JPH09264742A - Distance-measuring apparatus and safe running system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a distance-measuring apparatus which does not measure a distance erroneously even when a plurality of objects to be detected exist in a visual field range and to detect respective distances of the plurality of objects to be detected. SOLUTION: A right image which is outputted from a photodetector 23a on one side is shifted one pixel by one pixel by an image shift processing part 24. An agreement judgment part 25 compares the shifted right image with a left image which is outputted from a photodetector 23b on the other side, and it cuts out a region in which an agreement degree is high. A correlation computing part 26 finds the deviation amount between regions in which the agreement degree of the right and left images is found out to be high, and it finds the distance of an object to be detected on the basis of the deviation amount.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a distance measuring device and a safe traveling system. In particular, the present invention relates to a distance measuring device for measuring a distance to a detected object using a phase difference correlation method, and a safe traveling system using the distance measuring device.

[0002]

2. Description of the Related Art FIG. 1 shows a schematic configuration of a conventional distance measuring device 1 for measuring a distance to a detected object in a visual field by using a phase difference correlation method, and FIG. 2 shows an optical configuration thereof. .
This distance measuring device 1 uses low-resolution photodetectors 4a and 4b such as a photodiode (PD), and as shown in FIG. 2, a first light receiving lens 2a having an equal focal length f. And the second light receiving lens 2b have a constant baseline length Q
(Optical axes 3a, 3 of the first and second light receiving lenses 2a, 2b
b) and the first and second
First and second photodetectors 4a and 4b are disposed at positions separated from the light receiving lenses 2a and 2b by the focal length f.

However, in the distance measuring device 1 based on the phase difference correlation method, natural light or light emitted from a light projecting element (not shown) is reflected and diffused on the surface of the object to be detected OBJ. , The reflected light r is reflected by the first light receiving lens 2
The light enters the first photodetector 4a through the first optical path passing through a, and simultaneously enters the second photodetector 4b through the second optical path passing through the second light receiving lens 2b. From the first and second photodetectors 4a and 4b to the microprocessor (C
The light intensity distribution of the received light is output to the PU) 6 as image information. Then, in the microprocessor 6, the image 5a supplied from the first photodetector 4a and the second image 5a
The shift amount D of the barycentric position of the image 5b supplied from the photodetector 4b is calculated. This shift amount D corresponds to X ₁ + X ₂ in FIG. The distance L to the detected object OBJ is
Value of shift amount of center of gravity between images 5a and 5b D = X ₁
By using + X ₂ , it is obtained by the following formula based on trigonometry. L = Q · f / D ... When these operations are executed by the microprocessor 6, the result is output as distance data via the output circuit 7, as shown in FIG.

The principle of obtaining the distance L to the detected object OBJ based on the above equation, that is, the image shift amount D will be described with reference to FIG. Now, the detected object OBJ is located at a distance L from the center plane H of the first and second light receiving lenses 2a and 2b,
When it is located at a point q from the optical axis 3b of the light receiving lens 2b of the first photodetector 4a, the first photodetector 4a has the abscissa X _{1 based} on the position corresponding to the optical axis 3a of the first light receiving lens 2a. image 5a of the detected object OBJ (gravity center position) occurs, the horizontal coordinates X ₂ of the detection object OBJ from its position at that corresponds to the optical axis 3b of the second photodetector 4b second light receiving lens 2b Assuming that the image 5b (the position of the center of gravity) is generated, as shown in FIG. 3, there is a relation as shown below. Q−q = (L / f) X ₁ q = (L / f) X ₂ If q is eliminated from these two equations and X ₁ + X ₂ = D, the above equation is obtained. Therefore, in this way, the two photodetectors 4 are
By using a and 4b, it can be seen that the distance L to the detected object OBJ can be accurately obtained from the formula even if the detected object OBJ is located at an arbitrary position within the visual field.

However, since the object to be detected OBJ has a size, the first and second photodetectors 4a, 4a,
The light receiving pattern in 4b is not a point but a light intensity distribution pattern (image). Therefore, the definition of the deviation amount D cannot be used as it is, and it is necessary to expand the definition of the deviation amount D or the method of obtaining the deviation amount D. In the distance measuring device 1 using the phase difference correlation method, the shift amount D between the images 5a and 5b is obtained as follows. First, the reference line P
Define A and PB. FIG. 4A shows the first and second light receiving lenses 2a and 2b and the first and second photodetectors 4a and 4b.
FIG. 4B shows the first and second photodetectors 4a, 4
In the images 5a and 5b supplied from b, the horizontal axis represents the abscissas XA and XB on the photodetectors 4a and 4b, and the vertical axis represents the light intensity. The reference line PA shown in FIG.
It is a vertical line indicating the position of a in the screen and corresponds to the intersection A with the optical axis 3a on the first photodetector 4a. Similarly, the reference line PB is a vertical line indicating the position of the image 5b in the screen, and corresponds to the intersection B with the optical axis 3b on the second photodetector 4b. In order to obtain the shift amount D between the image 5a supplied by the first photodetector 4a and the image 5b supplied by the second photodetector 4b, first, both images 5a, 5
b is compared and the reference line PA and the reference line PB are aligned on the screen as shown in FIG.
The image 5b of b is moved together with the reference line PB so as to best match the image 5b of the other photodetector 4a. In this way, the reference line PA when the two images 5a and 5b are best matched,
The distance between the reference lines PB is defined as the deviation amount D at this time,
The distance L to the object to be detected OBJ is calculated using the formula. Note that the time when the two images 5a and 5b best match each other is as shown in FIG.
It is defined that the area of the non-overlapping area of both images 5a and 5b is the minimum area, such as the shaded area shown in (d).

In the distance measuring device 1 using such a phase difference correlation method, the two images 5a and 5b shift in the same direction even if there is a color pattern such as a pattern on the surface of the detected object OBJ. The displacement amount D is not affected, and the distance L to the detected object OBJ can be accurately measured even in such a case. Further, since the deviation amount D can be obtained with high accuracy by the correlation calculation, the distance L can be measured with high accuracy even when the detected object OBJ is at a long distance.

[0007]

However, in the method of calculating the distance by the phase difference correlation method using the low-resolution photodetector as described above, there are two or more detected objects in the visual field. , When two or more images of the object to be detected are formed on the photodetector, from the deviation amount when the area of the non-overlapping regions of the left and right images is minimum as described above, There is a problem that even if the distance is calculated, it does not match the distance to any detected object.

Now, two detected objects OBJ1 and OBJ
Suppose two are at different distances. For example, in FIG. 5A, the detected object OBJ1 is located at the distance L1 and the detected object OBJ1
J2 is present at a distance L2 (<L1), and the photodetector 4
As shown in FIG. 5B, an image 8a of the detected object OBJ1 and an image 9a of the detected object OBJ2 are generated on a, and the image is detected on the photodetector 4b as shown in FIG. 5C. Image 8b of detected object OBJ1 and image 9 of detected object OBJ2
b has occurred.

At this time, the image 8a is displayed as shown in FIG.
And images 8a and 8b so as to best match
b, 9a, 9b and the reference lines PA, PB are moved, and the distance obtained from the deviation amount D1 at that time matches the distance L1 to the detected object OBJ1. Further, as shown in FIG. 6B, the respective images 8a, 8b, 9a, 9b and the reference lines PA, PB are moved so that the images 9a and 9b are most matched, and the deviation amount D2 at that time is obtained. The distance matches the distance L2 to the detected object OBJ2.

However, in reality, as shown in FIG. 6 (c), when the degree of coincidence between the entire images 8a and 9a on the photodetector 4a and the entire images 8b and 9b on the photodetector 4b is the highest. The deviation amount D is calculated, and the deviation amount D at that time is D1.
Neither D2 nor two detected objects OBJ1, O
The images 8a, 8b, 9a, 9b of the BJ2 are the photodetectors 4a,
The amount of deviation D is D depending on the position on the image 4b.
Takes any value from 1 to D2.

Therefore, in the distance measuring device 1 by the phase difference correlation method using the low-resolution photodetectors 4a and 4b such as photodiodes, when there are two or more detected objects in the visual field. However, the distance data output from the distance measuring device 1 does not indicate the distance of any of the detected objects, which causes a problem that the distance measurement becomes unstable and the distance is erroneously measured. Moreover, when there are two or more detected objects in the field of view, it is extremely impossible to individually detect the distance to each detected object.

7 and 8 are a schematic diagram showing a conventional distance measuring device 10 using a high-resolution photodetector such as a CCD and a diagram showing the configuration of its optical system. In this distance measuring device 10, the CCD cameras 11a, 11
2D CCD 13 through lenses 12a and 12b
The images 5a and 5b of the object OBJ to be detected are formed on a and 13b (for example, consisting of 512 × 480 pixels), and in the microprocessor 14, the two-dimensional images from the CCD cameras 11a and 11b are matched. The distance to the detected object OBJ is calculated from the deviation amount at that time,
Distance data is output from the output circuit 15.

However, in the distance measuring device 10 using such a high-resolution photodetector, the distance measurement accuracy is good, but the cost is high because the high-resolution photodetector is used. Moreover, the memory for storing the output data from the photodetector also becomes large in capacity. Further, since a large amount of output data has to be processed, there is a problem that it takes a long processing time.

The present invention has been made in view of the above-mentioned drawbacks of the conventional examples, and an object of the present invention is to provide a low-resolution photodetector with a distance to a plurality of detected objects existing in the visual field. The object of the present invention is to provide a low-cost and high-accuracy distance measuring device by making it possible to measure with high accuracy.

[0015]

A distance measuring apparatus according to claim 1 receives light emitted from a predetermined visual field and transmits the light along first and second optical paths spatially separated from each other. And a plurality of light-responsive elements for receiving the light transmitted along the first optical path of the light-receiving optical system and outputting an electric signal representing the intensity distribution of the light. Means and
A second photo-detecting means having a plurality of photo-responsive elements for receiving the light transmitted along the second optical path of the light-receiving optical system and outputting an electric signal representing the intensity distribution of the light; And a calculating means for outputting a predetermined calculation result of the light intensity distribution output from the second light detecting means based on the amount of deviation on the photoresponsive element, as a distance to the detected object. The measuring device further comprises a judging means for judging a region having a high degree of coincidence from the signals representing the light intensity distributions output from the first and second light detecting means, and the calculating means comprises:
It is characterized in that the predetermined calculation is performed on the basis of the amount of shift on the photoresponsive element with respect to the region determined to have a high degree of coincidence by the determination means.

Here, the determination that the degree of coincidence of images is high is not limited to the determination of the images as they are, but the determination may be made after differentiating the image data or by extracting the extreme points. The determination may be performed after the contrasts of the images are equalized, or after the image is subjected to an arbitrary process.

According to the present invention, therefore, the determining means determines the portions having a high degree of coincidence between the two images, and obtains the distance to the detected object from the amount of deviation between the partial images determined to have a high degree of coincidence. ing. Therefore, it is possible to obtain a plurality of distances to the detected object corresponding to the parts with high coincidence,
Even if there are a plurality of detected objects in the field of view, the accuracy of distance measurement does not deteriorate. Further, it is possible to individually detect the distances of a plurality of detected objects.

According to a second aspect of the present invention, in the distance measuring device according to the first aspect, when the determination means determines a plurality of areas having a high degree of coincidence, the calculation means causes the respective areas to be determined. Is calculated, and a result of the predetermined calculation performed based on which of the deviation amounts is the largest is output as the distance to the detected object.

Demonstration when a plurality of areas having a high degree of coincidence are determined. By calculating the distance to the detected object based on the largest amount, the detected object located closest to the plurality of detected objects is detected. Only the distance of the object can be output.

According to a third aspect of the present invention, in the distance measuring device according to the first aspect, when the determination unit determines a plurality of regions having a high degree of coincidence, the calculation unit causes the respective regions to be determined. Is calculated, and all of the predetermined calculation results performed based on the deviations are output as the distance to the detected object.

When a plurality of areas having a high degree of coincidence are determined,
By calculating the deviation amount of the light intensity distribution for each area and calculating the distance from it to the detected object, the distance to multiple detected objects in the field of view can be affected by other detected objects. Instead, they can be individually separated and accurately determined.

The embodiment described in claim 4 is the embodiment described in claim 1
3. The distance measuring device according to 3, further comprising: an extraction unit that extracts a portion satisfying a predetermined condition from the region determined by the determination unit to have a high degree of coincidence, and the calculation unit includes the light intensity of the extracted portion. It is characterized in that the predetermined calculation is performed based on the deviation amount of the distribution on the photo-responsive element.

In this embodiment, the area determined to have a high degree of coincidence by the determining means is further divided by the extracting means according to a predetermined condition, and the deviation amount is calculated for each extracted partial area to calculate the distance. can do.

Therefore, it is possible to further finely measure the distance between the respective parts of the detected object corresponding to the area judged to have a high degree of coincidence by the judgment means, and it is possible to detect even the unevenness of the surface of the detected object. it can.

According to a fifth aspect of the present invention, in the distance measuring device according to the fourth aspect, when the extracting means extracts a plurality of portions satisfying a predetermined condition, the calculating means is configured to respectively It is characterized in that the shift amount of the light intensity distribution is obtained for the portion of (1) and the distance to the detected object calculated using the average value of the plurality of shift amounts is output.

When it is known that the deviation amounts are for the same detected object, erroneous distance measurement is prevented and the distance measurement reliability is improved by using the average value of the deviation amounts. You can

The embodiment described in claim 6 is defined in any one of claims 1 to
The distance measuring device described in No. 5 is further characterized by outputting the direction in which the detected object exists.

By detecting not only the distance of the detected object but also its direction, the state of the detected object can be determined more accurately.

[0029] The embodiment described in claim 7 is based on claim 1
The distance measuring device according to the fifth aspect is further characterized in that the size of the detected object is output.

By detecting not only the distance of the detected object but also its size, the state of the detected object can be determined more accurately.

The safe traveling system according to claim 8 is provided with the distance measuring device according to any one of claims 1 to 7, and the distance measuring device is mounted inside or outside the vehicle compartment to measure the distance to an object existing in the surroundings. It is characterized by that.

Since this safe traveling system is equipped with the distance measuring device, it is possible to reach a desired object to be detected (especially, the nearest front vehicle or rear vehicle) without being disturbed by the background or other objects to be detected. The distance can be detected with high accuracy.

According to a ninth aspect of the present invention, in the safe traveling system according to the eighth aspect, a display unit for displaying a distance to the object based on an output of the distance measuring device is provided near the vehicle speed display meter. It is characterized by being attached.

If a display unit for displaying the distance to the object based on the output of the distance measuring device is provided near the vehicle speed display meter, the inter-vehicle distance to the preceding vehicle can be displayed on the display unit. .

The embodiment described in claim 10 is the embodiment described in claim 8.
The safe traveling system according to claim 1, further comprising a determination device that determines, based on at least the output of the distance measuring device, whether or not the host vehicle is in a dangerous state in which it may collide with an object existing in the vicinity. Is characterized by.

If a judging device for judging whether or not there is a dangerous state in which there is a possibility of colliding with an object existing around based on the output of the distance measuring device, for example, an alarm is issued in advance. To indicate that there is a risk of collision,
A collision accident can be prevented by operating the brake.

The embodiment described in claim 11 is the same as claim 8.
The safe traveling system described above is characterized by including control means for controlling an accelerator opening degree or a brake based on at least the output of the distance measuring device.

If the control means for controlling the accelerator opening or the brake is provided based on the output of the distance measuring device, the speed of the vehicle can be controlled within a safe range according to the state of the vehicle.

The embodiment described in claim 12 is the embodiment described in claim 8.
The described safe traveling system is characterized in that it is provided with an informing means for informing the driver that the vehicle ahead has started based on at least the output of the distance measuring device.

If means is provided for notifying the driver that the vehicle ahead has started based on the output of the distance measuring device, the start of the vehicle ahead can be notified, for example, in the case of traffic congestion. You can make driving easier in a traffic jam.

[0041]

BEST MODE FOR CARRYING OUT THE INVENTION

(First Embodiment) FIG. 9 is a block diagram showing a configuration of a distance measuring device 21 according to an embodiment of the present invention. 22
Reference numerals a and 22b are lenses for focusing the light reflected and scattered by the detected object on the light receiving surfaces of the photodetectors 23a and 23b. Reference numerals 23a and 23b are low-resolution photodetectors such as photodiodes, which have a large number of pixels and output a signal (image signal) indicating the light intensity for each pixel. 24
Is an image shift processing unit that sequentially shifts an image generated by one photodetector, for example, the photodetector 23b on the right side by one pixel in the pixel array direction, and outputs the image. It is output to the coincidence determination unit 25, and the image shift number N at that time is output to the correlation calculation unit 26. Reference numeral 25 compares the images output from the photodetector 23b and sequentially shifted by the image shift processing unit 24 with the images output from the photodetector 23a, and determines a region (pixel region) having a high degree of coincidence between the two images. It is a matching degree determination unit. Therefore, the image shift processing unit 24 and the coincidence degree determination unit 25 shift one image by one pixel with respect to the other image, and discriminate a region having a high degree of coincidence. Reference numeral 26 is a correlation calculation unit that obtains the distance to the detected object from the determination information from the coincidence determination unit 25 and the image shift number N transmitted from the image shift processing unit 24. Two
An output circuit 7 outputs the distance to the detected object obtained by the correlation calculation unit as distance data.

According to the distance measuring device 21 shown in FIG.
As shown in (a), the distance to each of the detected objects can be measured even when there are a plurality of detected objects in the visual field. In FIG. 10A, two detected objects OBJ1 and OBJ4 are present at a distance L1 in the field of view, a detected object OBJ3 is present at a distance L2, and a detected object OB is present at a distance L3.
The case where J2 exists is shown. FIG. 10B shows an image (hereinafter, referred to as a left image) GL generated on the left photodetector 23a, and GL1 is a detected object OBJ1.
GL2 shows a partial image by the detected object OBJ2, GL3 shows a partial image by the detected object OBJ3, and GL4 shows a partial image by the detected object OBJ4. Further, FIG. 10C shows an image (hereinafter referred to as a right image) G generated on the right photodetector 23b.
R indicates GR, GR1 indicates a partial image by the detected object OBJ1, GR2 indicates a partial image by the detected object OBJ2, GR3 indicates a partial image by the detected object OBJ3, and GR4 indicates a part by the detected object OBJ4. The image is shown.

As described in the conventional example, when the distances between the detected objects are different, the deviation amounts of the left and right partial images are also different for each detected object. Therefore, for example, by shifting the right image by one pixel with respect to the left image and cutting out a region where the left and right images have a high degree of coincidence at each shift amount, the image region of each detected object can be separated, By setting a window having a phase difference correlation in that region, the distance for each detected object can be accurately measured. Hereinafter, this method will be specifically described.

Left image G as shown in FIGS. 10 (b) and 10 (c).
When the L and right images GR are obtained, first, the image shift processing unit 24 moves the right image GR received from the photodetector 23b, and the reference line PA of the left image GL is set as shown in FIG. The reference line PB of the right image GR is matched, and this state is used as the reference (origin) of the shift number of the right image GR. Next, the image shift processing unit 24 shifts the right image GR by one pixel (the number of image shifts N = 1) and transmits it to the matching degree determination unit 25. The coincidence determination unit 25 detects the left image G from the photodetector 23a.
In addition to directly receiving L, the right image GR shifted by one pixel from the image shift processing unit 24 is received. The left and right images GL and GR at this time are shown in FIG. The degree-of-coincidence determination unit 25 compares the left and right images GL and GR and searches for an area with a high degree of coincidence. In the case of FIG. 11B, the partial image GL
The degree of coincidence of 1, GR1 and GL4, GR4 is high. Therefore, the coincidence determination unit 25 determines that the partial images GL1 and G
It is determined that R1 and GL4, GR4 are regions with a high degree of coincidence, and the relevant information is output to the correlation calculation unit 26. When the correlation calculation unit 26 receives the information from the coincidence determination unit 25,
The image shift number N output from the image shift processing unit 24 is read. Partial images GL1, GR1 and GL4, G
If the image shift number N = 1 when R4 matches,
Since the pixel pitch is Δ and the shift amount D is D = NΔ = Δ, the left and right partial images GL1, GR1 and G are calculated by the equation.
Detected objects OBJ1 and OBJ4 corresponding to L4 and GR4
To the distance L1.

Next, the image shift processing section 24 determines the right image GR.
Is further shifted by one pixel (image shift number N = 2) and transmitted to the coincidence determination unit 25. The left and right images GL and GR received by the coincidence determination unit 25 at this time are shown in FIG. In this case, since there is no region with a high degree of coincidence, the degree-of-coincidence determination unit 25 transmits the information to the correlation calculation unit 26.

Next, the image shift processing section 24 determines the right image GR.
Is further shifted by 1 pixel (image shift number N = 3) and transmitted to the coincidence determination unit 25. At this time, the coincidence determination unit 25 receives and compares the left and right images GL and GR as shown in FIG. 12D, so that the partial images GL3 and GR3 are determined to be regions with high coincidence, and the correlation calculation unit 26 is instructed. Output the information. At this time, the correlation calculation unit 26 causes the image shift processing unit 24 to
From the image shift number N = 3 transmitted from
= NΔ = 3Δ is obtained, and further the distance L2 to the detected object OBJ3 corresponding to the partial images GL3 and GR3 is obtained.

Next, the image shift processing section 24 determines the right image GR.
Is further shifted by 1 pixel (image shift number N = 4) and transmitted to the coincidence determination unit 25. The left and right images GL and GR received by the coincidence determination unit 25 at this time are shown in FIG. In this case, since there is no region with a high degree of coincidence, the degree-of-coincidence determination unit 25 transmits the information to the correlation calculation unit 26.

Next, the image shift processing section 24 determines the right image GR.
Is further shifted by one pixel (image shift number N = 5) and transmitted to the coincidence determination unit 25. At this time, the coincidence determination unit 25 receives and compares the left and right images GL and GR as shown in FIG. 12 (f), so it is determined that the partial images GL2 and GR2 are regions with high coincidence, and the correlation calculation unit 26 Output the information. The correlation calculation unit 26 calculates the shift amount D = NΔ = from the image shift number N = 5 transmitted from the image shift processing unit 24.
5Δ is obtained, and further the distance L3 to the detected object OBJ2 corresponding to the partial images GL2 and GR2 is obtained.

The distance measuring device 21 sequentially shifts the right image GR one pixel at a time as described above, and repeats the process of determining the region with a high degree of coincidence. If there is a region with a high degree of coincidence, the corresponding image is detected. The distance from the shift number N to the detected object is calculated. FIG. 13 is a diagram showing a relationship between the image shift number N (distance L of the detected object) obtained as a result of the processing of FIGS. 11A to 12F and a region (pixel number) having a high degree of coincidence. Therefore, the hatched region indicates a region with a high degree of coincidence. As shown in this figure, L1, L2, L3 obtained as the distance to the detected object
Is output as distance data from the output circuit 27.

In the conventional phase-difference correlation method, the left and right images are superposed so as to best match each other, and the distance of the object to be detected is calculated from the deviation amount at that time. Therefore, only one distance data can be output. When there are a plurality of detected objects in the visual field, the distance measuring device outputs distance data which is not the distance of any detected object, and the measurement error is bad. On the other hand, the distance measuring device of the present invention sequentially shifts the other image with respect to one image, and if there is a region in which a part of one image substantially matches a part of the other image, The distance to the detected object is output from the time shift amount. Therefore, it is possible to output a plurality of distance data, and it is possible to accurately output each distance data corresponding to a plurality of detected objects.

(Second Embodiment) In the first embodiment,
Distance data to the detected object obtained, for example, L1, L
Although 2 and L3 are all output from the output circuit, only the shortest distance data (that is, the distance data having the largest displacement amount D), for example, L3 (<L2, L1) is output from the output circuit 27. You may This is because it is often the case that only the object to be detected (the preceding vehicle, the obstacle, etc. immediately before) needs to be detected, as in the case of a distance measuring device for a vehicle described later.

For this purpose, the correlation calculation unit 26 determines all distance data L1 to L3, and then the correlation calculation unit 26 or the output circuit 27 determines and outputs the shortest distance data L3. Good.

Alternatively, the calculation may be performed only for the pixel area where the distance to the detected object is the shortest. For example, as shown in FIG. 11A, after the reference lines PA and PB of the left and right images GL and GR are made to coincide with each other, the right image GR is largely shifted to a sufficiently large image shift number N from which the right image GR The image shift number N may be decreased by one pixel. That is, a region with a high degree of coincidence between the left and right images is obtained from the side with the largest number N of image shifts, and if a region with the largest number N of image shifts and a high degree of coincidence is found, the number N of image shifts at that time is detected. The distance to the object may be obtained and output, and the process may be ended there. According to this method, the operation of the distance measuring device is terminated after the data of the shortest distance is obtained, so that no other distance data is obtained, an extra calculation process is omitted, and the processing time is shortened. .

(Third Embodiment) FIG. 14 is a view showing the arrangement of a distance measuring apparatus according to another embodiment of the present invention. In the distance measuring device 28, the image signals GL and GR output from the left and right photodetectors 23a and 23b are differentiated by differentiating circuits 29a and 29b, and then the first embodiment is performed based on the differentiated image signals. Distance data is obtained in the same manner as. That is, assuming that the images GL and GR as shown in FIGS. 15A and 15B are obtained by the left and right photodetectors 23a and 23b, the differentiating circuits 29a and 29b generate the left image G.
The L and right images GR are differentiated and differential images DGL and DGR as shown in FIGS. 15C and 15D are output. The image shift processing unit 24 shifts the differential image DGR of the right image GR received from the differentiating circuit 29b by one pixel, and the coincidence determination unit 25.
Is the differential image DGR received from the image shift processing unit 24.
And the differential image D of the left image GL received from the differentiating circuit 29a.
A region having a high degree of agreement with GL is obtained. When it is determined that there is a region with a high degree of coincidence, the correlation calculation unit 26
Calculates the shift amount D from the image shift number N at that time, calculates the distance L of the detected object, and outputs it. It should be noted that this differential image shows the edge portions (both ends) of the detected object.

In this embodiment, the degree-of-coincidence determination unit 2
5, the differential image DG obtained by differentiating the left and right images GL and GR
Since it is determined whether or not the absolute value of the difference between L and DGR (differential value of light intensity) is less than or equal to a predetermined value, the left and right photodetectors 23a and 23b are affected by the shift of the left and right light receiving optical systems. Even if the light intensity distributions are different, it is possible to test the degree of coincidence by comparing only the change characteristics of the images, and it is possible to improve the accuracy of determination of the degree of coincidence.

(Highly Consistency Area) In the first and third embodiments, the image area having a high degree of coincidence has been described, but it is said that the left and right images (left and right differential images) have a “high degree of coincidence”. Means that the absolute value of the difference between the left and right images (the left and right differential images obtained by differentiating) is less than or equal to a predetermined value. To be more specific about the case of the first embodiment, FIGS.
As shown in FIG. 5, the area K having a high degree of coincidence has an absolute value of a difference between the light intensity value indicated by the left image GL and the light intensity value indicated by the right image GR in the corresponding pixel that is equal to or less than a predetermined value ΔP. Means an area.

When the heights (light intensities) of the left and right images with respect to the same object to be detected are substantially equal to each other, by setting the predetermined value ΔP to a small value, the partial images as described with reference to FIGS. 11 and 12 are obtained. The area K having a high degree of coincidence is output only when the areas coincide with each other.

However, in general, even if the images are generated by the same object to be detected, due to the influence of the optical system,
The heights of the left and right images GL and GR are not the same. Therefore, the value of the predetermined value ΔP needs to have a certain size. In that case, as shown in FIG. 16A, even when the partial images GL and GR do not match, the region K having a high degree of matching is output.

Further, when the value of the predetermined value ΔP is made extremely large, as shown in FIGS. 17A and 17B, the overlapping portion of the left and right images GL and GR is a region K having a high degree of coincidence.
Can be output as In this case, it can be determined by whether or not each pixel is receiving (light having a light intensity equal to or higher than a certain threshold value). Also in this case, FIG.
As shown in (a), even when the partial images GL and GR do not match, the region K having a high degree of matching is output.

As described above, when the area K having a high degree of coincidence is output even when the partial images do not coincide with each other, the number of pixel shifts N when the partial images coincide with each other is performed by performing the following processing. Need to be confirmed. For example, it is assumed that when the partial images GL and GR are sequentially shifted by the pixel shift numbers N1 to N5, the hatched regions in FIGS. 18A to 18E are output as the region K having a high degree of coincidence.
This is shown in FIG. 19 with the vertical axis representing the number of pixel shifts.
It is. Here, FIG. 16 (a) (b) or FIG.
As can be seen from (a) and (b), the width of the region K having a high degree of coincidence output when the partial images GL and GR coincide with each other has a high degree of coincidence outputted when they do not coincide. The width of the area K, which is larger than the width of the area K and has a high degree of coincidence, becomes maximum when the partial images GL and GR coincide with each other.
Therefore, the pixel shift numbers N3 and N1 corresponding to the largest width in the vertical axis direction among the regions K having a high degree of coincidence shown in FIG. 19 are selected and the distance is calculated to detect the detected object. The distance of an object can be calculated.

Similarly, when a differential image is to be handled, FIG.
As shown in (a) and (b), the region K having a high degree of coincidence is
In the corresponding pixel, it means a region in which the absolute value of the difference between the differential value of the light intensity indicated by the left differential image DGL and the differential value of the light intensity indicated by the right differential image DGR is equal to or smaller than a predetermined value ΔP.

Also at this time, the value of ΔP may need to have a certain magnitude. FIG.
As shown in (a) and (b), the left and right differential images DGL, DG
The overlapping portion of R can be output as the area K having a high degree of coincidence. Also in these cases, the differential partial images DGL and DGR as shown in FIGS.
Even if they do not match, the area K with a high degree of matching is output.

Therefore, in these cases, the same processing as in the case of the normal partial image may be performed. That is, the differential partial images DGL and DGR are sequentially pixel-shifted from N1 to N.
When shifted by 5, the hatched area in FIGS. 22A to 22E is output as the area K having a high degree of coincidence,
If the pixel shift number is plotted on the vertical axis as shown in FIG. 23, the pixel shift numbers N3 and N1 corresponding to the widest one of the regions K having a high degree of coincidence are selected, and the distance is calculated. Thus, the distance of the detected object can be obtained.

(Fourth Embodiment) FIG. 24 is a diagram showing the structure of a distance measuring device 30 according to still another embodiment of the present invention. In this distance measuring device 30, the extreme value detection unit 31a for detecting the extreme values (maximum value, minimum value and stationary value) of the images GL and GR output from the left and right photodetectors 23a and 23b, 31b is provided. Then, as shown in FIG.
As shown in FIG. 25, after the extreme value detectors 31a and 31b find the extreme points (indicated by white and black dots in FIG. 25) of the images GL and GR on the photodetectors 23a and 23b, the image shift processing is performed. The right image GR is shifted pixel by pixel by the unit 24, the matching determination unit 25 obtains an area where the extreme points of the left and right images GL and GR match, and the correlation calculation unit 26 shifts the number N of image shifts when the extreme points match. The distance of the detected object is calculated from
According to this embodiment, it is possible to deal with the difference in the light amount distribution due to the influence of the deviation of the optical system.

This embodiment is based on the fact that the shift amount of the image area and the shift amount M of the extreme point coincide with each other, as shown in FIG. Not only the case where they match, but also the case where they are within a certain distance like the part surrounded by the broken line in FIG.

28A to 28E show left and right images GL, G.
Area K in which extreme points coincide with each other by shifting R from N1 to N5
FIG. 29 is a diagram for explaining how to search for, and FIG. 29 is a diagram showing the search result. Also in this case, as can be seen from FIG. 27, when the left and right images GL and GR are coincident with each other, the width of the region K in which the extreme points are coincident with each other becomes maximum, so that the pixel shift number N1 and N3 is selected, and the distance to the detected object is obtained from this.

(Fifth Embodiment) FIG. 30 is a view for explaining a distance measuring device according to still another embodiment of the present invention. In each of the above embodiments, only the distance of the detected object is detected, but in this embodiment, the direction of the detected object is detected together with the distance of the detected object. That is, as shown in FIG. 30, the detected object OBJ exists at the position of the distance L in the direction connecting the center of gravity of the images GL and GR on the photodetectors 23a and 23b and the centers of the lenses 22a and 22b. Image G on the detectors 23a and 23b
Assuming that the coordinates of the center of gravity positions of L and GR are X ₁ and X ₂ , respectively, the detected object O from the centers of the lenses 22a and 22b.
The direction θ of the detected object OBJ drawn toward BJ is obtained from tan θ = (2LX ₁ −fQ) / 2fL = − (2LX ₂ −fQ) / 2fL. According to this equation, the correlation calculator 26 or the like calculates the direction θ of the detected object OBJ, and the output circuit 27 outputs the distance L and the direction θ.

(Sixth Embodiment) In addition, the distance measuring device can detect the distance L (and direction θ) of the object OBJ to be detected as well as the size (size) of the object to be detected. If the distance L of the detected object OBJ is obtained, the size E of the detected object OBJ can be obtained from E = (L / f) e from the similarity relationship shown in FIG. Here, e is a detected object OBJ generated on either photodetector 23a or 23b.
Width of the image GL or GR (or both photodetectors 23
a, the average value of the widths of the images GL and GR on 23b).

(Seventh Embodiment) FIG. 32 is a block diagram showing the configuration of a distance measuring device 32 according to still another embodiment of the present invention. Reference numerals 33a and 33b are cutout units that extract the left partial image and the right partial image that are determined to have a high degree of coincidence by the degree of coincidence determination unit 25. That is, the image shift processing unit 24 shifts the pixel by one pixel, and the matching degree determination unit 2
When it is determined by 5 that the partial images having the left and right images have a high degree of coincidence, the left partial image is cut out from the left image to the cutout unit 33a, and at the same time, the right partial image is cutout from the right image 33b. Is cut out to. An extraction unit 34 extracts a characteristic portion (hereinafter, referred to as an extracted partial image) that satisfies a predetermined condition (extraction condition) from the left partial image cut out to one cutout unit 33a. 3
A selection unit 5 selects a portion (hereinafter, referred to as a selected partial image) having a strong correlation with the extracted portion extracted by the extracting unit 34 from the right partial image cut out to the other cutout unit 33b. The meaning of the strong correlation between the extracted partial image and the selected partial image here will be clarified by each embodiment described later. Then, the correlation calculator 26
Calculates the distance D between the extracted partial image and the selected partial image, calculates the distance to the detected object based on the deviation D, and outputs the calculated distance data from the output circuit 27.

According to the distance measuring device 32, even when a plurality of objects to be detected exist in the field of view, the distances of the respective objects to be detected are detected separately, and even the unevenness of the surface of each object is detected. Can also be detected. Now, it is assumed that there are a plurality of detected objects in the field of view and a left image and a right image are generated on the photodetectors 23a and 23b. Image shift processing unit 24
When the right image is sequentially shifted by, and a certain pixel shift number is reached, the left partial image G is displayed as shown in FIG.
It is assumed that L1 and GL2 and the right partial images GR1 and GR2 are determined to have a high degree of coincidence (at this stage, the distance to the detected object may be obtained from the pixel shift number).
Therefore, the left partial images GL1 and GL2 and the right partial image G
It is determined that R1 and GR2 are images of the same detected object, and the left partial image GL3 and the right partial image GR3 are different partial images. Then, the left partial images GL1 and GL determined to have a high degree of coincidence with a certain pixel shift number.
2 is cut out to the cutout portion 33a (FIG. 33 (b)),
The right partial images GR1 and GR2 are cut out to the cutout portion 33b (FIG. 33 (c)). The extraction unit 34 includes the cutout unit 3
A part satisfying the extraction condition, for example, the left partial image GL1 as shown in FIG. 34A is extracted from the left partial images GL1 and GL2 generated in 3a, and the information is transmitted to the selection unit 35. The selection unit 35 searches the right partial images GR1 and GR2 generated in the cutout unit 33b, and
As shown in (b), the partial image GR1 having a strong correlation with the extracted partial image GL1 is selected. Then, the correlation calculation unit 26 obtains the shift amount between the extracted partial image GL1 and the selected partial image GR1. For example, as shown in FIG. 34C, the shift amount D1 is obtained from the distance between the reference lines PA and PB when the extracted partial image GL1 and the selected partial image GR1 are best matched. Then, the correlation calculation unit 26 calculates the distance L1 from the calculated displacement amount D1 to the corresponding position of the detected object based on the equation.

Similarly, as shown in FIG. 35 (a), left partial images GL1 and GL2 generated in the cutout portion 33a.
35, the left partial image GL2 is extracted, and the selecting unit 15 selects the right partial image GR2 having a strong correlation with the extracted partial image GL2, as shown in FIG. As shown in (c), if the deviation amount D2 is obtained from the distance between the reference lines PA and PB when the extracted partial image GL2 and the selected partial image GR2 are best matched,
It is possible to calculate the distance L2 to the relevant location of the detected object.

In this distance measuring device 32, an extracted partial image satisfying a certain extraction condition is extracted from the left image of the detected object, and a partial image having a strong correlation with the extracted partial image is selected from the right image and extracted. Since the correlation calculation is performed considering only the partial image and the selected partial image, it is possible to separate only a part of the detected object characterized by the above extraction condition and obtain the distance. Therefore, it is possible to accurately measure the distance of each unevenness of the detected object. Alternatively, it is possible to detect a plurality of approaching detected objects with high resolution.

More specifically, the extraction conditions used by the extraction unit 34 are as follows: from the extreme value (maximum value, minimum value or stationary value) of the image to the adjacent extreme value (local minimum value, maximum value or stationary value). It is possible to use the condition of a partial image in the section up to. For example, if the left partial image GL1 and the right partial image GR1 as shown in FIGS. 36A and 36B are cut out by the cutout units 33a and 33b, the extraction unit 34 will be described.
Is a partial image from the left partial image GL1 from the maximum value to the minimum value, for example, 36a (the portion indicated by the thick line), and the other partial images and background noise are cut to remove the partial image 36a as shown in FIG. Extracted partial image 37 in which other parts are interpolated
Generate a. On the other hand, the selection unit 35 selects the corresponding partial image 36b (the portion indicated by the thick line) from the right partial image GR1 and cuts the other partial images and background noise, as shown in FIG.
As in (d), a selected partial image 37b is generated by interpolating a portion other than the partial image 36b. Then, the correlation calculation unit 26
Calculates the distance of the part based on the extracted partial image 37a and the selected partial image 37b of FIGS. By calculating the distance by repeating such an extraction and selection operation, the partial images GL1 and GL1 shown in FIGS.
The unevenness of each part of the detected object characterized by GR1 is detected.

Particularly, since the unevenness of the detected object is represented by the section of the light intensity distribution waveform from the extreme point to the extreme point in the image, a partial image satisfying a predetermined condition is defined as follows from the extreme point. By making it a partial image of the section up to the extreme point,
It is possible to accurately measure the distance of each unevenness of the detected object.

Further, for example, when calculating the distance of a high-brightness detected object such as a tail lamp of a vehicle, the two cutout portions 33a, 33 are affected by the directivity of light.
Output from b (left partial image GL1, right partial image GR1)
There may be a difference in the amount of light. At this time, the left and right partial images GL1, supplied from the two cutout portions 33a, 33b
The distance of the detected object obtained from one set of extracted / selected partial images extracted / selected from GR1 and the distance of the detected object obtained from a different set of extracted / selected partial images are not equal, There is a risk that different distances will be calculated due to the influence of the difference.

As described above, when it is known that the brightness of the detected object is high and the distances between the extracted / selected partial images are equal, the deviation amounts obtained from the extracted / selected partial images are averaged, and the average is averaged. The distance may be obtained by using the deviation amount. As a result, it is possible to correct the shift of the image due to the difference in the light amount and output the correct distance.

(Eighth Embodiment) For example, the distance measuring device 32 may cause a data output failure due to the influence of electrical noise, optical disturbance, or the like. For example, a noise component 38 may occur in the partial image (light intensity distribution) generated in the cutout portions 33a and 33b due to electrical noise, optical disturbance, or the like, as shown in FIG. 37 (a). Therefore, with only the processing method as in the seventh embodiment, as a result of performing the correlation calculation on the noise component 38, there is a possibility that the distance data in which the detected object does not actually exist may be output.

Therefore, in this embodiment, the extraction unit 34
The extraction condition in is the partial image of the section from the extreme value (maximum value, minimum value or stationary value) of the image to the adjacent extreme value (local minimum value, maximum value or stationary value), and the width of that section (below , Edge width) W is greater than or equal to a predetermined value. That is, the extraction condition in the seventh embodiment is weighted at the point where the width (edge width) W of the partial image is equal to or larger than a predetermined value, and the small edge width is not extracted. Therefore, the extraction unit 14 from the image of FIG.
When the partial image is extracted by, the partial images 39 to 45 as shown in FIG. 37B are extracted, and the influence of noise can be removed to output accurate distance data.

Here, the predetermined value which gives the limit of the edge width W can be experimentally determined. That is, the test sample having the steepest edge (for example, the boundary between the white surface and the black surface) is arranged by changing the distance (set distance) from the distance measuring device 32, and then the edge of the test sample is changed. The edge width of the obtained image is measured.
FIG. 38 shows the set distance-edge width characteristic thus obtained. Then, the necessary minimum edge width W is read from FIG. 38 according to the set distance of the detected object to be detected, and the predetermined value is set to the edge width W or a value slightly smaller than that edge width. The extraction unit 34 does not extract a portion having a small edge width. By setting the predetermined value in this way, the partial image with the steepest edge can be extracted by the extraction unit 34, but the spike-shaped partial image due to the influence of electrical noise or the like can be prevented from being extracted. It is possible to prevent erroneous distance measurement to a detected object.

It should be noted that, here, in the case where the extraction unit 34 extracts a partial image in the section from the extreme value to the extreme value, the image having the width less than the predetermined value is not extracted. Even in the case of other extraction conditions such as those described above, it is possible not to extract those that are equal to or less than or equal to a predetermined value.

(Ninth Embodiment) In FIG. 39A, the right image GR is shifted by N1 pixels by the image shift processing unit 24, and the left and right images GL,
It shows a state in which regions with a high degree of GR coincidence are being compared. The state in which the degree of coincidence is high among the left and right images GL and GR is cut out to the cutout portions 33a and 33b. FIG.
(B) shows the left and right partial images GL1 and GR1 cut out to the cutout portions 33a and 33b with the reference lines PA and PB aligned. Then, the extraction unit 34 extracts an arbitrary extracted partial image 46 from the extreme point to the extreme point in the left partial image GL1, and the selecting unit 35 selects the corresponding selected partial image from the right partial image GR1.

Here, the left and right partial images GL1 and GR1 are
Since the pixel shift number is cut out as N1 when the matching degree is checked by the image shift processing section 24 and the matching degree determination section 25, the selected partial image 47 corresponding to the extracted partial image 46 extracted therefrom is also extracted. The number of image shifts should be N1. Therefore, this can be used as a criterion for determining the strength of the correlation in the selection unit 35. That is, with respect to the extracted partial image 46, the partial image 4 in which the number of shifts in units of pixels is relatively far from N1.
7, 49 to 52 are not the selected partial images, and the partial image 48 deviated by about N1 (may have a width of several pixels) in units of pixels is the selected partial image corresponding to the selected partial image. select.

Note that such a condition may not be a selection condition by itself, but may be an additional condition of another selection condition.

In this way, when the selecting unit 35 specifies the selected partial image, if the subsequent phase difference correlation processing is performed in consideration of the number of pixel shifts, for example, even partial images with a repetition period are not separated from each other. Can be prevented.

(Tenth Embodiment) To obtain the distance to the object to be detected by the correlation calculation, as described in the conventional example, the area where the two partial images do not overlap (hereinafter referred to as the correlation area value) is determined. It is calculated from the amount of deviation D when it becomes the minimum. However, for example, the light intensity distributions of the left and right photodetectors 23a and 23b are different due to the influence of optical misalignment.
As shown in (a), the contrast C of the left partial image 53
When the contrast CR between L and the right partial image 54 is different, the correlation area value between the two partial images 53 and 54 becomes minimum due to the difference in contrast CL and CR between the left and right partial images 53 and 54 (referred to as the area minimum point). There is a case where the point where the images overlap each other does not match (FIG. 40 (b)), and therefore an incorrect distance may be calculated. In order to solve this, a partial image output from one photodetector 23a or 23b as shown by a broken line 55 in FIG.
It is advisable to correct the light intensity distribution of No. 4 so that the contrasts of both images 53 and 55 match. By doing so, the point where the correlation area value of both partial images becomes the minimum can be correctly detected (FIG. 40 (c)).

(Eleventh Embodiment) FIG. 41 (a)-
What is shown in (d) is a view for explaining still another embodiment of the present invention. FIG. 41 (a) shows an image in which the left partial image GL1 extracted by the extracting unit 34 and the right partial image GR1 selected by the selecting unit 35 are interpolated (see FIG. 36) and the contrasts are matched. In the left and right partial images GL1 and GR1 in which the contrasts are matched in this way, when one of the left and right partial images is shifted, as shown in FIG.
The left and right partial images GL with the number of pixel shifts in which GR1 is most matched
1 and GR1 have the lowest correlation area value S, and when the distance S is a predetermined number, the left and right partial images GL1 and GR1
Since the contrasts of A and B match, the correlation area value S increases linearly regardless of whether the pixel shift number is large or small. By utilizing this property, the pixel shift number that minimizes the correlation area value S can be obtained by calculation as described below.

That is, as shown in FIG. 41B, for example, the correlation area value when the right partial image GR1 is shifted relatively leftward from the left partial image GL1 (pixel shift number n ₁ ) is S ₁ , As shown in FIG. 41 (c), the correlation area value when the right partial image GR1 is shifted to the right by a relatively large amount from the left partial image GL1 (pixel shift number n ₂ ) is S ₂ ,
If the change in the correlation area value when shifting by one pixel is ΔS, the relationship between the pixel shift number n and the correlation area value S is expressed by the following equation. S = −ΔS (n−n ₁ ) + S ₁ S = ΔS (n−n ₂ ) + S ₂ These two straight lines are shown in FIG. 41 (d). When the pixel shift amount (which is calculated and is not limited to an integer) at the intersection of the two straight lines is calculated, a detailed shift up to a decimal shift between the left and right partial images GL1 and GR1 when the correlation area value S is minimum is obtained. The quantity D can be obtained, and high-speed and highly accurate distance measurement can be realized.

Next, a vehicle equipped with the distance measuring device of the present invention will be described. The mounting location of the distance measuring device is not particularly limited, and it can be mounted at any location according to the application. For example, it can be attached to the inside of the vehicle such as the rear of the rearview mirror or on the dashboard, and can also be attached to the outside of the vehicle such as the bumper and front grille on the front of the vehicle. Of course, there is no problem even if it is attached to the rear or side of the vehicle to monitor the rear or side of the vehicle.

42 and 43 are a side view and an interior view showing an embodiment of such a vehicle 81, in which a distance measuring device 83 is attached to the rear surface of a rearview mirror 82.
A distance display device 85 is attached near the vehicle speed display meter 84. Then, the distance measuring device 83 is used to detect a detected object in the field of view in front of the vehicle 81 (for example, the front vehicle 86
The distance to the object to be detected in front is displayed in real time on the distance display device 85 to notify the driver of the distance to the object to be detected. Therefore, the driver can reliably drive while keeping a safe inter-vehicle distance without relying on the feeling, and it is possible to prevent an accident due to an excessively short inter-vehicle distance due to an illusion or driving fatigue.

Since the distance measuring apparatus according to the present invention can reliably calculate the distance to the detected object, it can be used not only as an inter-vehicle distance display system for detecting an inter-vehicle distance to a vehicle in front, but also as described below. It can be safely used for such a rear-end collision prevention system and a visual field assistance system.

FIG. 44 is a side view showing a vehicle 81 equipped with a rear-end collision prevention system 87 using the distance measuring device 83 according to the present invention. The distance measuring device 83 of the present invention is attached to the back side of the room mirror 82 in the vehicle interior,
The field of view is set above the ground in front of the vehicle 81. Further, a warning device 90 such as a buzzer or an alarm is provided in the driver's seat. The rear-end collision prevention system 87 is configured as shown in FIG. 45, and includes information indicating the inter-vehicle distance to the forward vehicle 86 detected by the distance measuring device 83 and the vehicle in operation detected by the speed sensor 88. The information indicating the speed of the vehicle (vehicle speed) is input to the rear-end collision prevention control unit 89, and the rear-end collision prevention control unit 89 determines the degree of danger based on the inter-vehicle distance to the preceding vehicle 86 and the own vehicle speed. Then, the alarm device 90 is turned on or the brake device 91 is activated to brake the vehicle 81.

FIG. 46 shows a method of determining the degree of risk by the rear-end collision prevention control section 89. The horizontal axis indicates the vehicle speed monitored by the speed sensor 88, and the vertical axis indicates the distance measuring device 83. It is the inter-vehicle distance from the vehicle ahead. In FIG. 46, C ₁ is an allowable area, C ₂ is a low-risk area, and C ₃ is a high-risk area, and the boundaries of these areas are defined as a function of the vehicle speed and the inter-vehicle distance. Although the rear-end collision prevention system 87 does not operate in the permissible area C ₁ ,
When the vehicle enters the low-risk area C ₂ , the alarm device 90 first turns on to alert the driver by a buzzer or an alarm, and when the vehicle enters the high-risk area C ₃ , the brake device 91 automatically operates. , The vehicle 81 is forcibly decelerated or suddenly stopped. Therefore, it is possible to prevent a rear-end collision accident due to the driver's carelessness or falling asleep.

FIG. 47 is a diagram showing an operation flow of the rear-end collision prevention system 87. When the distance measuring device 83 calculates the inter-vehicle distance to the preceding vehicle 86 (S101),
The rear-end collision prevention control unit 89 controls the inter-vehicle distance and the speed sensor 88.
The degree of danger is determined from the own vehicle speed monitored by (S102), and it is determined in which area in FIG. 46 the vehicle is located. Then, it is judged whether or not it is within the allowable region C ₁ (S
103), if it remains within the permissible area C ₁ , the risk level is continuously monitored (S101 to S103). If it is not within the allowable area C ₁ , it is judged whether it is the low-risk area C ₂ or the high-risk area C ₃ (S104), and if it is the low-risk area C ₂ , the alarm device 90 is activated (S105), Danger area C ₃
If so, the brake device 91 is operated (S106).

In the rear-end collision prevention system 87, the danger area is divided into the warning operation area and the brake operation area, but the brake operation area is further divided into two areas. May be forcibly decelerated and the vehicle may be stopped suddenly by applying a sudden brake when the brake operating range becomes higher.

FIG. 48 is a plan view showing a vehicle 81 equipped with a visual field assistance system 92 using the distance measuring device 83 according to the present invention. The distance measuring device 83 according to the present invention includes, for example, left and right side mirrors 93 and a bumper 94 on the rear surface of the vehicle.
Are attached to both left and right sides of the vehicle, and the fields of view (areas hatched in FIG. 48) are set to the vehicle rear side and the vehicle side portion rear side. The visual field assistance system 92 provides visual field assistance when changing lanes and visual field assistance during back parking. An alarm device 95 (or display device) is provided in the driver's seat. As shown in FIG. 49, the information indicating the distance to the detected object behind detected by the distance measuring device 83, the information indicating the own vehicle speed detected by the speed sensor 88, and the back lamp control signal 97. And a turn signal control signal 98 are input to the visibility assistance control unit 96. The visibility assistance control unit 96 determines the degree of danger at the time of changing lanes or at the time of back parking. Turn on and warn the driver.

FIG. 50 is a diagram showing an operation flow of the visibility assisting system 92 when changing lanes. When changing lanes, the turn signal on the turn side blinks for changing lanes, but when the control signal for the right or left turn signal is transmitted (S107), the visibility assist control unit 96 causes the turn signal control signal 98 to be displayed. Is detected, the distance measuring device 83 and the speed sensor 88 on the moving side are operated (S108), and a signal indicating the distance from the distance measuring device 83 to the following vehicle on the rear side is sent to the visibility assist control unit 96. A signal indicating the vehicle speed is output from the speed sensor 88 to the visibility assist control unit 96.
Output to The visibility assist control unit 96 determines the state of the vehicle 81 from these signals and determines the risk of lane change (the risk is a function of the distance to the detected object and the vehicle speed) (S109). , S110). As a result, when it is determined that the information is dangerous, the alarm device 95 is sounded (or displayed on the display device) to warn (S111). Therefore, when the warning is given, the driver can stand by without changing the lane until the warning is released, thereby preventing an accident due to the lane change.

Further, FIG. 51 shows the visibility assisting system 92.
It is a figure which shows the operation | movement flow at the time of back parking. When backing up, the back lamp blinks, but when this back lamp control signal is transmitted (S112), the visibility assist control unit 96 detects the back lamp control signal 97, and the back distance measuring device 83 Activate the speed sensor 88 (S1
13), the distance measuring device 83 outputs a signal indicating the distance to the object to be detected (parked vehicle, wall, child, etc.) to the visibility assist control unit 96, and the speed sensor 88 indicates the vehicle speed. The signal is output to the visual field assistance control unit 96. The visibility assistance control unit 96 determines the state of the vehicle 81 from these signals and determines the risk of back parking (S114, S115). As a result, when it is determined that the information is dangerous, the alarm device 95 is sounded (or displayed on the display device) to warn (S116). Therefore, when the warning is given, the driver can immediately apply the brakes to stop, and it is possible to prevent a collision accident or a contact accident with an obstacle or a child. In this case, an alarm may be sounded even outside the vehicle to warn a child or the like behind the vehicle.

FIG. 52 is a block diagram showing an automatic follow-up control system 121 using the distance measuring device 83 according to the present invention, for semi-automatically following the front vehicle 86 while keeping a substantially constant inter-vehicle distance. It is a device. Also in this case, the distance measuring device 83 of the present invention is attached to the back of the room mirror 82, for example, and the field of view is set forward (FIG. 42). The information indicating the inter-vehicle distance to the forward vehicle 86 detected by the distance measuring device 83 and the information indicating the own vehicle speed detected by the speed sensor 88 are input to the automatic tracking control unit 122, The automatic follow-up control unit 122 controls the accelerator control device 123 and the brake control device 12 according to the inter-vehicle distance and the vehicle speed.
You are controlling 4.

FIG. 53 is a diagram showing a determination method by the automatic tracking control unit 122. The curve E is an ideal curve of the inter-vehicle distance to the preceding vehicle 86 and the vehicle speed when the vehicle runs semi-automatically, and the area F ₁ including the curve E has an inter-vehicle distance that is wider than the ideal inter-vehicle distance or an ideal inter-vehicle distance. It is an ideal area that is allowed even if it is overly clogged. In addition, this ideal region F ₁
Out of the areas outside, the area F ₂ is an area where the inter-vehicle distance is too wide and needs to be adjusted, and the area F ₃ is an area where the inter-vehicle distance L is too narrow and needs to be adjusted. The automatic tracking control unit 122 determines from which output signal of the distance measuring device 83 and the speed sensor 88 the vehicle state is in FIG. 53, and when the vehicle state is in the ideal area F ₁ , the driving of the vehicle 81 is performed. However, when the state of the vehicle 81 deviates from the ideal region F ₁ , the accelerator control device 123 and the brake control device 124 are controlled according to the flowchart of FIG.
Keep track of a safe and optimal vehicle distance.

The control operation of the automatic tracking control system 121 will be described with reference to the flowchart of FIG. The inter-vehicle distance to the front vehicle 86 is calculated by the distance measuring device 83 (S131), and the automatic tracking control unit 122 determines the state of the vehicle 81 from the inter-vehicle distance and the own vehicle speed, and the state of the vehicle 81 and the ideal state. And are compared (S132). As a result, the vehicle 8
When it is determined that the state of 1 is in the ideal area F ₁ , (S
133), the driving is left to the driver without outputting the control signal to the brake control device 124 or the accelerator control device 123, and the comparison between the vehicle state and the ideal state is repeated (S131 to S133). On the other hand, the ideal area F
If it is determined that the _first outer further whether the region F _2, to determine whether the region F ₃ (S134), when it is determined that the area F ₂ of the past away following distance, A control signal is output to the accelerator control device 123, the accelerator opening is increased, and the distance L with the preceding vehicle 86 is reduced (S13).
5). On the other hand, when it is determined that the inter-vehicle distance is in the region C ₃ where the inter-vehicle distance is too close, a control signal is output to the accelerator control device 123 or the brake control device 124 to reduce the accelerator opening or activate the brake. To maintain a sufficient inter-vehicle distance (S136).

FIG. 55 is a block diagram showing a system 125 for supporting driving of a vehicle during traffic congestion. During a traffic jam, it is not possible to predict when the forward vehicle 86 will move forward. Therefore, the driver constantly pays attention to the forward vehicle 86 and immediately after the forward vehicle 86 moves, the vehicle 81
Have to move forward to reduce the distance between cars. For this reason, driving during a traffic jam is tired and irritable. This system is intended to facilitate driving during such traffic congestion. This vehicle 8
Even in No. 1, the distance measuring device 83 of the present invention is attached, for example, behind the room mirror 82, and the field of view is set forward (FIG. 42). The information indicating the inter-vehicle distance to the front vehicle 86 detected by the distance measuring device 83 and the information indicating the own vehicle speed detected by the speed sensor 88 are input to the front vehicle start detection control unit 126. Therefore, the forward vehicle start detection control unit 126 activates the alarm device 127 (or display device) to notify the driver when the inter-vehicle distance is a certain value or more. That is, as shown in FIG. 56, the forward vehicle start detection control unit 126 determines whether or not the own vehicle speed is 0 based on the signal from the speed sensor 88 (S141), and when the own vehicle speed is 0, the vehicle is stopped due to traffic congestion. It is determined that it is in the middle, and the following processing for traffic congestion is executed. The front vehicle start detection control unit 126 monitors the inter-vehicle distance to the front vehicle 86 calculated by the distance measuring device 83 (S142), and the inter-vehicle distance is a fixed value determined in consideration of an appropriate inter-vehicle distance during traffic congestion. It is determined whether the value is equal to or more than the value (S143). Then, when the front vehicle 86 moves forward and the inter-vehicle distance deviates by a certain value or more, the alarm device 12
7 is operated (S144), and the driver is urged to reduce the inter-vehicle distance. By using such a system 125, the driver is relieved of the burden of constantly paying attention to the movement of the forward vehicle 86, and the driver is nervous, for example, during a traffic jam.
Can be softened and have a lot of room.

[Brief description of drawings]

FIG. 1 is a schematic configuration diagram showing a conventional distance measuring device that measures a distance by a phase difference correlation method using a low-resolution photodetector.

FIG. 2 is a diagram showing an optical configuration of the above.

FIG. 3 is a diagram illustrating a principle for measuring a distance to a detected object in the above distance measuring device.

4 (a), (b), (c), and (d) are diagrams for explaining a method of obtaining a deviation amount in the above distance measuring device.

5A is a diagram showing a state in which two detected objects are present in the field of view, and FIGS. 5B and 5C are diagrams showing images of light intensity supplied from a photodetector in this case. .

6 (a), (b), and (c) are diagrams showing how the image shift amount is obtained by different methods in the situation as shown in FIG.

FIG. 7 is a schematic diagram showing a conventional distance measuring device using a high-resolution photodetector.

FIG. 8 is a diagram showing a configuration of an optical system in the above distance measuring device.

FIG. 9 is a diagram showing a configuration of a distance measuring device according to an embodiment of the present invention.

10A is a diagram showing a state in which a plurality of detected objects are present in the visual field of the above distance measuring apparatus, and FIGS. 10B and 10C are diagrams showing a left image and a right image at that time. .

FIG. 11A is a diagram showing a state in which the reference line of the left image and the reference line of the right image are matched. (B) is a diagram showing a state in which the right image is shifted by one pixel. (C) is a diagram showing a state in which the right image is shifted by two pixels.

FIG. 12D is a diagram showing a state where the right pixel is shifted by 3 pixels. (E) is a figure which shows the state which shifted the right pixel only 4 pixels. (F) is a diagram showing a state where the right pixel is shifted by 5 pixels.

FIG. 13 is a diagram showing a relationship between a pixel shift amount and a region having a high degree of coincidence of images, which is obtained in the specific example.

FIG. 14 is a diagram showing a configuration of a distance measuring device according to still another embodiment of the present invention.

15A and 15B are diagrams showing left and right images output from the photodetector, and FIGS. 15C and 15D are diagrams showing left and right differential images differentiated by the differentiating circuit.

16 (a) and 16 (b) are diagrams for explaining the definition of a region where the left and right images have a high degree of coincidence.

17 (a) and 17 (b) are diagrams for explaining different definitions of regions having a high degree of coincidence between left and right images.

18 (a) to 18 (e) are diagrams showing regions in which the images match when the left and right images are shifted from each other.

FIG. 19 is a diagram showing the areas where the left and right images in FIG. 19 coincide with each other together.

20A and 20B are diagrams for explaining the definition of a region having a high degree of coincidence between left and right differential images.

21 (a) and 21 (b) are diagrams for explaining different definitions of regions having a high degree of coincidence in left and right differential images.

22 (a) to 22 (e) are diagrams showing regions in which the images when the left and right differential images are shifted from each other match.

FIG. 23 is a diagram showing together the matching areas of the left and right differential images obtained in FIG. 22.

FIG. 24 is a diagram showing a configuration of a distance measuring device according to still another embodiment of the present invention.

FIG. 25 is a diagram illustrating a method of measuring a distance according to the above embodiment.

FIG. 26 is a view for explaining the principle of the above embodiment.

27 (a) and 27 (b) are diagrams showing a method of determining that extreme points match.

28A to 28E are diagrams showing regions in which extreme points match when left and right images are shifted from each other.

FIG. 29 is a diagram showing together areas where the extreme points obtained in FIG. 28 coincide.

FIG. 30 is a diagram showing a configuration of a distance measuring device according to still another embodiment of the present invention.

FIG. 31 is a diagram showing a configuration of a distance measuring device according to still another embodiment of the present invention.

FIG. 32 is a diagram showing a configuration of a distance measuring device according to still another embodiment of the present invention.

FIG. 33 is a view for explaining the above embodiment,
(A) is a diagram showing left and right images on the photodetector, (b) is a diagram showing a left partial image cut out by the cutout portion, (c)
FIG. 6 is a diagram showing a right partial image cut out by a cutout unit.

34A is a diagram showing an extracted partial image extracted by an extracting unit, FIG. 34B is a diagram showing a selected partial image selected by a selecting unit, and FIG. 34C is a diagram showing an extracted partial image and a selected partial image. It is a figure which shows a mode that correlation calculation is performed.

35A is a diagram showing another extracted partial image extracted by the extracting unit, FIG. 35B is a diagram showing another selected partial image selected by the selecting unit, and FIG. It is a figure which shows a mode that the correlation calculation is performed by the selected partial image.

36 (a) and (b) are diagrams showing left and right partial images cut out by a cutout unit, FIG. 36 (c) is a diagram showing extracted partial images interpolated by an extraction unit, and FIG. It is a figure which shows the selected partial image which was selected and interpolated.

FIG. 37 (a) is a diagram showing an image containing a noise component,
(B) is a figure which shows the image from which the noise component was removed by the extraction condition.

FIG. 38 is a diagram showing a set distance-edge width characteristic of the distance measuring device.

39A and 39B are diagrams for explaining another extraction condition.

40 (a), (b) and (c) are views for explaining the characteristics of still another embodiment of the present invention.

41 (a), (b), (c) and (d) are views for explaining still another embodiment of the present invention.

FIG. 42 is a side view showing a vehicle according to the present invention.

FIG. 43 is a partially cutaway schematic view showing the interior of the above vehicle.

FIG. 44 is a side view showing a vehicle equipped with the collision prevention system of the present invention.

FIG. 45 is a block diagram showing the collision prevention system of the above.

FIG. 46 is a diagram for explaining a method of determining a risk degree in the same collision prevention control unit.

FIG. 47 is a flowchart showing a processing procedure in the above collision prevention system.

FIG. 48 is a plan view showing a vehicle equipped with the visual field assistance system of the present invention.

FIG. 49 is a block diagram showing the same field-of-view assistance system.

FIG. 50 is a flowchart showing a processing procedure when changing lanes in the visual field assistance system.

FIG. 51 is a flowchart showing a processing procedure for back parking in the visibility assistance system.

FIG. 52 is a block diagram showing an automatic tracking system of the present invention.

FIG. 53 is a diagram showing an ideal region of a vehicle state in the above automatic tracking.

FIG. 54 is a flowchart showing a processing procedure in the above-mentioned automatic tracking system.

FIG. 55 is a block diagram showing a system for facilitating driving during traffic congestion according to the present invention.

FIG. 56 is a flowchart showing a processing procedure in the above system.

[Explanation of symbols]

 22a, 22b Lens 23a, 23b Photodetector 24 Image shift processing section 25 Matching determination section 26 Correlation calculation section 29a, 29b Differentiation circuit 31a, 31b Extreme value detection section 33a, 33b Cutout section 34 Extraction section 35 Selection section 81 Vehicle 87 Rear collision Prevention system 92 Visibility assist system 121 Automatic tracking control system 125 Driving support system for traffic congestion

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI Technical display location G01C 23/00 G01C 23/00 Z G08G 1/16 G08G 1/16 C // G01C 21/00 G01C 21/00 A (72) Inventor Junichi Takagi 10 Odoron-cho, Hanazono-cho, Ukyo-ku, Kyoto City Kyoto Prefecture

Claims (12)

[Claims] 1. A light receiving optical system for receiving light emitted from a predetermined visual field and transmitting the light along first and second optical paths spatially separated from each other, and a plurality of photoresponsive elements, The light receiving optical system includes a first photodetector that receives the light transmitted along the first optical path of the light receiving optical system and outputs an electric signal representing the intensity distribution of the light, and a plurality of photoresponsive elements. Second light detecting means for receiving the light transmitted along the second optical path of the system and outputting an electric signal representing the light intensity distribution; and light intensity distributions output by the first and second light detecting means. of,
A distance measuring device comprising: a calculation unit configured to output a predetermined calculation result based on a shift amount on the photoresponsive element as a distance to a detected object, wherein the first and second light detection units include From the signal representing the light intensity distribution to be output, a determination unit that determines a region with a high degree of coincidence is provided, and the calculation unit includes a region on the photoresponsive element for a region determined to have a high degree of coincidence by the determination unit. A distance measuring device, characterized in that the predetermined calculation is performed based on a deviation amount. 2. The distance measuring device according to claim 1, wherein when the determination unit determines a plurality of regions having a high degree of coincidence, the calculation unit causes the deviation amount of the light intensity distribution for each of the regions. And a predetermined calculation result obtained based on which one of the displacement amounts is the largest, is output as the distance to the detected object. 3. The distance measuring device according to claim 1, wherein when the determination unit determines a plurality of regions having a high degree of coincidence, the calculation unit causes the deviation amount of the light intensity distribution for each of the regions. Is calculated and all the predetermined calculation results obtained based on the deviations are output as the distance to the detected object. 4. The distance measuring device according to claim 1, further comprising: an extracting unit that extracts a portion satisfying a predetermined condition from a region determined by the determining unit to have a high degree of coincidence. A distance measuring device, wherein the means performs the predetermined calculation based on the amount of deviation of the light intensity distribution of the extracted portion on the photoresponsive element. 5. The distance measuring device according to claim 4, wherein when the extracting means extracts a plurality of portions satisfying a predetermined condition, the calculating means calculates the light intensity distribution of the respective portions. A distance measuring device, wherein a displacement amount is obtained, and a distance to a detected object calculated using an average value of a plurality of displacement amounts among the displacement amounts is output. 6. The distance measuring device according to claim 1, further comprising outputting a direction in which a detected object exists. 7. The distance measuring device according to claim 1, further comprising outputting the size of the detected object. 8. A safe traveling comprising the distance measuring device according to any one of claims 1 to 7, wherein the distance measuring device is mounted inside or outside a vehicle in order to measure a distance to an object existing in the surroundings. system. 9. The safe traveling system according to claim 8, wherein a display unit for displaying a distance to the object based on an output of the distance measuring device is attached near the vehicle speed display meter. Safe driving system. 10. The safe traveling system according to claim 8, wherein, based on at least the output of the distance measuring device, whether the host vehicle is in a dangerous state in which there is a possibility of colliding with an object existing in the surroundings. A safe traveling system characterized by comprising a judging device for judging. 11. The safe traveling system according to claim 8, further comprising control means for controlling an accelerator opening degree or a brake based on at least the output of the distance measuring device. 12. The safe traveling system according to claim 8, further comprising informing means for informing a driver that the vehicle ahead has started based on at least the output of the distance measuring device. Safe driving system.