JPH07333339A - Obstacle detector for automobile - Google Patents

Abstract

PURPOSE:To obtain an obstacle detector for automobile in which high reliability is ensured even under low visibility conditions of nighttime, for example. CONSTITUTION:A pulse light L1 is transmitted from a light transmitting means 1 and a pulse light L2, reflected on an object 30, is received by a light receiving means 3 where the received light is subjected to photoelectric conversion. A distance measuring means 6 then measures the distance R to the object based on the time difference between the transmitting timing of pulse light and the receiving timing of reflected pulse. Furthermore, means 11 picks up the image of the object based on the reflected pulse light and image processing means 12-15 recognize the object based on an image signal G received from the image pickup means 11 thus enhancing the sensitivity of the image pickup means using the pulse light for illumination.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention irradiates pulsed light toward an object and receives reflected pulsed light from the object,
The present invention relates to an obstacle detection device for an automobile that obtains a distance to an object based on the round-trip time of pulsed light and reflected pulsed light, and in particular, has improved reliability by recognizing the object based on an image signal that uses pulsed light as illumination. The present invention relates to an obstacle detection device for automobiles.

[0002]

2. Description of the Related Art Conventionally, a laser radar as disclosed in, for example, Japanese Patent Laid-Open No. 2-228579 is well known as an obstacle detecting device for an automobile of this type. FIG. 9 is a block diagram showing a configuration of a conventional vehicle obstacle detection device. In the figure, 10 is a laser radar mounted on the host vehicle, and 30 is an object such as a preceding vehicle to be detected. The laser radar 10 includes the following elements 1-6.
It consists of

Reference numeral 1 denotes a light transmitter for driving a light emitting element such as a laser diode to send pulsed light L1 toward an object 30.
Reference numeral 2 is a clock generator that generates a clock pulse CL that is the generation timing of the pulsed light L1, 3 is a light receiver that converts the reflected pulsed light L2 from the object 30 irradiated with the pulsed light L1 into an electric signal EL, and 4 is a clock A sample pulse generator 5 that generates a sample pulse SP based on the pulse CL is a sample hold circuit that holds the electric signal EL in response to the sample pulse SP.

Reference numeral 6 denotes a CPU including a distance measuring means as well as a control function and a calculation function, which controls the light transmitter 1 and the distance R to the object 30 based on the sample pulse SP and the sampled and held electric signal EL. To calculate.

Next, the operation of the conventional vehicle obstacle detection device configured as shown in FIG. 9 will be described with reference to the explanatory view of FIG. In FIG. 9, the horizontal axis represents time t,
The vertical axis represents the distance R from the vehicle to the object 30 at each time
Are shown respectively.

First, the clock generator 2 generates the clock pulse CL, and the light transmitter 1 transmits the pulsed light L1 in synchronization with the clock pulse CL. The pulsed light L1 is reflected by the object 30, becomes reflected pulsed light L2, and is received by the light receiver 3. The light receiver 3 photoelectrically converts the reflected pulsed light L2 into an electric signal EL, which is input to the sample hold circuit 5.

On the other hand, the sample pulse generator 4 counts the clock pulses CL, and delays the clock pulses CL by a time (N · Δt) obtained by multiplying the count value N by Δt (time corresponding to distance resolution). A sample pulse SP is generated. The sample hold circuit 5 has a sample pulse SP
In response to, sample the electrical signal EL from the light receiver 3,
This is held until the next sample pulse SP.

The CPU 6 compares the electric signal EL held by the sample hold circuit 5 with a predetermined threshold value L0,
The electric signal EL having a threshold value L0 or more is detected as a signal corresponding to the reflected pulsed light L2 from the object 30. And
The distance R from the host vehicle to the object 30 is calculated by the following equation (1).

R = NΔt × (c / 2) (1)

However, in the equation (1), N is the count value of the clock pulse CL in the sample pulse generator 4,
N · Δt is the delay time of the sample pulse SP, and c is the speed of light.

That is, the count value N of the clock pulse CL
The round-trip distance (= 2) to the object 30 by multiplying the time difference between the pulsed light L1 and the reflected pulsed light L2 obtained from
R) is obtained, and 1/2 of the round trip distance is the obtained distance R.
The count value N of the clock pulse CL returns to 0 when it reaches a value corresponding to the maximum detection distance.

By repeating the above operation as one cycle, the distance R can be continuously measured. Therefore, the CPU 6 can determine that the obstacle, that is, the object 30 is approaching the host vehicle when the time change of the calculated distance R shows a decreasing tendency as shown in FIG. 10, for example. In particular, it can be seen that the area A, which changes greatly with time, is in a state of rapid approach.

A scanning method can be used to measure the distance R to the object 30 more finely. Although not shown, this type of vehicle obstacle detection device is disclosed, for example, in Japanese Examined Patent Publication No. 55-1555, and radiates the pulsed light L1 in a plurality of directions by scanning the light transmitter 1 in the horizontal direction. The reflected pulsed light L2 from each direction is received by the light receiver 4. In this case, for example, when at least one of the distance data in a plurality of directions has a decreasing tendency as shown in FIG. 10, it is determined that the object (obstacle) 30 is approaching.

Further, a method of detecting an object by using an image picked up by a camera or the like is well known, and as an obstacle detecting apparatus for an automobile of this type, for example, "identification and tracking of a preceding vehicle on a highway" is used. (Computer vision,
1985.7.18) and "Driving support system using three-dimensional image recognition technology" (Preprint 9 of the Academic Conference of the Society of Automotive Engineers of Japan)
24, 1992-10) and the like.

Of the obstacle detection devices for automobiles using images as described above, the former content is a technique of limiting the obstacles on the road to the vehicle and utilizing the left-right symmetry of the vehicle.
The latter content is a method in which a stereo image sensor is divided into small areas to obtain distance data, and an obstacle is recognized from the spatial distribution of each distance data.

[0016]

As described above, the conventional obstacle detection device for an automobile detects a reflected pulse from the object 30 in the device for detecting the time difference between the light transmission and reception (the pulsed light L1 and the reflected pulsed light L2). The distance R to the object 30 is measured by receiving the light L2, and based on the information on the distance R, it is detected that the object 30 is an obstacle.
However, it is not possible to identify whether the object 30 is an obstacle on the road or a reflector of the guardrail based on only the reflected pulsed light L2.

Therefore, when the vehicle is traveling on a road having a large curve or a steep slope, it may be regarded as an obstacle even if the detected object 30 such as a road surface or a guardrail does not actually obstruct the traveling. However, there is a problem that reliability is low as data for final obstacle detection. Moreover, in the device using the image,
There was a problem that it was not practical because it became undetectable when the visibility was poor, such as at dusk or at night.

The present invention has been made to solve the above-mentioned problems, and an automobile having improved reliability by using a method of recognizing an object irradiated with pulsed light by image processing is also used. An object is to provide an obstacle detection device for use.

[0019]

According to a first aspect of the present invention, there is provided a vehicle obstacle detection device for transmitting a pulsed light and a pulsed light reflected from an object irradiated with the pulsed light. Then, the light receiving means for photoelectrically converting, the distance measuring means for measuring the distance to the object based on the time difference between the light sending timing of the pulsed light and the light receiving timing of the reflected pulsed light, and the object is imaged based on the reflected pulsed light. It is provided with an image pickup means and an image processing means for recognizing an object based on an image signal obtained from the image pickup means.

According to a second aspect of the present invention, there is provided the obstacle detecting device for an automobile according to the first aspect, wherein the image pickup means is integrally formed in close proximity to the light transmitting means and the light receiving means.

The obstacle detecting device for an automobile according to claim 3 of the present invention is the same as that according to claim 1 or 2.
The light transmitting means irradiates the pulsed light in a plurality of directions, and the distance measuring means measures the distance to the object in the plurality of directions.

According to a fourth aspect of the present invention, there is provided the vehicle obstacle detection device according to any one of the first to third aspects, in which the image processing means detects a white line and the lane in which the vehicle is traveling. The distance measuring means determines the distance to the object by validating only the reflected light in the lane.

According to a fifth aspect of the present invention, there is provided the vehicle obstacle detection device according to any one of the first to fourth aspects, wherein the image pickup means has a direction perpendicular to the optical axis of the reflected pulsed light. The image processing means are installed as a pair apart from each other.
The distance to the object is obtained based on the shift amount of each image signal from the pair of image pickup means.

[0024]

According to the first aspect of the present invention, the obstacle is recognized by using the distance measurement information based on the time difference between light transmission and reception and the object recognition information based on the object shape in the image, and the illumination of the light transmitting means is performed. By enhancing the sensitivity characteristic of the image pickup means, the reliability of obstacle recognition is improved.

Further, according to a second aspect of the present invention, the image pickup means, the light transmitting means and the light receiving means are integrally configured to downsize the entire apparatus.

Further, according to the third aspect of the present invention, the pulsed light is irradiated in a plurality of directions to measure the distance, and the reliability is further improved.

According to a fourth aspect of the present invention, the white line in the image is detected to identify the lane in which the vehicle is traveling,
The reliability is further improved by determining the distance to the object by validating only the reflected light in the lane.

Further, according to a fifth aspect of the present invention, a pair of image signals is acquired from a pair of image pickup means which are separated from each other in a direction perpendicular to the optical axis of the reflected pulsed light, and based on the deviation amount of each image signal. The reliability is further improved by calculating the distance to the object and comparing the distances based on the difference between the light transmission and reception times.

[0029]

【Example】

Example 1. (Corresponding to Claim 1, Claim 3 and Claim 4) Hereinafter, Embodiment 1 of the present invention will be described with reference to the drawings.
1 is a block diagram showing the configuration of a first embodiment of the present invention, in which 1 to 6, 10 and 30 are the same as those described above. Reference numeral 20 denotes an image sensor that captures an image of the object 30 based on the reflected pulsed light L2 and performs image processing, and includes the following elements 11 to 15.

Reference numeral 11 denotes a video camera having an image pickup visual field B, which constitutes an image pickup means for generating an image signal G within the image pickup visual field B. 12 is a digital image signal D for the image signal G
An AD converter for converting to G, an image memory 13 for storing a digital image signal DG for one screen, and an image memory 1 for 14
An arithmetic unit that performs various arithmetic processes on the data in 3
A CPU 5 performs processing such as white line detection in association with the image memory 13 and the arithmetic unit 14.

The AD converter 12 to the CPU 15 constitute image processing means for recognizing the object 30 based on the image signal G. In addition, the CPU 15 in the image sensor 20
Is cooperating with the CPU 6 in the laser radar 10, and C
Data of distance R is taken in from PU6.

Next, the operation of the first embodiment of the present invention shown in FIG. 1 will be described. In this case, the method for detecting the object (obstacle) 30 and the procedure for calculating the distance R in the laser radar 10 are the same as those in the conventional example, except that the image sensor 20 is added. In addition, here, in order to measure the distance R finely by scanning, the laser radar 1
It is assumed that 0 uses the scanning type light transmitter 1.

In the image sensor 20, the video camera 1
The image signal G obtained from 1 is converted into a digital image signal DG by the AD converter 12 and stored in the image memory 13 as image data. The CPU 15 is the image memory 1
The image data in 3 is set to be input to the calculator 14, the calculation start time and the calculation mode such as addition or subtraction are output to the calculator 14, and the result is fetched from the calculator 14.

At this time, the pulsed light L1 from the light transmitter 1 serves as a light source for illuminating the imaging visual field B of the video camera 11, and the influence of noise or the like is exerted even in a situation where the contrast is low such as twilight or night. A clear image signal G with less noise is obtained.

Next, the processing operation of the image sensor 20 according to the first embodiment of the present invention will be described in more detail with reference to the explanatory view of FIG. 2 and the flowchart of FIG.
FIG. 2 is an image or graph of the processing progress by the image sensor 20, (a) is an image captured by the video camera 11, (b) is a road surface area in the image (a),
(C) is a lateral edge point in the road surface area (b), (d) is a histogram of the lateral edge point (c), (e) is a measurement range of each distance data, (f) is a lateral edge point (c). And an image obtained by synthesizing the measurement range (e), (g) shows the object 3 in the image (a).
The respective states in which the distance to 0 is displayed are shown.

In FIG. 2 (c), H is the position of the obstacle candidate obtained from the histogram of the lateral edge points (FIG. 2 (d)), and W is the window used for distance measurement in Example 3 described later. Is. In addition, in FIG.
V is a vertical position for taking a histogram (frequency of lateral edge points), and Vp is a peak point of the histogram.
In (e), e1 to e5 are measurement areas of each distance data.

First, in the white line detection processing (step S40), white line candidate points are extracted from the image of FIG.
That is, by utilizing the fact that there is a brightness difference between the road part and the white line part, the brightness difference is binarized with a predetermined threshold value, or the brightness signal is differentiated to obtain the change rate of the brightness difference. Ask. Such white line detection processing is described in, for example, Japanese Patent Laid-Open No. 3734/1993.
This is a general method as disclosed in Japanese Patent Publication No. 98 and the like.

Subsequently, a road surface detection process (step S4)
In 1), the area surrounded by the white line candidates detected in the white line detection processing (step S40) is detected as a road surface, as indicated by the shaded area in FIG. 2B.

Next, a horizontal edge detection process (step S4)
2C, the points on the road detected in the road surface detection process (step S41) are compared with those in FIGS.
Then, the horizontal edge points are detected and the histogram of the horizontal edge points for each vertical position is taken. Such lateral edge detection processing is performed by, for example, “computer image processing” (industry publication,
This can be carried out by using a well-known horizontal edge detection filter or the like as described in page 20).

In addition, obstacle candidate selection processing (step S4)
In 3), search vertically from the bottom of the screen,
The peak point Vp of the histogram is obtained, and it is determined that there is an obstacle candidate, that is, the object 30 at the position H of the peak point Vp.
This processing is performed when many obstacles such as a car and a motorcycle are present in the driving lane of the own vehicle, and many lateral edge points are present on the contact surface between the road surface area (FIG. 2B) and the obstacle (object 30). (Fig. 2
This is a method of recognizing the shape of an obstacle by utilizing the occurrence of (c)).

Next, a distance data input process (step S4)
4), from the laser radar 10 to each measurement area e1
Data indicating a plurality of distances R within ~ e5 (for example, "1
The distance data composed of 0, 15, 15, 22, 80 and 35) is taken in as shown in FIG.

Finally, obstacle detection processing (step S4)
5), as shown in FIG. 2F, measurement areas e1 to e
5 and the road surface area (hatched portion) are combined, and the distance data (FIG. 2) calculated in the distance data input process (step S44) is calculated.
From (e)) and the position H of the obstacle candidate selected in the obstacle candidate selection process (step S43), the distance R to the object 30 (obstacle) is determined as shown in FIG. In this case, the distance R is displayed on the screen as the distance data “22 m” corresponding to the measurement area e3.

That is, the laser radar 10 is shown in FIG.
As in (e), the distance data of each distance measuring range e1 to e5 is output. Further, in the obstacle detection process (step S45), the image sensor 20 applies each distance data to the position corresponding to the input image (a), as shown in FIG. 2 (f), and within the road surface area (hatched portion). The distance "22" of the measurement area e3 at the position H including the largest number of horizontal lines (horizontal edge points) is regarded as the distance to the obstacle, and is added to the input image as shown in FIG.

Thus, based on the distance data from the laser radar 10 and the obstacle candidate position data from the image sensor 20, the distance R to the obstacle in the own lane is set.
Can be detected. At this time, since the pulsed light L1 from the light transmitter 1 acts as illumination, the video camera 1
The imaging sensitivity of 1 can be increased. Further, as described above, when at least one of the distance data in each direction due to the scanning of the light transmitter 1 shows a decreasing tendency as shown in FIG. 10, the existence of the dangerous object 30 approaching is recognized. be able to.

Here, the image sensor 20 first detects a white line, cuts out the road surface, and recognizes a lateral edge point within the range of the road surface as an obstacle candidate. However, other recognition processing, for example, a conventional example will be described. Even if the method utilizing the left-right symmetry of the vehicle is used, similarly reliable obstacle recognition can be realized. Further, in the laser radar 10, the white line position may be acquired from the image sensor 20, and only the reflected light within the own lane may be made effective to obtain the distance R.

Example 2. (Corresponding to Claim 2) In the first embodiment, the image sensor 20 is arranged apart from the laser radar 10, but the image sensor 20 is arranged in close proximity to the laser radar 10 so as to be integrated with the laser radar 10. Good. FIG. 4 is a block diagram showing a configuration of a second embodiment of the present invention in which a laser radar and an image sensor are integrated, and in FIG.
5 to 11 to 14 and 30 are the same as described above.

Reference numeral 40 denotes an obstacle detection unit in which the laser radar 10 and the image sensor 20 (see FIG. 1) are integrally formed, and a CP having the functions of both the CPUs 6 and 15.
Including U41. In this case, the distance calculation process and the obstacle detection process are the same as those described above, and thus the description thereof is omitted.

Like the obstacle detection unit 40 in FIG. 4, by integrating the laser radar 10 and the image sensor 20 in the first embodiment, errors due to disconnection of communication lines and the like can be avoided and the number of parts can be reduced. Cost reductions can be realized through reductions and miniaturization.

Example 3. (Corresponding to claim 5) In the first embodiment, the image signal G is obtained by using one video camera 11, but a stereo image signal may be obtained from a pair of video cameras. FIG. 5 is a block diagram showing a configuration of a third embodiment of the present invention using a pair of video cameras. In the figure, 1 to 6, 10 and 30 are the same as those described above. Further, 20A corresponds to the image sensor 20 and has the following configuration.

11a and 11b are a pair of video cameras constituting a stereo camera, B1 and B2 are imaging fields of view of the video cameras 11a and 11b, M is a distance between the video cameras 11a and 11b, and G1 and G2 are video. A pair of image signals imaged by the cameras 11a and 11b, and 12a and 12b are image signals G1 and G2, respectively.
AD converters for AD-converting DG, DG1 and DG2 are AD
It is a digital image signal after conversion.

Reference numerals 13a and 13b are image memories for storing the respective digital image signals DG1 and DG2, and 14A is a computing unit for performing various computations on the image signals, that is, the image data in the image memories 13a and 13b. 15A
Are the image memories 13a and 13b and the arithmetic unit 14
A CPU that controls A, reads out image data in each of the image memories 13a and 13b, and executes a predetermined arithmetic processing by the arithmetic unit 14A.

In this case, the procedure for calculating the distance R in the laser radar 10 is the same as that in the first embodiment, and only the stereo image processing configuration in the image sensor 20A is different. That is, in the image sensor 20A, the obstacle candidate selection process is the same as that in the first embodiment, but the two image cameras G1 and G2 are provided corresponding to the provision of the two video cameras 11a and 11b. Is provided with a pair of image processing means for processing.

Therefore, the arithmetic unit 14A and the CPU 1
5A employs a general method of obtaining a parallax from image shifts of two video cameras 11a and 11b in order to realize a distance measuring function by a stereo camera. This kind of method is, for example, “Measurement of inter-vehicle distance by stereoscopic vision using matching method for each partial area” in “Technical Report of Television Society” (Vol. 12, No. 24, pp. 1 to 6). Can be referred to.

Next, the processing operation of the third embodiment of the present invention shown in FIG. 5 will be described with reference to the flowchart of FIG. 6 and the explanatory view of FIG. 7 together with FIG. In FIG. 6, S40 to S44 are the same steps as in FIG. 3, and S45A corresponds to step S45,
The difference is that the obstacle candidate distance measurement processing step S50 is inserted.

First, a pair of video cameras 11a and 1a
One of the image data based on the image signals G1 and G2 from 1b is selected as a reference image, and the image processing steps S40 to S43 in FIG. 6 are executed for this reference image. That is, as shown in FIG. 2B, the white line detection process (step S40) and the road surface detection process (step S41) are performed from the image of FIG.
As described above, the lateral edge processing (step S42) and the obstacle candidate distance measuring processing (step S43) are performed.

Next, obstacle candidate distance measuring processing (step S5)
In 0), with respect to the reference image, a quadrangle centered on the obstacle candidate position H and inscribed in the own lane is set as the window W for distance measurement (see FIG. 2C), and the obstacle in the window W is set. Measure the distance RW to the object. That is, by using the window W set on the reference image as the reference window, the displacement amount n between the two images is obtained by detecting the most approximate window from the other image, and the distance RW is obtained from the displacement amount n. .

At this time, the distance RW to the obstacle in the reference window W is the focal length f of one video camera 11a corresponding to the reference image and the distance M between the two video cameras.
(Referred to as the baseline length), and is calculated from the following equation (2).

RW = f × M / n (2)

Subsequently, a distance data input process (step S
At 44), the distance data R based on the equation (1) is input from the laser radar 10. In the obstacle detection process (step S45A), the distance data R is compared with the distance data RW based on the equation (2), and the difference ΔR (= | R−RW |) between the distances R and RW is a predetermined value. If r or less, 2
It is determined that the two distances R and RW are the same and that the distance R to the obstacle is indicated, and the same processing as in the first embodiment is performed.

On the other hand, when the difference ΔR between the two distances is larger than the predetermined value r, it is necessary to select one of the distances R or RW depending on, for example, the running mode or the time series change of the distance data. For example, as shown in FIG. 7A, when the distance data R based on the laser radar 10 is unstable and the distance data RW based on the image sensor 20A is stable, the stable distance on the image sensor 20A side. Data RW is adopted.

However, as shown in FIG. 7B, at a certain time t
If the two distance data R and RW are completely different while remaining stable at o, it is regarded as a device error and neither distance data is adopted. At this time, an alarm display or the like may be performed if necessary. As a result, it is possible to obtain a highly safe device in which the risk determination based on the wrong distance is prevented and the fail-safe function is added.

In this case, in the image sensor 20A, the window W is set at the obstacle candidate position H, but the image is divided into small areas, the distance data RWi is obtained for each area, and the spatial distribution of each distance data RWi is obtained. A method of recognizing an obstacle may be used, and similarly highly reliable obstacle recognition can be realized.

Further, as described above, the image processing may be performed using the left-right symmetry of the vehicle, and in the laser radar 10, only the reflected light within the own lane is effective based on the white line position information from the image sensor 20A. The distance R may be calculated as Also, the pair of video cameras 11a and 11b are
If the direction is perpendicular to the optical axis of the reflected pulsed light L2,
Needless to say, they may be arranged in any direction such as up and down or left and right, and the same effect is obtained.

Example 4. Further, in each of the above embodiments, data communication is performed between the image sensor 20 or 20A and the laser radar 10, but the communication direction may be unidirectional or bidirectional, and a new obstacle recognition Each data may be transmitted to the unit.

FIG. 8 is a block diagram schematically showing a configuration example of the fourth embodiment of the present invention in which an obstacle recognition unit is added. In the figure, 10 and 20 are the same as those described above. Reference numeral 22 denotes the laser radar 10 and the image sensor 20.
Is an obstacle recognition unit connected to the laser radar 10 and the image sensor 20 and receives distance data and obstacle candidate data from the laser radar 10 and the image sensor 20 to recognize an obstacle.

In this case, the laser radar 10 and the image sensor 20 transfer the distance data R and RW, obstacle candidate data and the like to the recognition unit 22, and the obstacle recognition unit 22 uses the received data to detect obstacles. Performs recognition processing and the like. Therefore, the obstacle recognition unit 22 is
The functions of the CPUs 6 and 15 (see FIG. 1) in the laser radar 10 and the image sensor 20 can also be combined.

Also in the configuration as shown in FIG. 8, similarly to each of the above-described embodiments, the reception of the reflected light from the object other than the obstacle is prevented, and the highly reliable obstacle based on the reflected pulsed light L2 and the image data is obtained. Object recognition can be realized. That is, by using the distance measuring function based on the time difference between transmitting and receiving light and the object recognizing function based on image data, even in a poor visibility condition such as twilight or night, the obstacles can be reliably detected without erroneously detecting the guardrail or the sign. Can be detected.

[0068]

As described above, according to the first aspect of the present invention, the light transmitting means for transmitting the pulsed light, and the light receiving for photoelectrically converting the reflected pulsed light from the object irradiated with the pulsed light. Means, distance measuring means for measuring the distance to the object based on the time difference between the timing of sending the pulsed light and the timing of receiving the reflected pulsed light, an imaging means for imaging the object based on the reflected pulsed light, and an imaging means Since an image processing means for recognizing an object based on the image signal obtained from is provided and the sensitivity of the imaging means is increased by using pulsed light as illumination,
There is an effect that an obstacle detection device for an automobile having improved reliability can be obtained even in a poor visibility condition such as at night.

According to a second aspect of the present invention, in the first aspect, the image pickup means is integrally formed in close proximity to the light transmitting means and the light receiving means, so that the reliability is improved and the size is reduced. There is an effect that an obstacle detection device for an automobile that realizes the above can be obtained.

According to a third aspect of the present invention, in the first or second aspect, the light transmitting means irradiates the pulsed light in a plurality of directions, and the distance measuring means determines a plurality of distances to the object. Since the measurement is performed in the direction of, there is an effect that an obstacle detection device for a vehicle with further improved reliability can be obtained.

According to a fourth aspect of the present invention, in any one of the first to third aspects, the image processing means detects the white line to identify the lane in which the host vehicle is traveling,
Since the distance measuring means determines only the reflected light in the lane to obtain the distance to the object, there is an effect that an obstacle detecting device for a vehicle with further improved reliability can be obtained.

According to a fifth aspect of the present invention, in any one of the first to fourth aspects, a pair of the image pickup means are installed in a direction perpendicular to the optical axis of the reflected pulsed light. Then, the image processing means obtains the distance to the object based on the shift amount of each image signal from the pair of image pickup means, so that an obstacle detection device for a vehicle with further improved reliability can be obtained. There is.

[Brief description of drawings]

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

FIG. 2 is an explanatory diagram showing a processing operation of the image sensor according to the first embodiment of the present invention.

FIG. 3 is a flowchart showing the processing operation of the image sensor according to the first embodiment of the present invention.

FIG. 4 is a block diagram showing a configuration of a second embodiment of the present invention.

FIG. 5 is a block diagram showing a configuration of a third embodiment of the present invention.

FIG. 6 is a flowchart showing a processing operation of an image sensor according to Embodiment 3 of the present invention.

FIG. 7 is an explanatory diagram showing a distance measuring operation according to a third embodiment of the present invention.

FIG. 8 is a block diagram schematically showing a configuration of a fourth embodiment of the present invention.

FIG. 9 is a block diagram showing a configuration of a conventional vehicle obstacle detection device.

FIG. 10 is an explanatory diagram showing an example of general distance data.

[Explanation of symbols]

1 light transmitter, 2 clock generator, 3 light receiver, 4 sample pulse generator, 5 sample hold circuit, 6,
15, 15A, 41 CPU, 10 laser radar, 1
1, 11a, 11b Video camera, 12, 12a, 1
2b AD converter, 13, 13a, 13b image memory, 14, 14A arithmetic unit, 20, 20A image sensor, 30 object, 40 obstacle detection unit, EL electric signal, GG1, G2 image signal, H obstacle candidate position , L1 pulsed light, L2 reflected pulsed light, R distance,
S40 white line detection processing step, S41 road surface detection processing step, S43 obstacle candidate detection processing step, S
44 distance data input processing step, S45 obstacle detection processing step, S50 obstacle candidate distance measurement processing step.

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI Technical display location G08G 1/16 C

Claims (5)

[Claims] 1. A light sending means for sending pulsed light, a light receiving means for receiving and photoelectrically converting reflected pulsed light from an object irradiated with the pulsed light, a sending timing of the pulsed light, and the reflection Distance measuring means for measuring the distance to the object based on the time difference from the light receiving timing of the pulsed light, imaging means for imaging the object based on the reflected pulsed light, and an image signal obtained from the imaging means An obstacle detection device for an automobile, comprising: an image processing unit that recognizes the object based on the object. 2. The obstacle detecting apparatus for an automobile according to claim 1, wherein the image pickup means is formed integrally with the light transmitting means and the light receiving means in proximity to each other. 3. The light transmitting means emits the pulsed light in a plurality of directions, and the distance measuring means measures a distance to the object in the plurality of directions. The automobile obstacle detection device according to claim 1 or 2. 4. The image processing means detects a white line to identify a lane in which the host vehicle is traveling, and the distance measuring means determines a distance to the object by validating only reflected light in the lane. Claim 1 characterized by the above-mentioned.
The obstacle detection device for an automobile according to any one of claims 1 to 3. 5. The pair of the image pickup means are installed in a direction perpendicular to the optical axis of the reflected pulsed light, and the image processing means is provided with a shift amount of each image signal from the pair of image pickup means. The obstacle detection device for an automobile according to any one of claims 1 to 4, wherein the distance to the object is obtained based on the above.