane marker recognition processing increases, and the detection speed of the lane marker decreases. On the other hand, in the lane marker recognition device according to the present embodiment, one of the two lane marker recognition processing methods is appropriately selected to perform the recognition process, so that the lane marker can be reliably detected with a small amount of calculation. .
[0047]
When the two lane marker recognition processing methods are selected, if the amount of reflected and received laser light is equal to or less than the first threshold Th1, lane marker recognition processing by pattern matching is selected, so that the reflectance is low, such as Bottom Dots. Lane markers can be reliably detected. Further, when the amount of reflected light received by the laser beam is larger than the first threshold value Th1 and smaller than the second threshold value Th2, the lane marker recognition process based on the luminance change is selected. Can be detected.
[0048]
When the amount of reflected light received by the laser beam is equal to or greater than the second threshold value Th2 and the waveform of the amount of reflected light received is a predetermined pattern, that is, asymmetric, a lane marker recognition process based on a luminance change is selected. Therefore, the lane marker including the cat's eye and the white line can be reliably detected. In this case, when the amount of reflected light received by the laser beam is equal to or greater than the second threshold value Th2 and the waveform of the amount of reflected light received is symmetric, the lane marker recognition processing by pattern matching is selected. A lane marker consisting only of a cat's eye having a high cat's eye and a lane marker consisting of a cat's eye and Botts Dots can be reliably detected.
[0049]
In the lane marker recognition device according to the present embodiment, a lane marker is detected using at least one of a lane marker recognition method based on a luminance change and a recognition method based on pattern matching, and a distance W between the detected lane marker and the laser radar 2 is calculated. Then, since the scan angle θh is obtained based on the distance W and the laser light is emitted, the laser light can be reliably emitted toward the lane marker. Further, based on the calculated distance W, distances Dn and Df at which the edge of the laser beam intersects with the lane marker are calculated. Since selection is made, it is possible to prevent erroneous selection of selecting a lane marker recognition method based on the amount of reflected light received from an object other than the lane marker.
[0050]
The present invention is not limited to the above embodiment. For example, in the above-described embodiment, three types of elements constituting a lane marker, that is, a white line, Bottom Dots, and a cat's eye have been described as an example. Can be used. In this case, as shown in FIGS. 7A to 7F, the characteristics of the amount of reflected light received are grasped in advance, and a method based on luminance change or a method based on pattern matching is appropriately selected according to the characteristics. What should I do?
[0051]
The correspondence between the components of the claims and the components of the embodiment is as follows. That is, the CCD camera 1 constitutes the photographing means, the lane marker recognition processing device 3 constitutes the first lane marker recognizing means, the second lane marker recognizing means and the selecting means, and the laser radar 2 constitutes the signal sending means and the signal receiving means. I do. Note that each component is not limited to the above configuration as long as the characteristic functions of the present invention are not impaired.
[Brief description of the drawings]
FIG. 1 is a diagram showing a configuration of an embodiment of a lane marker recognition device according to the present invention.
FIGS. 2A to 2F are diagrams showing various types of lane markers.
FIG. 3 is a flowchart showing a processing procedure of a lane marker recognition process using a luminance change.
FIGS. 4A to 4E are diagrams for explaining a lane marker recognition process using a luminance change.
FIG. 5 is a flowchart showing a processing procedure of lane marker recognition processing by pattern matching.
FIGS. 6A to 6E are diagrams for explaining lane marker recognition processing by pattern matching; FIGS.
FIGS. 7 (a) to 7 (f) are diagrams showing the amount of reflected light received when the lane marker shown in FIGS. 2 (a) to 2 (f) is irradiated with laser light.
FIG. 8A is a diagram of a state in which laser light is emitted from a laser radar viewed from a side of a vehicle, and FIG.
FIG. 9 is a flowchart illustrating a procedure of a lane marker recognition process performed by the lane marker recognition device according to the present embodiment.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 ... CCD camera, 2 ... Laser radar, 3 ... Lane marker recognition processing apparatus, 4 ... Automatic steering control unit, 5 ... Steering actuator, 6 ... Vehicle speed sensor, 7 ... White line, 8 ... Botts Dots, 9 ... Cat's eye, 10 ... Guard rail, 11, 14 window, 12, 15 lane marker candidate point, 13, 16 detection lane marker, 20 vehicle, L1 optical axis of laser beam, L2 axle

Claims (6)

Imaging means mounted on the vehicle to photograph the front of the vehicle;
A first lane marker recognizing unit that performs a lane marker recognizing process based on a luminance change of an image captured by the image capturing unit;
A second lane marker recognizing unit that performs a lane marker recognizing process by pattern matching between the image captured by the image capturing unit and a predetermined template;
Signal transmission means for transmitting an optical signal forward of the vehicle;
Signal receiving means for receiving a reflected signal of the optical signal transmitted from the signal transmitting means,
A selecting section for selecting one of the first lane marker recognizing section and the second lane marker recognizing section based on a reception amount of the reflected signal received by the signal receiving section. Lane marker recognition device. The lane marker recognition device according to claim 1,
The lane marker recognizing device, wherein the selecting means selects the second lane marker recognizing means when the amount of the reflected signal received is equal to or less than a first received amount. The lane marker recognition device according to claim 1 or 2,
The lane marker recognizing device according to claim 1, wherein the selecting means selects the first lane marker recognizing means when the amount of the reflected signal received is greater than a first received amount and less than a second received amount. apparatus. The lane marker recognition device according to any one of claims 1 to 3,
The selecting means selects the first lane marker recognizing means when the reception amount of the reflection signal is equal to or more than a second reception amount and the pattern of the reception amount of the reflection signal is a predetermined pattern. If the received signal amount of the reflected signal is equal to or more than a second received amount and the pattern of the received signal amount of the reflected signal is not the predetermined pattern, it is preferable that the second lane marker recognizing means is selected. Characteristic lane marker recognition device. The lane marker recognition device according to any one of claims 1 to 4,
The apparatus further includes distance detecting means for detecting a lane marker by at least one of the first lane marker recognizing means and the second lane marker recognizing means, and detecting a distance between the detected lane marker and the signal transmitting means based on a detection result. And a signal transmitting unit that transmits the optical signal in a direction determined based on a distance from the lane marker detected by the distance detecting unit. The lane marker recognition device according to claim 5,
The selection unit is configured to select, based on a reception amount of a reflection signal from a distance range determined based on a distance between the lane marker detected by the distance detection unit and the signal transmission unit, of the reflection signals received by the signal reception unit. A lane marker recognizing device for selecting one of the first lane marker recognizing means and the second lane marker recognizing means.

Images