JP2002109549A - White road line recognition device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a white road line recognition device capable of recognizing a white road line accurately by setting thresholds for recognizing the white road line in consideration of a road surface area and a white road line area in plural windows provided on a road image for searching for the white road line. SOLUTION: The plural windows 20 for searching for the white road line 18 are provided on the road image acquired by an image acquisition means 14, and candidate points 21 of the white road line 18 are recognized by the thresholds vn set on each window 20 respectively, and a road model is estimated from the assembly of the white line candidate points 21. The area of the white road line 18 supposed to occupy inside the next window 20 from a preceding road model estimation value in repeated white road line recognition is estimated. The thresholds vn are set so that the area of pixels having brightness equal to or higher than the threshold vn becomes equal to or larger than the area of the white road line 18 estimated by a road area estimation means 103 relative to each window 20.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for acquiring a road ahead in the traveling direction of a vehicle as an image and recognizing a road white line, for example, for automatic driving of a vehicle and preventive safe driving, and the like. The present invention relates to a road white line recognition device for estimating the white line.

[0002]

2. Description of the Related Art As a conventional method for recognizing a road white line from an image of this kind, a vehicle traveling environment recognition apparatus disclosed in Japanese Patent Application Laid-Open No. 7-244717 and a Japanese Patent Application Laid-Open No.
On-board image processing apparatus. The former JP-A-7-244717 attempts to overcome the problem of edge detection by spatial integration, and the latter JP-A-9-35065 overcomes the problem by time integration. Further, in addition to these methods, a method is known in which the average value of the luminance values of the windows provided on the acquired image is used as the threshold value of the window, and the luminance value equal to or higher than the threshold value is regarded as a road white line.

[0003]

However, in such a conventional method of recognizing white lines, when spatial integration is performed as disclosed in Japanese Patent Application Laid-Open No. Hei 7-244717, a change in integration is calculated as a result. Poor removal performance,
In the case of time integration in Japanese Patent Application Laid-Open No. 9-35065, the road moves on the image with time, so that the output image of the time integration becomes blurred and the performance of detecting road white lines deteriorates.

Further, in a method of detecting a road white line by setting a threshold value in a window, an average value serving as a threshold value often does not come between the luminance of an actual road white line and the luminance of a road surface. FIG. 14 shows an example of the cumulative luminance distribution at that time, in which the right axis represents the luminance and the vertical axis represents the distribution frequency of the luminance, and the luminance increases from the right side and changes from 0 to 255. Is 100
Percent has been reached. In the figure, the frequency T0 corresponds to the luminance V0, which means that the frequency equal to or higher than the luminance V0 is T0%. In addition, luminance 0
(Black) Since the above means all luminances, the intercept on the vertical axis is always 100%.

[0005] Here, the average luminance of the luminance distribution is Vav
e, when the road white line luminance is Vref (Vave <Vref),
Since the entire screen is dark and the road white line occupies a small portion, the average brightness is set closer to black.
Therefore, in this case, the luminance existing between Vave and Vref is a target of erroneous detection.

[0006] On the contrary, FIG. 15 shows a similar cumulative luminance distribution in which Vave> Vref. In the case where the luminance of some noises is high and the average luminance is raised, Vdd> Vref is obtained.
The luminance existing between ave and Vref cannot be recognized as a road white line.

The present invention has been made in view of such a conventional problem, and sets a threshold value for recognizing a road white line in a plurality of windows provided on a road image for searching for a road white line. An object of the present invention is to provide a road white line recognizing apparatus which can set road white lines with high accuracy by setting the area in consideration of the area and the road white line area.

[0008]

According to a first aspect of the present invention, there is provided an image acquisition means for acquiring a road image ahead of a vehicle, and a window setting for providing a plurality of windows for searching for a road white line in the acquired road image. And a white line candidate point selecting means for recognizing a road white line candidate point based on a threshold value set for each window, and estimating a road model from a set of white line candidate points. A road area estimating means for estimating an area of a road white line assumed to be occupied in the next window from a previous road model estimation value by repeated road white line recognition, and a threshold value for each window The threshold value is set individually to set the threshold value such that the area of the pixel having the above brightness is equal to or larger than the area of the road white line estimated by the road area estimation means. And summarized in that provided with means.

According to a second aspect of the present invention, there is provided an image acquiring means for acquiring a road image ahead of a vehicle, a window setting means for providing a plurality of windows for searching for a road white line in the acquired road image, A white line candidate point selecting means for recognizing road white line candidate points based on a set threshold value, wherein a road model is estimated from a set of white line candidate points, and the road white line recognition device is repeatedly executed. Road area estimating means for estimating the area of the road white line assumed to occupy the next window from the previous road model estimation value in road white line recognition; and Threshold value setting means for setting the threshold value so as to be equal to or larger than the area of the road white line estimated by the road area estimation means; And gist applying to window.

According to a third aspect of the present invention, in the road white line recognizing device of the second aspect, a predetermined window is selected as a window through which a road white line close to the image acquisition means is seen.

The present invention described in claim 4 is the first to third aspects of the present invention.
In any one of the road white line recognition devices, the gist is to repeat the road recognition and assume the next window position from the previous road model estimation value.

The present invention according to claim 5 provides the invention according to claims 1-4.
In any one of the road white line recognition device, the upper and lower ends of the window are assumed to overlap with the road white line, and the width of the road is assumed to be a trapezoid, with the upper and lower bottoms as respective widths.
The gist is to approximately determine the area.

[0013] The present invention described in claim 6 provides the invention according to claims 1-4.
The road white line is recognized as a parallelogram, with the width of one of the upper and lower edges of the window assumed to overlap the road white line, as one side, and its area is approximately determined. Is the gist.

[0014]

According to the first aspect of the present invention, the road white line is searched in a plurality of windows provided on the road image in front of the vehicle, and is set individually in each window. The road white line can be determined based on the luminance equal to or higher than the threshold value (hereinafter, referred to as high luminance). At this time, a plurality of widths in the road width direction of each window may be set so as to include an area on the image where a road is assumed to exist.
In addition, the high-luminance pixels include road conditions other than road white lines,
For example, it may appear due to reflection of water or direct or reflected light of a lamp or sunlight, which becomes a pseudo white line and is confused with an original road white line. Therefore, the high brightness pixels including the pseudo white line are taken in as road white line candidate points, and then the road white line can be detected from the continuity of the road white line candidate points.

At this time, in each window, the area of a pixel having high luminance, that is, the area of a pixel of a road white line candidate point including a pseudo white line is set to be equal to or larger than the area of a road white line assumed to occupy the window. Since the threshold value is set, a set of pixels having high brightness includes a road white line without fail. Since the threshold value is determined by the comparison between the area of the pixel having high luminance and the area of the road white line, even when the ratio of the road surface area in the window becomes large, the threshold value becomes black. Deviation, that is, setting to the low luminance side is prevented. Also, when high-luminance noise appears in a part of the window, the threshold value is hardly affected if the area is small.
That is, the threshold value is determined in consideration of the road white line area occupied in the window with respect to the entire road surface area in the window, whereby the road white line candidate point in the window can be reliably recognized, As a result, the road white line can be accurately recognized from the road white line candidate points.

According to the second aspect of the present invention, the threshold value is set in the same manner as in the road white line recognition device of the first aspect. Since the threshold value set in a predetermined window is set as a common threshold value and is set in all of a plurality of windows, the setting calculation of this threshold value is performed once for all windows, that is, Only one time is required for one screen. Therefore, since the threshold value can be set at a high speed, the recognition of the road white line can be performed in more real time. Therefore, it is advantageous for a system that requires high-speed processing. For example, it is possible to reliably and safely guide a vehicle traveling at high speed.

According to the third aspect of the present invention, in the road white line recognizing device of the second aspect, the predetermined window is selected as a window through which the road white line close to the image acquisition means is seen. In a window close to the image acquisition unit, the road white line is enlarged compared to a window farther away, so that a change in the road white line area is suppressed. Accordingly, since the noise is less likely to affect the threshold value, the threshold value setting that is repeatedly performed can be made faster, which is advantageous in a system that requires high-speed processing.

According to a fourth aspect of the present invention, there is provided the road white line recognition device according to any one of the first to third aspects,
Since the road recognition is repeatedly performed and the next window position is assumed from the previous road model estimation value, the window position can be appropriately set each time according to the road model. Therefore, when the window is fixedly provided, it is necessary to cover all the areas where the road can exist. However, the window to be set each time needs only to secure a sufficient area for acquiring the road white line, and the area of the window Can be reduced. In other words, by reducing the window in this way, the probability that a pseudo white line other than the road white line is captured is reduced, and the setting accuracy of the threshold value is improved.

According to a fifth aspect of the present invention, there is provided the road white line recognizing device according to any one of the first to fourth aspects, wherein each of the upper and lower ends of the window is assumed to overlap the road white line. Is assumed to be a trapezoid, and the area of the white line is approximately determined. Therefore, since the vertical height of the window is known, the area of the road white line is equal to the upper bottom and the lower bottom. The calculation can be performed at high speed only by obtaining the length of the portion.

According to a sixth aspect of the present invention, there is provided the road white line recognition device according to any one of the first to fourth aspects, wherein one of the upper and lower ends of the window is assumed to overlap with the road white line. The road white line is regarded as a parallelogram as one side, and its area is approximately determined. In this case, only the length of one of the upper and lower ends of the window where the road white line overlaps is determined, and the road white line is determined. The area can be determined, and the operation can be further speeded up.

[0021]

Embodiments of the present invention will be described below in detail with reference to the accompanying drawings.

FIG. 1 is a road white line recognition apparatus 1 according to the present invention.
0, a basic configuration showing one embodiment, an image acquisition unit 14 for acquiring a road image ahead of a vehicle, and a window setting unit 101 for providing a plurality of windows 20 for searching for a road white line 18 in the acquired road image. The road white line 18 is determined by the threshold value vn set for each window 20.
And a white line candidate point selecting means 102 for recognizing the candidate point 21 of the road line. The road model is estimated from the set of the white line candidate points 21. A road area estimating means 103 for estimating an area sn of the road white line 18 assumed to occupy the next window 20;
The area of a pixel having a luminance equal to or greater than the threshold value vn with respect to 0 is the road white line 18 estimated by the road area estimation unit 103.
And a threshold value setting means 104 for individually setting the threshold value vn so as to be equal to or more than the area sn.

FIG. 2 is a schematic view showing an embodiment of the vehicle 12 equipped with the road white line recognition device 10. The image acquisition means 14 and an image signal (hereinafter simply referred to as an image) acquired by the image acquisition means 14 are shown. A processor 16 for processing is installed in the vehicle 12. The image acquisition means 14 is preferably a CCD camera or the like capable of high-sensitivity imaging, and this image acquisition means (hereinafter, referred to as a camera) 14 is attached to the front central portion of the indoor ceiling 12a so as to face forward and downward, and passes through the windshield 12b. A road R image ahead of the vehicle 12 is acquired. The processor 16 is provided at an appropriate location where the vehicle 12 is not affected by heat or wind and rain.
Control for recognizing the road white line 18 from the road R image acquired in 4 and estimating a road model (road shape) is executed.

FIG. 3 shows a flowchart for executing a program P1 for assuming a road model by the processor 16. That is, in this flowchart, after a vehicle front image is captured by the front image capture process (step S
1), a plurality of windows are provided in an image by a window setting process (step S2). Next, steps S4 to S
After it is determined whether or not the processing of Step 7 is completed (Step S3), a road model is assumed by the road parameter estimation processing (Step S8).

In step S4, the next window is selected in order so that the processing in steps S5 to S7 is performed on all of the plurality of windows provided in the image. Then, after capturing pixels having a luminance higher than the threshold value by the white line candidate point selection processing in step S5, the white line can be regarded as a white line from the continuity of the white line candidate points captured in step S5 in the white line candidate line calculation processing in step S6. A possible line is calculated, and the intersection coordinates between the white line candidate line and the upper side of the window are obtained in the road coordinate calculation process in step S7. Step S8
After returning to step S1, the program returns to step S1 to
Is executed every predetermined time.

Hereinafter, the processing of the program P1 will be described in detail for each of steps S1 to S8.

(1) Capture of Forward Image (S1) An image signal captured by the camera 14 is captured every predetermined time for processing the program P1.

(2) Window Setting (S2) As shown in FIG. 4, a plurality of windows 20 are provided along the extension direction of the road white line 18, and each of the windows 20 is determined by a road model obtained by the previous image processing result. Each position is set so that the road white line 18 is taken into the window 20. The vertical direction in the figure indicates the y-axis direction of the image coordinates, and the lower part in the figure is y plus. The horizontal direction in the figure indicates the image coordinate x-axis direction, and the right side in the figure is x-plus. As shown in FIG. 5, each window 20 (one window 20 is shown as an example in FIG. 5) so that the width W of the window 20 in the x direction covers an area on an image where a road is assumed to exist. ) Is set. The coordinates y0 of the upper side 20a and the coordinates y1 of the lower side 20b of each window 20 are fixed. The road model is obtained by Equation 1 (see Japanese Patent Application Laid-Open No. 6-20189).

[0029]

(Equation 1) The parameters A, B, C, D, and H correspond to the lateral position of the own vehicle with respect to the lane, the road curvature, the yaw angle, the pitch angle, and the camera height of the own vehicle with respect to the lane, respectively. E0 is the lane width (the distance between the inner sides of the left and right white lines), and fV and fh are camera perspective transformation constants representing the vertical direction and the horizontal direction, respectively. (X, y) is the coordinates on the image inside the white line with respect to the own vehicle (hereinafter referred to as road coordinates), and in the following, x and y mean the coordinates on the image. Also, i is “0” on the left white line,
The right white line represents “1”.

Equation 1 shows one method of representing a road model. For example, a method in which E0 is a variable, H is fixed, and the lane width is the distance between the center of the left white line and the center of the self-white line. , (X, y) can be defined as the image coordinates of the center of the white line.

Here, the upper side 20a of the N-th window
Let y be the y coordinate of A, and let the previous parameter estimation result be ｛A
(-1), B (-1), C (-1), D (-1), H
If (−1)｝, the center coordinate xest of the current window 20 is expressed by Equation 2 from Equation 1.

[0032]

(Equation 2) There is also proposed a method of setting the width of the window 20 in the x direction in a fixed manner, or rationally setting the dispersion of parameters as disclosed in Japanese Patent Application Laid-Open No. 8-261756.

(3) All windows completed? (S3) The processing in the loop of steps S3 to S7 is
It is determined whether or not execution has been performed for all.

(4) To the next window (S4) The processing in the loop of steps S3 to S7 is performed for each window 20.
If the processing is not executed for all the windows, the unprocessed windows 20 are sequentially selected.

(5) White Line Candidate Point Selection (S5) The program P2 for processing this white line candidate point selection is shown in FIG.
This is executed according to the flowchart shown in FIG. That is, the program P2 estimates the road area (step S10).
After that, a frequency distribution is calculated (step S11), and a threshold value is set from these (step S12).
Selection based on a threshold value is performed (step S13).

Here, the processing of the program P2 will be described in detail for each of steps S10 to S13. (5-1) Road Area Estimation (S10) The road area in the corresponding window 20 is estimated from the result of the previous processing. The upper side 20 of the N-th window 20
Let y be the y coordinate of a. From Equation 1 representing the road model,
The expected x-coordinate xn in this window 20 is shown by Expression 3.

[0037]

(Equation 3) A represents the lateral displacement of the host vehicle with respect to the lane. The displacement on the image when the vehicle is displaced by the white line width represents the width of the white line width on the image. Therefore, assuming that the width is the actual width of the white line (for example, 15 cm), the width wn of the white line on the image is obtained from Expression 3.
Is shown by equation 4. In this case, the subscript n represents the N-th.

[0038]

(Equation 4) The road area is normalized by the area of the window 20. That is, if the window area matches the area, the area = 1
00 percent. The area of the window 20 is known in step S2 of the program P1, and is defined as un. In this case, the road white line 18 is regarded as a parallelogram whose upper and lower bases are in contact with the window frame. That is, window 2
The width of one of the upper side 20a and the lower side 20b of 0 is assumed to overlap the road white line 18, and the road white line 18 is regarded as a parallelogram, and the area thereof is approximately determined. The normalized road white line area sn in this case is expressed by Expression 5. Here, dn represents the width of the window 20 in the y direction.

[0039]

(Equation 5) (5-2) Frequency Distribution Calculation (S11) The luminance distribution is checked by scanning all the pixels in the window 20. By the way, in examining this luminance distribution, FIG.
The prior art in which the luminance changes from 0 to 255 according to the accumulated luminance distribution of 5 has been described.
If ~ 255 are prepared separately, the memory of the processor 16 becomes enormous. For this reason, in the present embodiment, as shown in FIG. 7, the counters are grouped for every 10, and the luminances 0 to 9 are replaced as the counter 0, and the luminances 10 to 19 are sequentially replaced as the counter 10. In this case, the counter is partially omitted in FIG.

(5-3) Threshold value setting (S12) The sum of each counter shown in FIG.
Is the total number of pixels of the road bn, the number of pixels kn on the road white line 18 is given by Expression 6.

[0041]

(Equation 6) Further, the number of counters for each luminance is assumed to be c0 to c25. However, c1 is defined like a counter of pixels of luminance 0-9. Also, only c25 is a counter for 250 to 255 pixels. Next, a variable jn for accumulating the counter value is prepared from the counter with the higher luminance, and the equation 7 is repeatedly executed, starting from 25 for m, decreasing by one. Then, when the condition (jn ≧ kn) is satisfied during the repetition, the calculation is stopped.

[0042]

(Equation 7) If the condition shown in the following Expression 8 is satisfied during the repetition, the calculation is stopped.

[0043]

(Equation 8) At this time, it means that the road white line area is included in the area equal to or larger than the threshold value for the first time. Then, the threshold value vn is calculated by Expression 9 using the value of m at the time when the calculation is stopped.

[0044]

(Equation 9) (5-4) Selection Based on Threshold Value (S13) The pixels in the window 20 are re-scanned, and a luminance equal to or higher than vn is set as a white line candidate point 21. Thereby, as shown in FIG. 8, the white line candidate points 21 are represented as a set group in which a large number of plotted points are plotted in the window 20.

Next, the description returns to the program P1.

(6) White Line Candidate Line Calculation (S6) A white line candidate line 22 is searched from the continuity of the white line candidate points 21 plotted in the window 20. For this search, a Hough transform, a least squares method, or the like can be used. For example,
In the Hough transform linear approximation, as shown in FIG.
Of the straight lines passing through the inside, the one passing through the white line candidate point 21 most is selected.

(7) Calculation of Road Coordinates (S7) As shown in FIG. 10, the intersection coordinates (xn, yn) between the white line candidate line 21 and the upper side 20a of the window 20 are obtained. This becomes the road coordinates in the window 20, and the white line candidate line 2
A set of road coordinates (x0, y0) to (xn0,
yn0) is obtained.

(8) Road Parameter Estimation (S8) Based on the set of road coordinates (x0, y0) to (xn0, yn0) and the structure of the above equation 1, the road model (A, B, C, D,
H) is estimated. As a representative method of performing this estimation, a least square method or a Kalman filter can be used.

Hereinafter, the road white line recognition device 10 of the present embodiment will be described.
The operation of will be described. That is, the road white line recognition device 10
The threshold value set for each window 20 is the camera 14
It is determined appropriately taking into consideration the road surface area and the road white line area in the road screen acquired by the above, and the road white line 18 can be recognized with high accuracy. By using this, the road model (shape) can be accurately determined. Can be estimated.

That is, the road white line recognition device 10 selects the white line candidate point 21 in step S5 of the program P1, and the white line candidate point 21 is used to estimate the road area (step S10) and calculate the frequency distribution (step S10) in the program P2. It is detected based on the threshold value vn set through S11). At this time, the high-brightness captured on the screen is not limited to the road white line 18, and, for example, a backlit window of a vehicle ahead, a lamp, and other direct or reflected light of sunlight have high brightness. However, window 20
If they are restricted to within, the probability that they will enter greatly phenomena. That is, the possibility that the road white line 18 shows high luminance in the window 20 increases, and thus the road white line 18 is surely included in a set of pixels having a luminance equal to or higher than a predetermined value. Here, the setting of the threshold value vn is problematic as described above, and by solving this problem according to the present invention, the road white line 18 can be recognized with high accuracy.

That is, in the present embodiment, assuming that the minimum luminance of the road white line 18 is Vref as shown in FIG. 11, the frequency of a pixel having luminance equal to or higher than Vref is Tref percentage. This means that the white line area occupying the window 20 is Tref percent, for example, when there is no road line having a luminance higher than the white line 18. Therefore, window 2
By limiting to 0, it is unlikely that there is a pixel having luminance higher than the white line, and since the area of the window 20 is known, if the area of the road white line 18 is known, FIG. Conversely, the luminance at which the distribution is equal to or less than Tref percent may be set as the threshold value vn.

In other words, although the conventional averaging method shown in FIG. 15 as a conventional example takes a blackish threshold value and thus causes erroneous recognition, the method of this embodiment has a problem in that the road surface (asphalt) area and Since the threshold value vn is set in consideration of the white line area of the road, the threshold value vn is not set to the black side, no matter how much the area occupied by the road surface.

As a conventional example, in the case of FIG. 16, a loss occurs in order to take a threshold value for whitening, but even in this case, in this embodiment, the road surface area and the road white line area are considered as described above. , The threshold value vn is set, so that even if high-luminance noise appears in a part, the influence is less likely to occur if the area is small. For example, when the luminance of one pixel is extremely high as shown in FIG. 12, the average luminance changes accordingly, but only the extremely high luminance value changes in the frequency distribution. vn does not change.

Further, the road white line recognition device 10 of the present embodiment
In this case, road recognition is repeatedly performed by the processing of the program P1, and the position of the next window 20 can be assumed from the previous road model estimation value. Can be set to For this reason, when the window 20 is fixedly provided, it is necessary to cover all areas where a road may exist, but in the window 20 set each time, it is sufficient to secure an area sufficient to acquire the road white line 18. The area 20 of the window can be reduced. That is, by making the window 20 smaller in this way, the probability that a pseudo white line other than the road white line 18 will be captured is reduced, so that the setting accuracy of the threshold value vn can be improved.

In the present embodiment, equations 4 and 5 are used to estimate the area of the road white line 18, and the road white line 1 is estimated.
8 was approximated as a parallelogram, so the road white line 1
Since the area of the road white line 18 can be determined only by obtaining the length of one of the upper and lower ends of the window 20 where the window 8 overlaps, the calculation can be speeded up.

In the above-described embodiment, the case where the threshold value vn for recognizing the road white line 18 is set for each window 20 is disclosed. The threshold value vn can be set. That is, the common threshold value vn
Is set in the same way as in FIG. 1 shown in the above embodiment, except that the threshold value vn set by the threshold value setting means 104 is set to a plurality of windows 20. Will be executed in one of the windows 20. The threshold value vn set in one window 20 is applied to all other windows 20.

Therefore, by applying the common threshold value vn to all windows 20 in this manner, the calculation for setting this threshold value can be performed only once for one screen. Therefore, since the threshold value can be set at a high speed, the recognition of the road white line 18 can be executed in more real time, which is advantageous for a system that requires high-speed processing. Will be able to do it.

In the above embodiment, the case where the area of the road white line 18 is estimated by approximating the area as a parallelogram is disclosed. However, the estimation of the road area is not limited to the parallelogram. Road 18 can be regarded as a trapezoid. FIG. 13 shows a method of capturing the road white line 18 as a trapezoid. When the y coordinate of the upper side 20a of the N-th window 20 is yun and the same coordinate of the lower side 20b is ydn, the width wun of the white line at the window upper side 20a is shown. Is shown in Equation 10.

[0059]

(Equation 10) Similarly, on the image, the width wd of the white line at the lower side 20b of the window
n is represented by Expression 11.

[0060]

[Equation 11] Then, the road area is normalized by the area of the window 20. That is, if the area matches the area of the window 20, the area is set to 100%. The area of the window 20 is already set in the program P1, and the area is defined as un. Then, when the upper and lower bases of the road white line 18 are approximated to be trapezoids in contact with the window frame, the normalized road white line area Sn is expressed by Expression 12. At this time,
dn represents the width of the window 20 in the y direction.

[0061]

(Equation 12) Therefore, in this case, the road white line 18 is regarded as a shape closer to the actual one than appears on the screen, and a more accurate white line area can be obtained, and the threshold value can be optimized.

FIG. 14 shows still another embodiment, in which a window 20 through which the road white line 18 closest to the camera 14 is seen.
Is a program P3 which sets a threshold value and applies it to all windows 20. That is, this program P3 is the same as step S of program P1 shown in FIG.
The process is executed in place of the white line candidate point selection processed in step 5, and it is first determined in step S20 whether the window 20 is the first window 20 to be searched. That is, the first window 20 is set closest to the camera 14.

If the window 20 is the first window, the road area estimation in step S21, the frequency distribution calculation in step S22, and the threshold setting in step S23 are sequentially executed. Then, the set threshold value is selected in step S24, and high-luminance pixels are plotted on the screen. On the other hand, if it is determined in step S20 that it is the second or later, jump to steps S21 to S23 and proceed to step S24, where a white line candidate point is selected based on the threshold value set in the first window 20.

Accordingly, by using the threshold value set in the first window 20 closest to the camera 14 in all the windows 20, the road white line 18 is enlarged compared to the window 20 located farther away. Because
The change in the road white line area is suppressed. Therefore,
Since the noise hardly affects the threshold value, the threshold value setting that is repeatedly performed can be made faster, which is advantageous in a system that requires high-speed processing.

[Brief description of the drawings]

FIG. 1 is a schematic configuration diagram showing a basic structure of an embodiment of a road white line recognition device according to the present invention.

FIG. 2 is a schematic configuration diagram showing one embodiment of a vehicle equipped with a road white line recognition device according to the present invention.

FIG. 3 is a flowchart for executing an image processing program according to an embodiment of the road white line recognition device according to the present invention.

FIG. 4 is an explanatory diagram showing an arrangement state of a plurality of windows set on a screen in an embodiment of the road white line recognition device according to the present invention.

FIG. 5 is an explanatory diagram showing an arrangement state of windows set in an embodiment of the road white line recognition device according to the present invention.

FIG. 6 is a flowchart for executing a white line candidate point selection by an embodiment of the road white line recognition device according to the present invention.

FIG. 7 is an explanatory diagram showing a luminance distribution of a window set in an embodiment of the road white line recognition device according to the present invention.

FIG. 8 is an explanatory diagram showing a plot state of white line candidate points in a window set in one embodiment of the road white line recognition device according to the present invention.

FIG. 9 is an explanatory diagram showing a selection state of a white line candidate line in a window set in one embodiment of the road white line recognition device according to the present invention.

FIG. 10 is an explanatory diagram showing road coordinates at which a white line candidate line comes into contact with a window set in an embodiment of the road white line recognition device according to the present invention.

FIG. 11 is an explanatory diagram showing a selection state of a threshold value for a luminance distribution of a window set in an embodiment of the road white line recognition device according to the present invention.

FIG. 12 is an explanatory diagram showing the effect of high luminance noise acting on the luminance distribution of a window set in one embodiment of the road white line recognition device according to the present invention.

FIG. 13 is an explanatory diagram showing a state in which a road white line is approximated to a trapezoid in a window set in another embodiment of the road white line recognition device according to the present invention.

FIG. 14 is a flowchart for executing white line candidate point selection by another embodiment of the road white line recognition device according to the present invention.

FIG. 15 is an explanatory diagram showing a positional relationship of average luminance with respect to a luminance distribution obtained by conventional road white line recognition.

FIG. 16 is an explanatory diagram showing another positional relationship between the average luminance and the luminance distribution obtained by the conventional road white line recognition.

DESCRIPTION OF SYMBOLS 10 Road white line recognition device 12 Vehicle 14 Camera (image acquisition means) 16 Processor 18 Road white line 20 Window 21 White line candidate point 22 White line candidate line 101 Window setting means 102 White line candidate point selection means 103 Road area estimation means 104 Threshold setting means

 ──────────────────────────────────────────────────続 き Continued on the front page F term (reference) 5B057 AA06 AA16 BA02 CA01 CA08 CA12 CA16 CB01 CB08 CB12 CB16 CC03 DA08 DB02 DB06 DB09 DC04 DC23 5H180 AA01 CC04 CC24 LL01 LL08 LL09 5L096 AA02 AA06 BA04 BA18 FA02 FA17 GA41 GA51

Claims (6)

[Claims] 1. An image acquisition means for acquiring a road image ahead of a vehicle, a window setting means for providing a plurality of windows for searching for a road white line in the acquired road image, and a threshold value set for each window. A road white line recognition device comprising a white line candidate point selecting means for recognizing a road white line candidate point, and estimating a road model from a set of white line candidate points. A road area estimating means for estimating an area of a road white line assumed to be occupied in the next window from the model estimation value; Threshold value setting means for individually setting the threshold values so as to be equal to or larger than the area of the road white line estimated in the above. Recognition device. 2. An image acquiring means for acquiring a road image ahead of a vehicle, a window setting means for providing a plurality of windows for searching for a road white line in the acquired road image, and a road set by a threshold value set in each window. A white line candidate point selecting means for recognizing a white line candidate point, wherein a road model is estimated from a set of white line candidate points. The road area estimating means for estimating the area of the road white line assumed to be occupied in the next window from the estimated value, and the area of a pixel having a luminance equal to or higher than a threshold value in a predetermined window is estimated by the road area estimating means. Threshold setting means for setting the threshold so as to be equal to or larger than the area of the road white line, and applying the threshold to all windows. Road white line recognition device. 3. The road white line recognition device according to claim 2, wherein the predetermined window is selected as a window through which a road white line close to the image acquisition means is seen. 4. The method according to claim 1, wherein
A road white line recognition device characterized by repeatedly performing road recognition and assuming a next window position from a previous road model estimation value. 5. The method according to claim 1, wherein:
An apparatus for recognizing a road white line, wherein the width of the upper and lower ends of the window is assumed to overlap the road white line, and the width of the road white line is regarded as a trapezoid, and the area thereof is approximately determined. 6. In any one of claims 1 to 4,
A road white line recognition device, characterized in that a width assumed to be one of the upper and lower ends of a window and a road white line is regarded as one side, and the road white line is regarded as a parallelogram, and its area is approximately determined.