JP3855102B2 - Lane boundary detection device - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a lane boundary detection device.
[0002]
[Prior art]
Conventionally, as a lane boundary detection device for detecting a lane boundary position (white line or the like) drawn on a road surface, for example, a device described in Japanese Patent Application Laid-Open No. 63-40913 is known. The apparatus described in the publication takes a runway image using a CCD camera, and acquires luminance data as image data in association with a pixel position based on this image. At this time, the luminance data of the pixel position corresponding to the lane boundary position drawn in the black part of the asphalt becomes larger than the luminance data of the pixel position corresponding to the position of the surrounding asphalt. Therefore, this apparatus detects the lane boundary position by comparing the acquired luminance data with a predetermined threshold value.
[0003]
[Problems to be solved by the invention]
However, on the actual road, there are changes in various situations such as the time zone, the presence of shade and the sun, and thus the luminance data changes significantly. Therefore, an apparatus that uniformly detects the lane boundary position by providing a certain threshold value as described above cannot cope with a change in the situation and may erroneously detect the lane boundary position.
[0004]
The present invention has been made to solve the above problems, and an object of the present invention is to provide a lane boundary detection device capable of improving the detection accuracy of the lane boundary position according to a change in the situation. is there.
[0005]
[Means for Solving the Problems]
According to the first aspect of the present invention, the image data of the road is acquired in correspondence with the pixel position by the image acquisition means mounted on the vehicle, and the image data is drawn on the road surface based on the size comparison between the image data and the threshold value. In the lane boundary detection device for detecting a pair of lane boundary positions, the threshold value is set to one of a first threshold value and a second threshold value larger than the first threshold value. And the threshold value setting means is one of the pair of lane boundary positions based on a magnitude comparison between the image data and a threshold value set as the first threshold value. only When the value cannot be detected, the threshold value is set to the second threshold value.
[0006]
According to a second aspect of the present invention, in the first aspect of the invention, the threshold value setting unit is configured to compare the pair of the image data with a threshold value set as a first threshold value. One of the lane boundary positions only The threshold value is set to the second threshold value when the state in which the detection of the error cannot be detected continues for a predetermined time.
[0008]
( Action)
According to the invention described in claim 1, one of the lane boundary positions only Is no longer detected, the threshold value for detecting the lane boundary position is large, that is, the second threshold value is set to be strict. Therefore, under unstable road conditions where the lane boundary position cannot be detected in pairs, increasing the threshold value suppresses erroneous detection of noise or the like as the lane boundary position, resulting in higher reliability. A lane boundary position is detected.
[0009]
According to invention of Claim 2, either one of lane boundary positions only The threshold value is set to the second threshold value when the state in which no detection is possible continues for a predetermined time. Therefore, one of the lane boundary positions only It is avoided that the threshold value is changed unnecessarily due to the temporary loss of detection.
[0011]
DETAILED DESCRIPTION OF THE INVENTION
(First embodiment)
A first embodiment of the present invention will be described below with reference to FIGS.
[0012]
FIG. 2 is a perspective view showing a vehicle 1 to which the present embodiment is applied. As shown in the figure, a mounting frame 3 of a license plate 2 is disposed at a substantially central portion of the rear door 1a of the vehicle 1, and an imaging CCD camera 11 as an image acquisition means is disposed above the mounting frame 3. is set up.
[0013]
The CCD camera 11 acquires an image (video) of the road surface behind the vehicle 1 in the manner shown in FIG. 3, and the boundary line of the lane 4 (the lane defined by the lane marking) as a running road that runs within the imaging range. Are mounted so as to include images of the left white line 5L and the right white line 5R.
[0014]
FIG. 1 is a schematic configuration diagram illustrating a lane departure determination apparatus 10 to which the exemplary embodiment is applied. As shown in the figure, the lane departure determination apparatus 10 includes the CCD camera 11, a main switch 12, an indicator 14, a buzzer 15, and a controller 16 constituting threshold value setting means.
[0015]
The CCD camera 11 outputs the acquired image to the controller 16 as an analog video signal.
The main switch 12 is an operation switch for system operation / stop operated by a user (driver or the like), and outputs a signal corresponding to the operation state to the controller 16.
[0016]
The indicator 14 is provided, for example, on an instrument panel in the passenger compartment, and is lit and driven by the controller 16 so that the user can confirm the operation of the system. For example, when white lines are detected on both sides of the vehicle 1, the controller 16 drives the indicator 14 to light.
[0017]
The buzzer 15 is uttered and driven by the controller 16 when it is determined that there is a possibility of lane departure in the process described later.
The controller 16 includes a control microcomputer 21, a luminance signal extraction circuit 22, an input circuit 23, an indicator signal output circuit 25, a buzzer signal output circuit 26, and a RAM (Random Access Memory) 27. ing.
[0018]
The luminance signal extraction circuit 22 receives the video signal from the CCD camera 11 and extracts the luminance signal, and outputs this luminance signal to the control microcomputer 21.
The input circuit 23 inputs a signal corresponding to the operation state from the main switch 12 and outputs it to the control microcomputer 21.
[0019]
The control microcomputer 21 performs processing such as lane departure determination described later based on the signals input via the luminance signal extraction circuit 22 and the input circuit 23, and temporarily stores various data relating to the processing in the RAM 27. Further, the control microcomputer 21 outputs a drive signal to the buzzer signal output circuit 26 in accordance with the determination result of the lane departure, and drives the buzzer 15 through the buzzer signal.
[0020]
FIG. 4 is a flowchart showing a processing mode of lane departure determination in the present embodiment. This process is repeatedly executed by a scheduled interruption every predetermined time on the assumption that the main switch 12 is turned on.
[0021]
When the processing shifts to this routine, the control microcomputer 21 executes various data input processing in step 101.
Next, the control microcomputer 21 proceeds to a subroutine of step 200 and executes the threshold value changing process of FIG.
[0022]
The control microcomputer 21 that has shifted to the subroutine of Step 200 first determines whether or not the value of the single tracking end flag end_flag stored in the RAM 27 in Step 201 is “0”. The one-tracking end flag end_flag is set to a value “1” or a value “0” in the course of lane recognition processing in step 500 described later, and is temporarily stored in the RAM 27.
[0023]
When it is determined in step 201 that the single tracking end flag end_flag is “0”, the control microcomputer 21 proceeds to step 202 and sets the threshold value (th) to the first threshold value. On the other hand, if it is determined in step 201 that the single tracking end flag end_flag is not “0” (value “1”), the control microcomputer 21 proceeds to step 203 and sets the threshold (th) to the second value. Set to threshold. The first threshold value and the second threshold value are set such that the absolute value of the first threshold value is smaller than the absolute value of the second threshold value.
[0024]
The threshold value (th) is referred to in white line candidate point detection processing in step 300 described later, and the detection sensitivity is adjusted so that the larger the value, the harder it is to be extracted as a white line candidate point.
[0025]
When the threshold value (th) is set in step 202 or step 203, the control microcomputer 21 returns to the routine of FIG.
In step 102, the control microcomputer 21 performs camera video input processing. Specifically, the control microcomputer 21 reads the luminance signal extracted from the video signal of the CCD camera 11 and performs A / D (analog / digital) conversion for each pixel, and the RAM 27 as luminance data associated with the pixel position. Temporarily store. This pixel position is defined according to the imaging range of the CCD camera 11 (see FIG. 3). Note that the luminance data is larger as the corresponding luminance is brighter (white) and smaller as it is darker (black). For example, the luminance data is represented by 8 bits (0 to 255). The luminance is closer to the value “255”, and the luminance is closer to the value “0”.
[0026]
When the camera image input process is completed, the control microcomputer 21 proceeds to a subroutine of step 300 and performs the white line candidate point detection process of FIG.
The control microcomputer 21 that has shifted to the subroutine of Step 300 reads the luminance data of each pixel temporarily stored in the RAM 27 for the n-th line in the image horizontal direction in Step 301.
[0027]
FIG. 6 is a graph showing an example of luminance data corresponding to the position of each pixel arranged in the horizontal direction. In the example shown in the figure, the luminance data of the line shows a peak so as to become brighter at positions corresponding to the left white line 5L and the right white line 5R of the lane 4.
[0028]
The line into which the luminance data is read is set in a range suitable for detecting white line candidate points in the vertical direction of the image, and the control microcomputer 21 stores the luminance data in the range in the order of the line numbers. Read.
[0029]
Next, the control microcomputer 21 proceeds to step 302 and calculates a differential value (dBi) of luminance for each pixel (i-th pixel) of the line. The luminance differential value (dBi) of the i-th pixel is obtained by the difference between the luminance data (Bi) of the i-th pixel and the luminance data (Bi + 1) of the adjacent i + 1-th pixel. FIG. 7 is a graph showing a luminance differential value (dBi) corresponding to the position of each pixel arranged in the horizontal direction based on the luminance data shown in FIG.
[0030]
Next, the control microcomputer 21 proceeds to step 303 and determines whether the luminance differential value (dBi) of the i-th pixel is larger than the threshold value (th) set in the subroutine of step 200. When it is determined that the luminance differential value (dBi) of the i-th pixel is larger than the threshold value (th), the control microcomputer 21 proceeds to step 304 and sets the corresponding pixel position as the white line candidate point in the RAM 27. Is temporarily stored, and the process proceeds to step 305. On the other hand, if it is determined that the luminance differential value (dBi) of the i-th pixel is smaller than the threshold value (th), the control microcomputer 21 proceeds to step 305 as it is.
[0031]
Note that the position of the pixel (i-th pixel) in the line is set in a range that is considered appropriate for detecting the white line candidate point, and the differential value (dBi) is sequentially applied to all the pixels in the set range. ) And extraction as white line candidate points (steps 302 to 304).
[0032]
Here, the extraction of the white line candidate points will be further described with reference to FIG. For example, in FIG. 7, assuming that the threshold value (th) is the first threshold value, the luminance differential value (dBi) is larger than the first threshold value at the four positions 40 to 43. The control microcomputer 21 selects a position (41, 42) that exceeds the threshold (th) for the first time on both sides of the center axis in the front-rear direction of the host vehicle 1 among the detected positions 40 to 43 as white line candidate points. Extract as That is, for the positions 40 and 41 detected on the left side, the right position 41 is extracted, and for the positions 42 and 43 detected on the right side, the left position 42 is extracted as a white line candidate point.
[0033]
Next, in step 305, the control microcomputer 21 adds the number (n + 1) obtained by adding 1 to the number (n) of the line from which the luminance data was read in step 301, and is the maximum horizontal line number that is set to read. Judge whether it is larger.
[0034]
If the number (n + 1) is equal to or smaller than the number of the maximum horizontal line, the control microcomputer 21 determines that there is still a horizontal line from which luminance data is read, and proceeds to step 306. In step 306, the control microcomputer 21 sets a number obtained by adding “1” to the number n as a new number, and returns to step 301.
[0035]
On the other hand, if the number (n + 1) is larger than the number of the maximum horizontal line, the processing of steps 301 to 304 has been completed for all the lines set to read the luminance data, and the control microcomputer 21 performs the routine of FIG. Return to.
[0036]
As described above, the control microcomputer 21 reads the luminance data for each line in the horizontal direction and extracts white line candidate points.
The control microcomputer 21 that has returned to the routine of FIG. 4 moves to the subroutine of step 400 and executes the white line candidate straight line detection process of FIG.
[0037]
The control microcomputer 21 that has shifted to the subroutine of step 400 separates the white line candidate point group temporarily stored in the RAM 27 in step 401 into right and left with the center of the image (the center of each line) as a boundary.
[0038]
Next, in step 410, the control microcomputer 21 performs a left candidate line detection process of detecting candidate lines from the candidate point group separated on the left side among the candidate point groups separated on the left and right. Hereinafter, the left candidate line detection process performed by the control microcomputer 21 in step 410 will be described in detail.
[0039]
When the process reaches the left candidate line detection process in step 410, the control microcomputer 21 presets the number (points) of candidate points constituting the white line candidate point group separated on the left side of the image in step 402. Judge whether it is more than the threshold score. This threshold score is set in advance to a value suitable for determining that a white line (candidate line) exists in the image.
[0040]
If the number of candidate points separated on the left side of the image in step 402 is larger than the threshold score, the control microcomputer 21 resets the left straight line presence counter i_l to “0” in step 403, and proceeds to step 404.
[0041]
In step 404, the control microcomputer 21 determines that there is a white line on the left side of the image, and sets and stores a flag left_exist indicating the state of the white line on the left side of the image at “1”. That is, the flag left_exist indicates that a white line exists on the left side of the image when the value is “1”. The flag left_exist is set and stored to “0” when it is determined that there is no white line on the left side of the image in the manner described later.
[0042]
If it is determined in step 404 that a white line exists on the left side of the image, then in step 405, the control microcomputer 21 reads the white line candidate points temporarily stored in the RAM 27 and applies this point group to a straight line.
[0043]
As a method of fitting to this straight line, for example, Hough transformation is based on literature (“Takashi Matsuyama et al .: Computer Vision, 149/165, New Technology Communications: 1999”, “PVC Hough: Methods and means for recognizing. complex patterns, U.S. Patent No. 3069654 (1962) "). Further, this point group may be applied to a straight line by the method of least squares. Alternatively, various other methods such as feature amount extraction may be employed. In this embodiment, approximation to a straight line is performed using the least square method.
[0044]
When the straight line is detected, the control microcomputer 21 proceeds to step 420.
On the other hand, if the number of candidate points separated on the left side of the image in step 402 is equal to or less than the threshold score, the control microcomputer 21 adds “1” to the left straight line presence counter i_l in step 406 and temporarily stores it in the RAM 27. The process proceeds to step 407.
[0045]
In step 407, the control microcomputer 21 determines whether or not the left straight line presence counter i_l is greater than the number of times corresponding to the distance between the broken lines. The number of times corresponding to the distance between the broken lines is set so that the white line can be detected correctly even when the white line is partially blurred or the white line is drawn with a broken line even though the lane originally exists. Therefore, even if the white line is not temporarily detected due to the running road condition, the control microcomputer 21 continues the determination state assuming that the white line exists for a while.
[0046]
When the left straight line presence counter i_l is equal to or less than the number of times corresponding to the distance between the broken lines in step 407, the control microcomputer 21 determines that the white line is temporarily interrupted, and proceeds to step 404. Similarly, the processing of steps 404 and 405 is performed.
[0047]
On the other hand, if the left straight line presence counter i_l is larger than the number corresponding to the distance between the broken lines in step 407, the control microcomputer 21 determines that the white line on the left side of the image is not continuously detected, and sets and stores the flag left_exist to “0”. . When the flag left_exist is set to “0” and stored, the control microcomputer 21 proceeds to step 420.
[0048]
In step 420, the control microcomputer 21 executes the right candidate lane detection process, and performs setting storage of the value of the flag right_exist indicating the presence state of the white line on the right side of the image, and detection of a straight line if it exists. The right candidate lane detection process in step 420 is performed in the same manner as the left candidate lane detection process in step 410 described above, and thus detailed description thereof is omitted.
[0049]
When candidate straight line detection processing is performed for both sides of the image in this way, the control microcomputer 21 temporarily stores the result of the detection processing in the RAM 27 and returns to the routine of FIG.
[0050]
The control microcomputer 21 that has returned to the routine of FIG. 4 proceeds to the subroutine of step 500 and performs the lane recognition process of FIG.
The control microcomputer 21 whose processing has shifted to step 500 determines whether both the values of the flags left_exist and right_exist indicating the presence state of the left and right white lines (candidate lines) of the image temporarily stored in the RAM 27 in step 501 are “1”. Judge whether.
[0051]
If the values of the flags left_exist and right_exist are both “1”, that is, if it is determined that white lines exist on the left and right sides of the image, the control microcomputer 21 proceeds to step 502.
[0052]
In step 502, the control microcomputer 21 resets the single tracking count count_one_line to “0”, and proceeds to step 503. One tracking count count_one_line is either the left or right side of the image of white line only Represents the number of computations in which the state in which no is detected is continuously detected (the number of computations in the routine of FIG. 10), and either the left side or the right side of the image of white line only This corresponds to the duration of the state where is not detected.
[0053]
In step 503, the control microcomputer 21 temporarily stores the single tracking end flag end_flag described above as “0” in the RAM 27, and proceeds to step 504.
[0054]
In step 504, the control microcomputer 21 turns on the indicator 14 and returns to the routine of FIG.
On the other hand, if it is determined in step 501 that at least one of the flags left_exist and right_exist is not “1”, that is, it is determined that there is no white line on at least one of the left side and the right side of the image, the control microcomputer 21 proceeds to step 505. Transition.
[0055]
Next, in step 505, the control microcomputer 21 determines whether or not the values of the flags left_exist and right_exist are both “0”.
[0056]
If the values of the flags left_exist and right_exist are both “0”, that is, if it is determined that there is no white line on either the left or right side of the image, the control microcomputer 21 proceeds to step 506.
[0057]
In step 506, the control microcomputer 21 sets the single tracking count count_one_line to “0”, sets the single tracking end flag end_flag to “0”, temporarily stores it in the RAM 27, and proceeds to step 507.
[0058]
In step 507, the control microcomputer 21 turns off the indicator 14 and returns to the routine of FIG.
If it is determined in step 505 that the values of the flags left_exist and right_exist are not “0”, that is, one of the left side and the right side of the image of white line only If it is determined that is not detected, the control microcomputer 21 proceeds to step 508.
[0059]
In step 508, the control microcomputer 21 determines whether or not the single tracking number count_one_line is larger than the single tracking number upper limit count_one_max.
[0060]
In step 508, the single tracking number count_one_line is larger than the single tracking number upper limit count_one_max, that is, either the left side or the right side of the image. of white line only If the state in which no is detected is continued longer than the set period, the control microcomputer 21 proceeds to step 509.
[0061]
In step 509, the control microcomputer 21 resets the single tracking number count_one_line to “0” and temporarily stores the single tracking end flag end_flag in the RAM 27 as “1”, and proceeds to step 507.
[0062]
In step 507, the control microcomputer 21 turns off the indicator 14 as described above, and returns to the routine of FIG.
In step 508, the single tracking number count_one_line is less than or equal to the single tracking number upper limit count_one_max, that is, either the left side or the right side of the image. of white line only If it is determined that the state in which no is detected has not reached the set period, the control microcomputer 21 proceeds to step 510.
[0063]
In step 510, the control microcomputer 21 adds “1” to the one tracking count count_one_line, temporarily stores it in the RAM 27, and proceeds to step 503.
[0064]
The control microcomputer 21 that has completed the processing of step 504 or 507 in the above manner returns to the routine of FIG.
In step 107, the control microcomputer 21 determines whether the vehicle 1 has deviated from the lane 4. Specifically, the control microcomputer 21 is a vehicle based on information such as the width of the vehicle 1 stored in advance in its ROM (read only memory) area and the state of the lane 4 (white lines 5L and 5R) stored in the RAM 27. It is determined whether 1 may deviate from the lane 4 in the left direction or the right direction. When the process of step 107 is finished, the control microcomputer 21 once finishes the subsequent process.
[0065]
As described above in detail, according to the first embodiment, the following effects are obtained.
(1) By changing the one tracking end flag end_flag according to the detection status of the white lines 5L and 5R, the threshold value (th) used when detecting the white line candidate point is set as the first threshold value. It was possible to change between the second threshold value. Accordingly, instability such that the white lines 5L and 5R cannot be detected in pairs under the road surface condition, the control microcomputer 21 sets the threshold value to the second threshold value by setting the one tracking end flag end_flag to “1”. Change to As a result, the threshold value is set to be larger than the first threshold value, it is possible to suppress erroneous detection of noise or the like as the lane boundary position, and the lane boundary position can be detected more accurately.
[0066]
(2) White line on either the left or right side of the image only If it is determined that there is not, the time (number of operations) that the state continues is monitored by one tracking number count_one_line, and the threshold (th) is changed when the one tracking number upper limit count_one_max is exceeded . Therefore, one of the white lines only Since the threshold value is changed after an undetected state continues for a certain period of time, it is possible to prevent the threshold value from being changed unnecessarily. And the fall of the detection accuracy accompanying this can also be suppressed.
[0067]
(3) White line on either the left or right side of the image only When it is determined that there is not, the control microcomputer 21 sets the threshold value high to make it difficult to detect the white line candidate point. Therefore, since a white line candidate point is not detected until a white line with higher luminance that can be detected even at a higher threshold appears, a more reliable white line can be detected.
[0068]
(Second Embodiment)
A second embodiment embodying the present invention will be described below with reference to FIGS.
[0069]
In the second embodiment, the aspect of the threshold value changing process in step 200 of the invention described in the first embodiment is changed, and other configurations are the same as those in the first embodiment. It is almost the same as that described in the embodiment. Accordingly, the same components as those in the first embodiment are denoted by the same reference numerals, and detailed description thereof is omitted.
[0070]
FIG. 11 is a schematic configuration diagram illustrating a lane departure determination device 28 to which the second exemplary embodiment is applied. As shown in the figure, the lane departure determination device 28 includes a CCD camera 11, a main switch 12, an indicator 14, a buzzer 15, a headlight lighting switch 30, a determination unit and a threshold setting unit. And a controller 29 constituting the above.
[0071]
The headlight lighting switch 30 is an operation switch for turning on / off a headlight (not shown) operated by a user (driver or the like), and outputs a signal corresponding to the operation state to the controller 29. A signal output from the headlight lighting switch 30 is an on / off signal for lighting the headlight.
[0072]
The controller 29 includes an input circuit 31 for inputting a signal from the headlight lighting switch 30. The input circuit 31 receives a signal from the headlight lighting switch 30 and outputs it to the control microcomputer 21. The control microcomputer 21 determines the brightness of the driving environment based on the signal. That is, the control microcomputer 21 determines that the driving environment is dark (for example, nighttime) when the on signal is input, and determines that the driving environment is bright (for example, daytime) when the off signal is input.
[0073]
FIG. 12 is a flowchart showing a processing mode of lane departure determination in the second embodiment. This process is also repeatedly executed by a scheduled interruption every predetermined time on the assumption that the main switch 12 is turned on.
[0074]
When the processing shifts to this routine, the control microcomputer 21 executes various data input processing in step 101. For example, the control microcomputer 21 inputs a signal corresponding to the operation state of the headlight lighting switch 30 via the input circuit 31.
[0075]
Next, the control microcomputer 21 proceeds to a subroutine of step 600 and executes the threshold value changing process of FIG.
The control microcomputer 21 that has shifted to the subroutine of step 600 first determines in step 601 whether or not the headlight lighting switch 30 is turned on.
[0076]
If it is determined in step 601 that the headlight lighting switch 30 is turned on, the control microcomputer 21 proceeds to step 602 and sets the threshold value (th) to the first threshold value. On the other hand, if it is determined in step 601 that the headlight lighting switch 30 is not turned on, the control microcomputer 21 proceeds to step 603 and sets the threshold value (th) to the second threshold value.
[0077]
The first threshold value is set to a value that can suitably detect the white line state in a state where the headlight lighting switch 30 is turned on (for example, at night). For example, when the headlight is turned on, the taillight of the vehicle 1 is also turned on. The first threshold value is set in consideration of the influence of lighting of the taillight on the image of the camera 11. The second threshold value is set to a value that can preferably detect the white line state in a state where the headlight lighting switch 30 is turned off (for example, daytime).
[0078]
When the threshold value (th) is set in step 602 or step 603, the control microcomputer 21 returns to the routine of FIG.
The control microcomputer 21 performs camera image input processing and white line candidate point detection processing in step 102 and step 300 as in the first embodiment, and proceeds to step 110.
[0079]
The control microcomputer 21 performs white line candidate straight line processing in step 110, and further proceeds to step 111 to perform lane recognition processing. The white line candidate straight line detection process and the lane recognition process performed in step 110 and step 111 are performed by a conventionally known method, and detailed description thereof is omitted here.
[0080]
When the lane recognition process is performed in step 111, the control microcomputer 21 performs the determination process based on the lane departure in step 107 as in the first embodiment, and then terminates the subsequent processes.
[0081]
As described in detail above, according to the second embodiment, the following effects are obtained.
(1) The control microcomputer 21 determines whether the driving environment is bright or dark when the headlight lighting switch 30 is turned on or off, and changes the threshold value to the first or second threshold value. Therefore, the threshold value can be changed to a suitable value according to the light and darkness of the driving environment, and the lane boundary position with high reliability can always be detected regardless of the lightness and darkness of the driving environment.
[0082]
In addition, embodiment of this invention is not limited to the said embodiment, You may change as follows.
In the first embodiment, the duration in which no white line is detected on either the left side or the right side of the image is measured based on the number of computations of a predetermined routine. On the other hand, for example, the duration of a state in which no white line is detected on either the left side or the right side of the image may be directly measured by a timer.
[0083]
In the first embodiment, when either the left or right white line is detected from the state where no white line is detected, the temporary indicator 14 is turned on. However, when detecting a white line from a state in which no white line is detected, control may be performed so that the indicator 14 is lit only when white lines on both sides of the image are detected.
[0084]
In the second embodiment, the first and second threshold values are changed based on the on signal and the off signal of the headlight lighting switch 30. However, the change of the first and second threshold values may be performed based on other signals, for example, at least one of a vehicle small light, a fog lamp, a rear fog lamp, and a taillight. It may be controlled by an on / off signal or its lighting state.
[0085]
In this case, the threshold value may be set according to the lighting state of each light and lamp.
In each of the above embodiments, the white line candidate point is detected by calculating the deviation of the luminance data of each pixel in the horizontal direction from the pixel adjacent to the position as a luminance differential value, and comparing the magnitude of the value with a threshold value. Was done by On the other hand, for example, as shown in FIG. 14, the white line candidate points 50 to 53 may be detected by directly comparing the first and second threshold values similar to the luminance data.
[0086]
In each of the above embodiments, the white line candidate points used when determining the white line are selected as positions 41 and 42 that first exceed the threshold (th) on both sides of the center axis of the vehicle 1 in the front-rear direction. Was done by doing. However, the positions 40 to 43 of all detected white line candidate points may be used for determining the white line.
[0087]
Further, for example, positions 40 and 43 that finally exceed the threshold value (th) on both sides of the central axis may be used.
Further, for example, an intermediate position (or center of gravity position) between a position where the threshold value (th) is exceeded for the first time and a position where the threshold value (th) is finally exceeded on each side of the central axis may be used. .
[0088]
In the above embodiments, two types of threshold values, the first and second threshold values, are used. However, the third and fourth threshold values may be set according to the road surface condition.
In each of the above embodiments, the brightness differential value calculated based on the brightness data obtained by digitizing the brightness signal extracted from the video signal of the CCD camera 11 and the threshold value are compared when detecting the white line candidate point. On the other hand, the luminance signal extracted from the video signal of the CCD camera 11 may be compared with an analog value corresponding to the threshold value as it is. Similarly, the luminance signal may be differentiated while being analog, and the magnitude (absolute value) of the differentiated signal may be compared with an analog value corresponding to the threshold value according to the above.
[0089]
In each of the above embodiments, the luminance signal is extracted from the video signal from the CCD camera 11, and the white line is detected based on the luminance data. In contrast, for example, in a color type camera, hue (hue) data may be extracted from a video signal, and white line detection may be performed based on the extracted data.
[0090]
In the above embodiment, the rear image of the vehicle 1 is acquired by the CCD camera 11, and the white lines 5L and 5R are detected by image recognition based on the acquired image. On the other hand, for example, the CCD camera 11 is installed on the side or front of the vehicle 1. Then, a side image or a front image of the vehicle 1 may be acquired by the CCD camera 11, and the white lines 5L and 5R may be detected by image recognition based on the acquired image. Even if such a change is made, the same effect as in the above embodiment can be obtained.
[0091]
In the above embodiment, the rear image of the vehicle 1 is acquired by the CCD camera 11 mounted on the vehicle 1, and the white lines 5L and 5R are detected by image recognition based on this. On the other hand, the white lines 5L and 5R may be detected by receiving and acquiring video from a camera disposed on a road and performing image recognition based on the video. Even if such a change is made, the same effect as in the above embodiment can be obtained.
[0092]
In the above embodiment, the CCD camera 11 is used for imaging, but an infrared camera, a CMOS camera, or the like may be used.
In the above-described embodiment, the white lines 5L and 5R are detected as the boundary lines of the lane 4;
[0093]
For example, the present invention may be applied to a vehicle system capable of automatic traveling that can be applied to an automatic guided vehicle, a robot, a route bus, an automatic warehouse, and the like.
Next, technical ideas that can be grasped from the above embodiments will be described below together with their effects.
[0094]
(A) A pair of images drawn on the road surface based on a size comparison between the image data and a threshold value based on an image from an imaging unit mounted on a vehicle and capturing an image of a running road in correspondence with a pixel position. In the lane boundary detection device for detecting the lane boundary position of the pair of lane boundary positions, the threshold value setting means is configured to detect the pair of lane boundary positions based on a magnitude comparison between the image data and a threshold value set as a second threshold value. A lane boundary detection device characterized in that when either one of them cannot be detected, the threshold value is set to a first threshold value. Thus, even if it comprises a lane boundary detection apparatus, the effect similar to the invention of Claim 1 is acquired.
[0095]
(B) A lane drawn on the road surface based on a comparison between the image data and a threshold value based on an image from an image pickup means that is mounted on the vehicle and picks up an image of the road and corresponding to the pixel position. In the lane boundary detection device for detecting the boundary position, a determination unit that determines the brightness of the driving environment and a threshold setting unit that sets the threshold value differently according to the determination of the lightness and darkness of the driving environment by the determination unit A lane boundary detection device comprising: Thus, even if it comprises a lane boundary detection apparatus, the effect similar to the invention of Claim 3 is acquired.
[0096]
(C) In the lane boundary detection device according to claim 4,
The lane boundary detection device according to claim 1, wherein the vehicle light is at least one of a headlight, a small light, a fog lamp, a rear fog lamp, a taillight, and a brake light. Thus, even if it comprises a lane boundary detection apparatus, the effect similar to the invention of Claim 4 is acquired.
[0097]
【The invention's effect】
As described above in detail, according to the present invention, it is possible to provide a lane boundary detection device capable of improving the detection accuracy of the lane boundary position according to a change in the situation.
[Brief description of the drawings]
FIG. 1 is a schematic configuration diagram showing a first embodiment of the present invention.
FIG. 2 is a perspective view showing a vehicle to which the first embodiment is applied.
FIG. 3 is a schematic diagram showing an image of a CCD camera.
FIG. 4 is a flowchart showing a processing mode of lane departure determination.
FIG. 5 is a flowchart showing a threshold value changing process.
FIG. 6 is an explanatory diagram of luminance data.
FIG. 7 is an explanatory diagram of a luminance differential value.
FIG. 8 is a flowchart showing an aspect of white line candidate point detection processing;
FIG. 9 is a flowchart showing a mode of white line candidate straight line detection processing;
FIG. 10 is a flowchart showing an aspect of lane recognition processing.
FIG. 11 is a schematic configuration diagram showing a second embodiment.
FIG. 12 is a flowchart illustrating a lane departure determination process according to the second embodiment.
FIG. 13 is a flowchart showing a threshold changing process.
FIG. 14 is an explanatory diagram of another example of white line candidate point detection processing.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 ... Vehicle, 4 ... Lane as lane, 5L, 5R ... Left white line and right white line as lane boundary position, 11 ... CCD camera as image acquisition means, 16 ... Controller as threshold value setting means, 29 ... Determination And controller as threshold setting means.

Claims (2)

A lane for acquiring a pair of lane boundary positions drawn on the road surface by acquiring image data of the road corresponding to the pixel position by an image acquisition means mounted on the vehicle and comparing the size of the image data with a threshold In the boundary detection device,
Threshold setting means for setting the threshold to one of a first threshold and a second threshold greater than the first threshold;
The threshold value setting means detects when only one of the pair of lane boundary positions cannot be detected based on a magnitude comparison between the image data and the threshold value set as the first threshold value. A lane boundary detection device characterized in that a threshold value is set to a second threshold value. The lane boundary detection device according to claim 1,
The threshold value setting means is in a state in which only one of the pair of lane boundary positions cannot be detected based on a magnitude comparison between the image data and the threshold value set as the first threshold value for a predetermined time. A lane boundary detection device, characterized in that the threshold value is set to a second threshold value when continued.

Images