JPH08297728A - Vehicle front monitoring device and method - Google Patents

Abstract

PURPOSE: To provide a vehicle front monitoring device which can correctly recognize a white line without being effected by any influence of noises. CONSTITUTION: A vehicle front monitoring device is provided with a window means 20 which sets a prescribed range of a screen as a window to successively move this window on the screen and also to successively reads the binary signals corresponding to the image pickup elements included in the window, plural reference windows 30 which have the ranges equal to the window of the means 20 and also have the binary signals in continuous patterns previously decided in the window means 20, a decision means 31 which performs comparison to decide the coincidence between the window 20 and the windows 30, and a white line recognition means which recognizes the white line that is drawn on a road and can be viewed in front of a vehicle based on the output of the means 31.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a vehicle front monitoring apparatus and method using a camera mounted toward the front of a vehicle, and more particularly to recognizing a white line drawn on a road in front of the vehicle. It is related to white line recognition.

[0002]

2. Description of the Related Art FIG. 10 is a block diagram showing the configuration of a conventional vehicle front monitoring system. In the figure, 1 is a camera that is mounted toward the front of the vehicle to take a picture of the front, and 2 is an analog video signal output from a plurality of image pickup devices provided on the screen of the camera, and this signal is converted into a digital signal. An A / D converter 3 for conversion receives the output signal of the A / D converter, classifies this signal into two types of values, and detects the contour line of the image photographed on the screen.
It is white line recognition means for recognizing a white line drawn on the road ahead based on the output of the value conversion means.

The operation of the conventional device thus configured will be described with reference to FIG. FIG. 11 shows an image taken by the camera 1. In the figure, 5 is a screen on which a plurality of image pickup devices 6 are arranged, and 7 is a white line drawn on the road.
The image captured by the camera 1 is received by the image sensor 6 and output as an analog signal. This signal is converted into a digital signal by the A / D converter 2 and given to the binarizing means 3. The binarization unit 3 has, for example, a predetermined reference luminance difference, and each image pickup is performed depending on whether or not the luminance difference between each image pickup element and the image pickup element adjacent thereto is larger than the reference luminance difference. The video signal corresponding to the element 6 is classified into two types of values. If the luminance difference between the adjacent image pickup elements 6 is equal to or larger than the reference luminance difference, 1 is output and the luminance difference between the adjacent image pickup elements 6 is set as the reference. If it is smaller than the brightness difference, 0 is output. The state of this processing will be described with reference to FIG. In the figure, (a) shows the brightness information of each image sensor 6 by numerical values. (B) is a calculation of the brightness difference between the adjacent image pickup devices 6 in (a). Similarly, (c) is a calculation of the brightness difference between the vertically adjacent image pickup devices 6 in (a). The information of (b) and (c) is compared with a predetermined reference luminance difference of 10. If the luminance difference is 10 or more, 1 (hatched portion in the drawing) is given, and the luminance difference is less than 10. Is given 0 (white part in the figure) to obtain (d) and (e), respectively. These pieces of information (d) and (e) are combined by logical sum (OR) to obtain the information of the contour line (hatched portion: 1, white portion: 0) represented by the binary value shown in (f). That is, after the process of the binarizing means 3, a signal of 1 or 0 is given corresponding to each image pickup element 6, and the screen 5 is represented by a binary signal of 1 and 0. .

The white line recognition means 4 recognizes the white line photographed on the screen based on the information of the screen 5 obtained as described above. The basic idea of white line recognition is that the white line is drawn on a dark road with white paint, so there is a larger difference in brightness than normal at the boundary, and the difference in brightness larger than normal is It is only necessary to recognize a certain point as a white line. An example of the condition for recognizing the white line is as follows. (1) There is a predetermined difference in brightness between image pickup devices that are horizontally adjacent to the screen, (2) A binary signal is detected in the horizontal direction from the center of the screen toward the edge of the screen, and the binary signal is first detected. Is 0 to 1, or 1
(3) The points satisfying the above (1) and (2) are detected in the right half and the left half of the screen, respectively, and the distance between them is approximately 4 m. The white line information thus obtained is output from the white line recognizing means 4 and used for monitoring control connected to the subsequent stage.

[0005]

By the way, in the conventional apparatus, when the binary signal of the image pickup device is detected, the method of sequentially detecting the entire area from the center of the screen to the edge of the screen in the horizontal direction is not adopted. A predetermined detection range is set in the vicinity of the detection point, and only the binary signal of the image sensor within the detection range is detected. This is to prevent erroneous recognition of a white line as a line that has nothing to do with the white line if the detection range is too wide. Since the white line has continuity, it is unlikely that the white line is far from the detection point of the previous white line, and the white line is surely detected by setting the detection range to the predetermined range as described above. It becomes possible. In FIG. 11, 81 is a detection range having a predetermined range. Reference numeral 82 denotes a detection range newly set based on the position of the white line detected in the detection range 81, and thereafter, the detection range is sequentially set based on the previous detection position of the white line. However, even the conventional apparatus configured as described above is vulnerable to noise and cannot detect the white line in some cases. In FIG. 11, reference numeral 9 denotes noise, which is caused by, for example, a shadow cast on the road, a dent on the road, a repair mark on the road, or raindrops on the windshield during rainy weather.

The detection range 84 is set based on the previous detection range 83, and the noise 9 and the white line 7 exist in the detection range 84. By the way, since the white line which is closest to the center of the screen is recognized as a white line, in this case, the white line recognition means 4 recognizes that the noise 9 is a white line. Further, the white line recognition means 4 sets the detection range 85 as the next detection range, and the noise 9 is also detected in the detection range 85.
Is recognized as a white line, and the detection range 86 is set as a new detection range. In the detection range 86, the white line 7 is outside the detection range as shown, and the white line can no longer be detected.

The present invention has been made to solve such a problem, and an object thereof is to obtain a vehicle front monitoring device capable of correctly recognizing a white line without being affected by noise.

Another object of the present invention is to provide a vehicle front monitoring method capable of correctly recognizing a white line without being affected by noise.

[0009]

A vehicle front monitoring device according to the present invention sets a predetermined range of a screen as a window and sequentially moves the window on the screen.
Window means for sequentially reading binary signals corresponding to the image sensor existing in this window, and a range equal to the window means, and the binary signal is set in this window in a pattern having a predetermined continuity. A plurality of reference windows, a determination means for comparing the windows and the reference window to determine whether or not they match each other, and a white line drawn on the road in front of the vehicle based on the output of the determination means. And a white line recognition means for recognizing.

Further, in the vehicle front monitoring apparatus according to the present invention, the window means arranges storage elements corresponding to the image pickup elements existing in the window, and the storage elements sequentially output signals from the binarization means. The memory elements corresponding to the image sensors that are adjacent to each other in the horizontal direction of the screen are connected in series to form a plurality of memory element arrays, and the memory element arrays that are adjacent to each other in the vertical direction of the screen. Are configured to be connected in parallel via the delay means.

The vehicle front monitoring apparatus according to the present invention has a plurality of sets of window means and determination means.

Further, in the vehicle front monitoring method according to the present invention, a predetermined range of the screen is set as a window and the window is sequentially moved on the screen, and at the same time, a binary value corresponding to an image pickup device existing in the window is set. Sequential reading of the signal and comparing the window with a plurality of reference windows having a range equal to the window and having binary signals set in a pattern having a predetermined continuity within the window. And a step of recognizing a white line drawn on the road in front of the vehicle based on the result obtained by the step of determining whether the two match. is there.

Further, in the vehicle front monitoring method according to the present invention, a predetermined range of the screen is set as a window, the window is sequentially moved on the screen, and a binary value corresponding to the image pickup device existing in the window is set. Sequential reading of the signal and comparing the window with a plurality of reference windows having a range equal to the window and having binary signals set in a pattern having a predetermined continuity within the window. And a step of determining whether or not they match each other are repeated a plurality of times.

[0014]

The vehicle front monitoring device according to the present invention sets a predetermined range on the screen as a window, compares the binary signal of this window with the binary signal of the reference window, and compares the binary signal of the window with the reference window. It is determined whether or not they match the pattern of No. 2, and if it is determined that they do not match, it is determined that the binary signal of the window does not have the pattern of the reference window, and this is ignored. Works like.

In the vehicle front monitoring device according to the present invention, the storage elements corresponding to the image pickup elements existing in the window are arranged, and the storage elements corresponding to the image pickup elements adjacent in the horizontal direction of the screen are respectively connected in series. To form a plurality of storage element arrays, and connect the storage element arrays adjacent in the vertical direction of the screen in parallel via the delay means to form the window means.

Further, the vehicle front monitoring device according to the present invention compares the binary signal in the window with the binary signal in the reference window and determines whether or not they match each other a plurality of times.

Further, in the vehicle front monitoring method according to the present invention, a predetermined range of the screen is set as a window, the binary signal of this window is compared with the binary signal of the reference window, and the binary signal of the window is detected. It is determined whether or not they match the pattern of the reference window, and if it is determined that they do not match, it is determined that the binary signal of the window does not have the pattern of the reference window. Take the step of ignoring.

Further, in the vehicle front monitoring method according to the present invention, the step of comparing the binary signal in the window and the binary signal in the reference window and determining whether or not they match each other is performed a plurality of times.

[0019]

【Example】

Example 1. In Example 1, paying attention to the fact that white lines have continuity, pattern matching is performed before white line recognition to remove noise 9 unrelated to white line detection from the screen 5, and the screen from which this noise 9 has been removed The white line is detected based on FIG. 1 is a block diagram showing the configuration of the first embodiment. In the figure, the same or corresponding parts as those described above are designated by the same reference numerals. In the figure, 1 is a camera which is mounted toward the front of the vehicle to take a picture of the front, and 2 is an A which converts an analog video signal from a plurality of image pickup devices 6 arranged on a screen 5 of the camera 1 into a digital signal. / D converter,
Reference numeral 3 is a binarizing means for comparing the output signal from the A / D converter 2 with a predetermined reference luminance into two kinds of values of 1 or 0 and outputting a binary signal. 10 is a binarizing means 3. The pattern matching means 11 compares the window set in the predetermined range of the screen 5 with the reference window on the basis of the output to determine whether the patterns of the binary signals match or not. The white line recognition means performs white line recognition on the basis of the image, and the output of the white line recognition means 11 is used for the monitoring process connected to the subsequent stage.

Next, pattern matching will be described with reference to the drawings. FIG. 2 is an explanatory diagram for explaining the state of pattern matching. In the figure, the image sensor 6 having diagonal lines
Has a luminance equal to or higher than the reference luminance, and one of the binary signals is given. Reference numeral 7 denotes a white line, which is represented here in correspondence with the size of the image sensor 6. Reference numeral 12 is a window formed of a predetermined range of the screen 5, that is, a block of the image sensor 6 in 4 rows and 4 columns. This window 1
2 moves sequentially from the upper left of the screen to the lower right of the screen in synchronization with the scanning of the image sensor. Another window shown below the window 12 in the figure is a case where the window 12 is sequentially moved according to the scanning order of the image sensor and moved to a position including the noise 9. FIG. 3 shows an example of a reference window, and shows a reference window of 4 rows and 4 columns in which a binary signal is preset in a pattern having continuity.

The operation will be described below. Image sensor 6
Scan 1 row 1 column, 1 row 2 column, ... 1 row n column, 2 row 1
Columns, 2 rows, 2 columns, ... M rows and n columns are sequentially performed from the upper left of the screen toward the lower right of the screen. First, when the image sensor 6 of 4 rows and 4 columns is being scanned, the window 1 of 4 rows and 4 columns up to 1 row and 1 column is started from the image sensor 6 of 4 rows and 4 columns.
2 is set. When the window 12 is set, next, it is determined whether or not the binary signal pattern in the set window 12 and the binary signal pattern of the reference window shown in FIG. 3 match. The window setting method and the window comparison / determination method will be described in detail later.

In FIG. 2, all the binary signals in the windows set from the 4th row and the 4th column to the 1st row and the 1st column are 0 (the state where there is no hatched portion). In contrast, none of the reference windows prepared in FIG. 3 match such a pattern. Therefore, at this time, the pattern matching means 10 determines that they do not match and outputs 0. Next, it is assumed that the scanning of the image sensor 6 is moved from 4th row and 4th column to 4th row and 5th column. At this time, the window 12 is set to have a range of 4 × 4 from the 4th row and the 5th column to the 1st row and the 2nd column. In this case as well, the window 12 is compared with the reference windows 1 to 8 in the same manner, and when it is determined that they match each other, the pattern matching means 10 outputs 1 and when it is determined that they do not match. Outputs 0. The window 12 is sequentially moved for each scanning of the image sensor as described above, a new range is set each time, and the window 12 is compared with the reference window to be judged. Therefore, when the image sensor 6 of 6 rows and 8 columns is being scanned,
Since the pattern of the binary signal in the window 12 and the binary signal in the reference window 3 match, the pattern matching means 10 outputs 1 at this time. On the other hand, 1
When the image sensor 6 of row 1 and column 9 is being scanned, there is no reference window that matches the binary signal pattern of the window 12. Therefore, at this time, the pattern matching means 10 outputs 0. That is, the patterns of the reference windows 1 to 8 are set to have continuity, whereas the noise 9 generally does not have continuity. Therefore, it does not match the pattern of any of the reference windows shown in FIG. 3, in which case the noise 9 is ignored.

As described above, the pattern matching means 1
According to 0, among the images obtained on the screen 5, only images having continuity are selected, and images having no continuity are ignored. Therefore, the image signal is held in the case of the white line 7 having continuity, but the image signal is ignored in the case of the noise 9 having no continuity. Therefore, the image signal of the noise 9 is removed from the image signal supplied to the white line recognition means 11, and the white line recognition means 11 performs the above-described white line recognition processing based on the image signal from which the noise 9 is removed. As described above, the inconvenience that the white line cannot be recognized due to the influence of the noise 9 does not occur.

Next, how to set the window 12 will be described. FIG. 4 is a block diagram showing the window means,
This window means is included in the pattern matching means 10. FIG. 5 is an explanatory diagram for explaining the correspondence between the window means and the image sensor. In FIG. 4, 20 is a window means, 21 is a storage element composed of a D-type flip-flop, and 22 is a delay means for delaying the output signal of the binarizing means 3 for a predetermined period. The storage element 21 is adjusted to the range of the window to be set, and in this case, it is adjusted to the 4th row and 4th column of the window 12. The storage elements 21 adjacent to each other in the scanning direction of the image pickup element 6 (in this case, the horizontal direction of the screen) are connected in series, and the storage elements a to d, e shown in the figure.
.About.h, i to 1, m to p constitute four memory element arrays. These storage element arrays are connected in parallel,
The delay means 22 is interposed between the storage element arrays. The delay period by the delay unit 22 is set to a time period from scanning the image sensor 6 from the left end of the screen to the right end of the screen and returning to the left end of the screen again, that is, a period corresponding to one horizontal scanning period.

Window means 20 constructed in this way
The operation will be described. As shown in FIG.
Are sequentially scanned one by one from the upper left of the screen to the lower right of the screen, and the video signal obtained by the scanning is converted into a binary signal of 1 or 0 by the binarizing unit 3 and output. It The output of the binarizing means 3 is output every time the image pickup device 6 scans and is given to the window means 20. The window means 20 sequentially transmits the output of the binarizing means 3 to the storage element 21 and stores and updates the binary signal stored in the storage element 21. That is, each of the four storage element arrays has
Binary signals of four continuous image pickup devices are stored. Further, a delay means 22 for transmitting a signal for only one horizontal scanning period is interposed between the storage element columns. Therefore, 4
Each of the two storage element arrays is delayed by one horizontal scanning period, which is continuous in the vertical direction of the screen when viewed on the screen. Accordingly, the a to p storage elements 21 of the window means 20 are stored in the a to p imaging elements 6 shown in FIG.
The value signals are correspondingly stored and updated.

Next, a method of comparing and determining the set window and the reference window will be described. FIG. 6 is a detailed block diagram of the pattern matching means 10. In the figure, reference numeral 30 is a reference window in which a binary signal of a predetermined pattern is stored, and has reference windows 1 to N (N is a natural number), where the reference windows are 1 to 8. Each of them is given the reference windows 1 to 8 shown in FIG. Reference numeral 31 is a judging means for comparing the binary signal of the window means 20 and the binary signal of the reference window 30 to judge whether or not they match each other, and 32 is an N input terminal and an output terminal. Is an OR circuit having

As described above, in the window means 20, the binary signal is stored and updated every time the image pickup device 6 is scanned.
The determination unit 31 compares the output of the window unit 20 that sequentially changes for each scan of the image sensor 6 with the reference window 30. This comparison operation is performed as shown in FIG. 7, for example. FIG. 7 is an explanatory diagram showing the operation of the determination means 31. The output of the window means 20 is compared with the reference windows 1 to 8 by each judging means 31, respectively. Judgment means 3
1 determines whether or not the portion where the binary signal is 1 in the reference window (hatched portion in the drawing) is also 1 in the output of the window means 20. That is, in other words, the determination unit 31 determines that the binary signal pattern set in the reference window 30 is the window 12 set this time.
It is determined whether it is inside. Therefore, in the case of FIG. 7, the reference window 5 is output to the window means 20.
Pattern exists, it is determined that the two match, and the determination unit 31 outputs 1. Note that FIG.
The output of the window means 20 has a part in which the binary signal is 1 at the lower right, but this part is ignored because it is out of the determination target. Therefore, also in this judging means 31, small noises are removed.

The logical sum circuit 32 receives the output of the judging means 31, outputs 1 if any one of the plurality of judging means 31 outputs 1, and also outputs the window means 2
When the output of 0 does not match any of the reference windows and all the determination means 31 output 0, 0 is output.

The binary signal containing no noise obtained as described above is given to the white line recognizing means 11, and the white line recognizing means 11 accurately detects the white line based on the binary signal containing no noise.

Therefore, according to the first embodiment, noise can be removed and accurate white line recognition can be performed.

Further, a window can be set on the screen with a simple structure.

Although the size of the window 12 is 4 rows and 4 columns in the above embodiment, for example, 3 rows and 3 columns, 5 rows and 5 columns,
It may have any range of 4 rows by 5 columns, 3 rows by 5 columns, and the like. In this case, as the pattern of the reference window having continuity, a pattern corresponding to each of the above ranges is created according to the example shown in FIG.

Although the reference windows 1 to 8 are prepared as the reference window 30 in FIG. 6, the reference windows 1 to 8 in FIG. 3 are rotated rightward by 90 °.
The reference window 30 may be prepared in 32 ways by preparing the one rotated by 180 ° and the one rotated by 270 °. At this time, 32 determination means 31 are also prepared, and 32 input terminals of the OR circuit 32 are also prepared. By doing so, not only the white line on the left side of the road described above but also the white line on the right side of the road can be recognized. Further, even if the road is curved and the white line is curved, it can be recognized without any problem.

Although the configuration of the device and its operation have been described above, a vehicle forward monitoring method can be obtained by performing the operation in the order of operation of the device.

Example 2. In the first embodiment, the pattern matching is performed to remove the noise from the video signal. However, when the noise is large and has a certain degree of continuity, the noise is completely removed by the first embodiment. Sometimes it cannot be removed. The second embodiment solves this problem, and specifically, the pattern matching is performed a plurality of times.

FIG. 8 is a block diagram showing the configuration of the second embodiment. Compared with FIG. 1, one stage of pattern matching means 10 is added. That is, in the case of FIG. 8, the output of the OR circuit 32 of the pattern matching means 10 in the previous stage is input to the window means 20 of the pattern matching means 10 in the subsequent stage, and the pattern matching is performed again.

FIG. 9 shows the processing state of the second embodiment. In the figure, (a) is an image obtained from the output of the binarizing means 3,
(B) is an image obtained from the output of the pattern matching means 10 in the former stage, and (c) is an image obtained from the output of the pattern matching means 10 in the latter stage. From this figure, it can be seen that even relatively large noise can be removed by performing pattern matching a plurality of times.

Although the configuration of the device and the operation thereof have been described above, it is also possible to obtain the vehicle front monitoring method by performing the operation in the order of operation of the device.

[0039]

As described above, according to the vehicular front monitoring apparatus of the present invention, a predetermined range of the screen is set as a window, the window is sequentially moved on the screen, and an image existing in the window is picked up. 2 corresponding to the element
Window means for sequentially reading the value signal, a plurality of reference windows having a range equal to the window, and the binary signal being set in the window in a pattern having a predetermined continuity, the window and the reference Since the determination means for comparing the windows with each other to determine whether the two coincide with each other and the white line recognition means for recognizing the white line drawn on the road ahead of the vehicle based on the output of the determination means are provided, continuous Accurate white line recognition can be performed by removing noise having no property.

Further, according to the vehicle front monitoring device of the present invention, the window means arranges the memory elements corresponding to the image pickup elements existing in the window, and the memory elements cause the signal from the binarization means to be arranged. The storage elements corresponding to the image pickup elements that are adjacent in the horizontal direction of the screen are connected in series to form a plurality of storage element rows, and the storage elements that are adjacent in the vertical direction of the screen are stored. Since the element arrays are configured to be connected in parallel via the delay means, the window means can be configured with a simple configuration.

Further, according to the vehicle front monitoring device of the present invention, since the plurality of sets of window means and determination means are provided, it is possible to perform pattern matching a plurality of times to remove even a relatively large noise. .

Further, according to the vehicle front monitoring method according to the present invention, a predetermined range of the screen is set as a window, the window is sequentially moved on the screen, and at the same time, it corresponds to the image pickup device existing in the window. Sequentially reading the binary signal and comparing the window with a plurality of reference windows having a range equal to the window and in which the binary signal is set in a pattern having a predetermined continuity. And a step of deciding whether or not they match each other, and a step of recognizing a white line drawn on the road in front of the vehicle based on the result obtained by the step of determining whether or not both match. Therefore, noise having discontinuity can be removed and accurate white line recognition can be performed.

Further, according to the vehicle front monitoring method according to the present invention, a predetermined range of the screen is set as a window, the window is sequentially moved on the screen, and the image pickup device existing in the window is dealt with. Sequentially reading the binary signal and comparing the window with a plurality of reference windows having a range equal to the window and in which the binary signal is set in a pattern having a predetermined continuity. Since the step of determining whether or not they match each other is repeated a plurality of times, even a relatively large noise can be removed.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a first embodiment.

FIG. 2 is an explanatory diagram showing a state of pattern matching.

FIG. 3 is a diagram showing an example of a reference window.

FIG. 4 is a block diagram showing a configuration of window means.

FIG. 5 is an explanatory diagram illustrating a correspondence between window means and an image sensor.

FIG. 6 is a block diagram showing a configuration of a pattern matching unit.

FIG. 7 is an explanatory diagram showing the operation of the determination means.

FIG. 8 is a block diagram showing a configuration of a second embodiment.

FIG. 9 is an explanatory diagram showing the operation of the second embodiment.

FIG. 10 is a block diagram showing a configuration of a conventional device.

FIG. 11 is a diagram showing an image taken by a camera of a conventional device.

FIG. 12 is an explanatory diagram showing a process of a binarizing unit.

[Explanation of symbols]

1: camera, 2: A / D converter, 3: binarization means,
4: White line recognition means, 5: Screen, 6: Image sensor, 7: White line, 9: Noise, 10: Pattern matching means, 1
1: white line recognition means, 12: window, 20: window means, 21: storage element, 22: delay means, 30: reference window, 31: determination means, 32: logical sum circuit, 81
~ 86: Detection range

Claims (5)

[Claims] 1. A camera for photographing the front of a vehicle, and a binarizing means for classifying video signals from a plurality of image pickup devices arranged on the screen of the camera into two kinds of values and outputting a binary signal. A window means for setting a predetermined range of the screen as a window, sequentially moving the window on the screen, and sequentially reading the binary signal corresponding to the image pickup device existing in the window; And a plurality of reference windows in which binary signals are set in a pattern having a predetermined continuity in the window, and the windows are compared with each other by comparing the reference windows with each other. And a white line recognition means for recognizing the white line drawn on the road in front of the vehicle based on the output of the determination means. Front monitoring apparatus for a vehicle according to claim. 2. The window means is a means for arranging storage elements corresponding to the image pickup elements existing in the window, and sequentially storing and updating the signal from the binarizing means by the storage elements. The storage elements corresponding to the image pickup elements that are adjacent in the direction are connected in series to form a plurality of storage element rows, and the storage element rows that are adjacent in the vertical direction of the screen are respectively delayed by delay means. The vehicle front monitoring device according to claim 1, wherein the front monitoring devices are connected in parallel. 3. The vehicle front monitoring device according to claim 1, further comprising a plurality of sets of window means and determination means. 4. A step of classifying video signals from a plurality of image pickup devices arranged on a screen of a camera for photographing the front of a vehicle into two kinds of values and outputting a binary signal, and a predetermined range of the screen. Is set as a window and the window is sequentially moved on the screen, and the binary signals corresponding to the image pickup devices existing in the window are sequentially read, and a range equal to the window is provided. Comparing a plurality of reference windows in which binary signals are set with a pattern having a predetermined continuity in the window with the window to determine whether the two coincide with each other; Recognizing the white line drawn on the road in front of the vehicle based on the result obtained by the step of whether or not Forward monitoring method for vehicles. 5. A predetermined range of the screen is set as a window, the window is sequentially moved on the screen, and 2 corresponding to the image pickup device existing in the window is set.
Sequential reading of the value signal, and comparing the window with a plurality of reference windows having a range equal to the window and having binary signals set in a pattern having a predetermined continuity within the window. 5. The step of determining whether or not the two match with each other is repeated a plurality of times.
The vehicle front monitoring method described.