JPH09311998A - Object detection device and parking lot management system using it - Google Patents

Abstract

PROBLEM TO BE SOLVED: To eliminate the malfunctions and to ensure the stable detecting operations of an object detection device by measuring the distance between a light receiving part and a detected object or a background object and deciding whether or not the detected object exists between the background object and the light receiving part. SOLUTION: The images sent from the right and left optical paths are detected, and the distance L is measured between a sensor 1 and a mark 11. Then it's decided whether or not the measured distance L is equal to the distance Lm between the sensor 1 and the mark 11 which is previously stored in a storage part. When both distances L and Lm are equal to each other, the absence of a parking vehicle 26 is decided. On the other hand, the presence of the vehicle 26 is decided when both distances L and Lm are not equal to each other or the distance L cannot be measured. That is, the sensor 1 detects the image of the mark 11 (Fig. a) and the distance L is calculated. Meanwhile, an upper image of the vehicle 26 is detected (Fig. b) and the distance is detected between the upper part of the vehicle 26 and the sensor 1. Thus, the distance L is detected by calculating the shift value X between the right and left images. Then the presence or absence of the vehicle 26 is decided based on the distance L.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an object detecting device for measuring a distance to a detected object or a background object and detecting the presence or absence of the detected object based on the distance, and a parking lot management system using the same. is there.

[0002]

2. Description of the Related Art Conventionally, a system using a TV camera is known as a parking detection method for detecting whether or not a vehicle is parked in a parking space. FIG. 19 shows Japanese Patent Laid-Open No. 63-2.
It is a figure which shows the parking detection method shown by 05800 publication. In this parking detection method, a clear mark 102 is provided on the road surface in the parking space 101, and the mark 102
The vehicle 104 is detected by the V camera 103.
It is to determine whether or not the car is parked. FIG. 1A shows a case where the vehicle 104 does not exist in the parking space 101, and the mark 1 is displayed in the field of view 105 of the TV camera.
Since the image No. 02 can be detected, it is determined as an empty space. FIG. 7B shows the case where the vehicle 104 exists in the parking space 101, and since the mark 102 is not detected in the visual field 105 of the TV camera, it is determined that the parking space 101 is closed. In this method, the presence or absence of parking is detected by pattern matching whether or not the mark 102 exists in the image detected by the TV camera 103, and therefore the mark 102 must be stably detected.

FIG. 20 shows a state in which the TV camera 103 is looking at the mark 102 when the ground 106 is dark and the mark is bright, and a one-dimensional image in the X-axis direction obtained by the TV camera 103 at this time. It is a figure which shows one. Parts A and B are dark parts corresponding to the ground, and C
The part is a part corresponding to the bright mark 102. As can be seen from the figure, pattern matching of the image corresponding to the mark is not so difficult.

[0004]

However, in the parking detection method described above, when the surface of the mark 102 becomes dirty or scratched, or the shadow of the adjacent vehicle is reflected by the mark 102.
It becomes difficult to carry out pattern matching when FIG. 21 is a diagram showing a mark 102 that is soiled or scratched by the surface being stepped on by a tire or the like, and a one-dimensional image in the X-axis direction obtained by the TV camera 103 at this time. Further, FIG. 22 is a diagram showing a mark 102 on which a shadow of an adjacent vehicle is approaching and a one-dimensional image in the X-axis direction obtained by the TV camera 103 at this time. FIG.
As shown in FIG. 1 and FIG. 22, when the surface of the mark 102 is soiled or scratched, or when the shadow of an adjacent vehicle is on the mark 102, the one-dimensional image in the X-axis direction obtained by the TV camera 103 is not obtained. Since it becomes a rule and it is difficult to perform pattern matching to determine that the image is the same as the image shown in FIG. 20, the presence / absence detection of the parked vehicle becomes unstable. The present invention has been made in order to solve the above-described problems, and does not malfunction even if the detected object or background object is dirty, scratched, or changes in the surrounding environment, and the presence or absence of the detected object is stable. An object of the present invention is to obtain an object detection device capable of discriminating between a vehicle and a parking lot management system using the same.

[0005]

In order to achieve the above object, the present invention receives light from an object to be detected or a background object existing in a light receiving field and outputs a light receiving signal based on the received light amount. Distance measuring means including one light receiving means, and measuring the distance between the light receiving means and the detected object or the background object based on the triangulation distance measuring method in which the distance between the two light receiving means is the base line length, and the distance measuring means. The object detecting device is provided with a judging means for judging whether or not the detected object is present between the background object and the light receiving means based on the distance measurement output from the means. In this configuration, the distance to the object to be detected or the background object existing in the light receiving field of view is detected by the triangulation method, and the object to be detected is determined when this distance is shorter than the distance between the light receiving means and the background object. Since it is judged to be present, as compared with the method of detecting a detected object by pattern matching using a TV camera, it does not malfunction even if the detected object or a background object is stained, scratched, or the surrounding environment changes, The presence / absence of an object to be detected can be stably determined.

Further, the present invention is an object detecting device, which is a passive type detecting device for detecting the presence or absence of an object to be detected by receiving light from the object to be detected or a background object. The distance may be measured based on the amount of displacement between the two images. In this configuration, the distance is detected based on the amount of shift between the left and right images, and the presence / absence of the detected object is determined based on the detected distance. The presence or absence can be detected.

Further, in the present invention, the background object may be a mark installed on a road surface of a parking space. In this configuration, the distance to the mark is detected by the triangulation method, and when a distance shorter than the mark is detected, it is determined that there is a parked vehicle, so the surface of the mark is not stained or scratched. The presence or absence of a vehicle can be stably detected even if the vehicle is in the shadow or is shaded by an adjacent vehicle.

Further, the present invention is a parking lot management system comprising the above-mentioned object detecting device, and detecting whether or not there is a parked vehicle in the parking space based on the discrimination output by this device. In this configuration, the above-described object detection device detects whether or not there is a vehicle in the parking space, and thus an administrator can grasp which parking space is vacant.

[0009]

Embodiments of the present invention will be described below with reference to the drawings. (First Embodiment) FIG. 1 is a schematic diagram showing the configuration of a parking detection sensor (object detection device) according to the first embodiment and a light intensity pattern of an image obtained by a light receiving portion of this sensor. The parking detection sensor 1 is a passive type sensor that performs detection by receiving reflected light β from a background object or a detected object of natural scattered light (for example, a lighting device or sunlight) at a place where the parking detection sensor 1 is placed. The light receiving portion of the sensor 1 includes the light receiving lenses 12a and 12b (focal length f) for receiving the reflected light β from the mark 11 (object to be detected) and the CCD.
And image detectors 13a and 13b formed of a photodiode (hereinafter referred to as PD) array or the like. The optical axes 14a and 14b of the two light receiving lenses 12a and 12b are arranged in parallel at an interval L0, and the image detectors 13a and 13b are rearwardly separated from the light receiving lenses 12a and 12b by a distance f equal to their focal length. Light receiving lens 1 at the condensing surface (focal surface)
It is arranged perpendicular to the optical axes 14a and 14b of 2a and 12b. Further, L is the mark 11 and the light receiving lenses 12a and 12
The distance between the light receiving lens 12a and the light receiving lens 12a,
Image detectors 13a, 1 from optical axes 14a, 14b of 12b
The distance to the images 15a and 15b on 3b is shown.

The scattered reflection light β when the surface of the mark 11 is irradiated with the light from the surrounding light sources is the light receiving lens 12a,
The light is collected on the light receiving surfaces of the image detectors 13a and 13b via 12b, and images 15a and 15b (light intensity patterns) as shown in FIG. 1 are generated on the left and right.

FIG. 2 is a block diagram of the sensor 1. The signal processing method of the sensor 1 will be described with reference to FIG. The images 15a, 15b are image detectors 13a, 1
The image signal Sa is converted from 3b and output. This image signal Sa is converted into an image signal S by the quantizers 16a and 16b.
It is converted into an electric signal b and is transferred to the signal processing unit 17. The distance calculation unit 18 (distance measuring means) of the signal processing unit 17 calculates the distance L from the signal Sb to the mark, and this distance L
Is compared with the distance Lm to the mark stored in advance in the storage unit 20 by the comparison unit 17 (determination means), and the comparison result Sc is output via the output unit 21.

FIG. 3 is a schematic view showing the positional relationship between the sensor 1, the mark 11 and the like, and the distance measuring method of the sensor 1 will be described with reference to this drawing. Light receiving lens 12 for mark 11
Assuming that the lateral displacement of b from the optical axis 14b is q (the direction away from the optical axes 14a and 14b of the light projecting lens is a positive direction), L0-q = x1 (L / F) The relationship of q = x2 · (L / f) is established. If q is deleted from these two equations and the mutual shift amount (shift amount) X = x1 + x2 of the images 15a and 15b on the image detectors 13a and 13b is set, L = L0.f / X is obtained. The distance L to the mark 11 can be obtained by detecting the mutual shift amount X of the images 15a and 15b on the detectors 13a and 13b.

FIG. 4A shows the quantizers 16a and 16 described above.
b and the image detectors 13a and 13b for explaining the principle of obtaining the mutual shift amount (deviation amount) X of the images 15a and 15b in the signal processing unit 17, and FIGS. Indicates an image (light intensity pattern) obtained by the image detectors 13a and 13b. When the reflected light β reflected and diffused on the surface of the mark 11 is incident on the image detectors 13a and 13b through the respective light paths that pass through the light receiving lenses 12a and 12b, the image is detected on the image detectors 13a and 13b. , Images 15a and 15b having light intensity patterns as shown in FIG. In FIG. 6B, the horizontal axis represents the horizontal coordinates (pixel arrangement direction) XA, XB on the image detectors 13a, 13b, and the vertical axis represents the light intensity. The reference PA is a vertical line indicating the position of the image 15a on the screen, and corresponds to the intersection A with the optical axis 14a on the image detector 13a. Also, the reference PB
Is a vertical line indicating the position of the image 15b on the screen, and corresponds to the intersection B with the optical axis 14b on the image detector 13b.

Image 1 provided by image detector 13a
Image 15b supplied by 5a and image detector 13b
To find the mutual shift amount X of
As shown in (c), the reference PA and PB are matched on the same screen, and then the image 15b of one image detector 13b is obtained.
Are shifted together with the reference PB so as to best match the image 15a of the other image detector 13a. Thus, the reference PA and the reference P when the two images 15a and 15b best match
Let the distance between B be the mutual shift amount X. The state in which the two images 15a and 15b are most in agreement means a state in which the area of the non-overlapping region 15a and 15b of both images is the minimum, as shown by the hatched region in FIG. Thereby, the distance L from the sensor 1 to the mark 11 can be detected.

Next, a vehicle detection method using the sensor 1 will be described with reference to FIGS. 5 and 6. FIG.
(A) is a perspective view of the vicinity of the sensor 1 when the parked vehicle 26 is present in the parking space 25, and (b) is a perspective view of the vicinity of the sensor 1 when the parked vehicle 26 is not present in the parking space 25. . Further, FIG. 6 shows the quantizers 16 a, 1 a of the sensor 1.
6b and the signal processing unit 17 are flowcharts for determining the presence / absence of a parked vehicle 26 in the parking space 25. First,
Images from two right and left optical paths are detected (S1), and the distance L is obtained by the above method (S2). Next, it is determined whether or not the detected distance L is equal to the distance Lm to the mark 11 stored in the storage unit 20 in advance (S3), and when they are equal (YES in S3), it is determined that there is no parked vehicle 26. Yes (S4). If they are different or if the distance cannot be detected (NO in S3), it is determined that there is a parked vehicle 26 (S).
5). In the case of FIG. 5A, since the sensor 1 detects the image of the mark 11, the distance L to the mark 11 is calculated. Further, in the case of FIG. 5B, since the upper image of the vehicle 26 is detected, the distance to the upper portion of the vehicle 26 is detected. However, if there is no contrast on the upper portion of the vehicle 26, the accurate distance is not calculated, and it is determined that the parked vehicle 26 is not present.

As described above, in the parking detection method using the present sensor 1, the distance L is detected by obtaining the deviation amount X between the left and right images 15a and 15b, and the distance L is used to determine the presence / absence of the vehicle 26. The presence or absence of the parked vehicle 26 can be detected regardless of the distribution of the left and right images 15a and 15b. In other words, no malfunction occurs when detecting marks with dirt or scratches on the surface of the tire ((b) in FIG. 21) or marks with a shadow of the adjacent vehicle ((b) in FIG. 22). The presence / absence of the vehicle 26 can be detected more accurately than in the parking detection method by pattern matching using the TV camera.

FIG. 7 is a diagram showing a structure for parking detection according to a modification of this embodiment. The parking detection method shown in FIG. 5 detects the presence or absence of the vehicle 26 in one parking space 25 with one sensor 1, but the parking detection configuration shown in FIG. 7 has a plurality of parking spaces 25a, 25b. , 2
Marks 11a, 11b, 11 on 5c, 25d, respectively
c and 11d are provided, and all the marks 11a and 11d are provided.
The sensor 1 is installed so that b, 11c, and 11d are in the visual field. Parking space 25 as shown in FIG.
When there is no vehicle at a, 25b, 25c, 25d, all marks 11a, 11b, 11c, 1 are detected by the sensor 1.
1d is placed in the light receiving field, and the marks 11a, 1
Since the distances to 1b, 11c, and 11d can be detected, it is determined that there are no parked vehicles in the parking spaces 25a, 25b, 25c, and 25d. Further, as shown in FIG. 7B, when the vehicles 26a and 26d are parked in the parking spaces 25a and 25d, only the mark 11b and the mark 11c are present in the light receiving visual field of the sensor 1.
The distance between b and mark 11c can be detected, but mark 1
The distance between 1a and the mark 11d cannot be detected. Therefore, in the parking spaces 25a and 25d, the parked vehicles 26a,
26b, and it is determined that there are no parked vehicles in the parking spaces 25b and 25c. In this way, if one sensor 1 determines the distances to a plurality of marks, a plurality of parking spaces can be managed by one sensor 1.
Also, since the wiring on the ceiling is reduced, the installation cost can be reduced.

If the image detectors 13a and 13b in which photodiodes (hereinafter referred to as PDs) are arranged only in the one-dimensional direction are used, the size of the sensor 1 can be reduced and the cost can be further reduced. Can be achieved. Also, the number of pixels of the images 15a and 15b is reduced, so that the calculation time can be shortened (hereinafter, the image detectors 13a and 13b will be described as one-dimensional image detectors).

(Second Embodiment) FIG. 8 is a diagram showing a structure of a mark according to the second embodiment and an image of the mark detected by the parking detection sensor. The mark 28 has a pattern 29 on its surface, so that even if the mark 28 is installed on the ground 30 of the same color as the mark 28, the parking detection sensor 1 does not affect the color of the ground 30. , The mark 28 can be detected.

The reason will be described below with reference to FIG.
FIG. 9 is a diagram showing an image around the mark when the ground and the mark have the same color (white), and an image of the mark detected by the parking detection sensor at that time. The area surrounded by the broken line is sensor 1.
In the right (or left) field of view, the right image (or left image) by the sensor 1 has a pattern as shown in FIG. As described above, when the ground 30 and the mark 31 have the same color, the reflectance is the same, so that sufficient contrast cannot be obtained in the image, and the distance calculation by the sensor 1 described above becomes difficult. In the embodiment shown in FIG. 8, by giving a pattern 29 to the surface of the mark 28,
Even if the ground 30 and the mark 31 have the same color, a sufficient contrast can be obtained in the image detected by the sensor 1, so that the distance can be detected. If the mark 28 is paint applied to the ground 30, the mark will not be destroyed by the tire or the like.

FIG. 10 is a structural diagram of marks according to a modification of the second embodiment. The mark 33 has a shape elongated in the direction (Y-axis direction) orthogonal to the array direction (X-axis direction) of the PDs of the one-dimensional image detector that is the sensor 1, whereby the Y-axis of the sensor 1 is formed. This facilitates the adjustment of the light receiving visual field 34 in the direction. The reason will be described below with reference to FIG. FIG. 11 is a diagram showing the positional relationship between the mark 35 and the light receiving field 34 when the small mark 35 is arranged. In the one-dimensional image detector, the field of view in the vertical direction (Y-axis direction) is narrower than the field of view in the PD array direction (X-axis direction), so that the mark 35 is placed in the field of view 34 in Y direction.
Axial adjustments are difficult to make. Since the mark 33 shown in FIG. 10 extends long in the Y-axis direction, the field of view 3
The image of the mark can be captured regardless of the position of 4 in (a), (b), or (c), and the visual field adjustment in the Y-axis direction becomes easy.

12 and 13 are structural views of marks according to a modification of the second embodiment. Although the mark 33 shown in FIG. 10 described above facilitates the visual field adjustment in the Y-axis direction, the marks 36 and 37 shown in FIGS.
This makes it easy to adjust the visual field in the axial direction. The reason will be described below with reference to FIG. FIG. 14 is a diagram showing the positional relationship between the marks 35 and the light-receiving visual field 34 when the small marks 35 are arranged. In this case as well, visual field adjustment in the X-axis direction is necessary. Since the mark 36 shown in FIG. 12 has a striped pattern 36a on its surface, it is possible to obtain the contrast of the mark 36 at any position of the visual field 34 (a), (b), or (c). The visual field adjustment of the sensor 1 in the axial direction becomes easy. Further, as shown in FIG. 13, the same effect can be obtained by using a plurality of marks 37.

FIG. 15 is a diagram showing marks according to a further modification of the second embodiment. In some parking lots 40 having a plurality of parking spaces 25, numbers 41 indicating the respective positions and symbols (alphabets, etc.) are painted. If the number 41 or the symbol is used as a mark and the mark is detected by the sensor 1, it is not necessary to prepare the mark again. FIG.
As shown in (a), the partition line 4 between the parking spaces 25
2 may be used by marking, and in this case, the sensor 1 is installed so that the partition line 42 is seen from diagonally above, and the distance L to the partition line 42 is detected by this. As shown in FIG. 16B, when the parked vehicle 26 is present, the partition line 42 is covered by the vehicle 26, so that the distance to the partition line 42 cannot be detected, and the parked vehicle 2
6 is determined to be present.

FIG. 17 is a diagram showing the structure of a mark according to a further modification of the second embodiment. The mark 43 comprises a light emitting element 44 and a diffusion plate 45 for diffusing the light projected from the light emitting element 44. As a result, the sensor 1 can obtain a sufficient contrast of the image of the mark 43 even in a pitch dark space, so that the distance to the mark 43 can be detected. In addition, it is safe for the user because his / her feet are bright.

Further, the surface of the mark may be formed of a night-light or fluorescent material, and in this case, the mark 43 shown in FIG. 17 does not have a complicated structure. However, the same effect as the mark 43 can be obtained. Further, cost reduction can be achieved. Further, if the mark is paint, a luminous paint or a fluorescent paint may be used.

(Third Embodiment) FIG. 18 is a block diagram of a parking management system according to the third embodiment. The parking management system 50 has a parking detection sensor 1 and a mark for each parking space, and outputs a signal from each sensor 1 (for example, ON when the vehicle is parked, OFF when the vehicle is not parked). The information is collected by the statistic section 51, the location of the currently vacant parking space, the utilization rate of the parking space depending on the time zone, and the like are statisticized and displayed on the display 52. As a result, the manager can know which parking space is available and guide the visitors to the available parking space. Further, by directly transmitting the information to the display board 53 that displays the location of the empty space, the visitor can see the display board 53 and move to the nearest parking space.

[0027]

As described above, according to the present invention, the distance between the light receiving unit and the detected object or the background object is measured based on the triangulation method, and the background object and the light receiving unit are determined from the measurement result. Since it determines whether or not there is a detected object in between, compared to the conventional method of detecting the presence or absence of the detected object by pattern matching, it can detect stains, scratches, or changes in the surrounding environment on the detected object or background object. Even if it does not malfunction, stable detection can be performed. Further, according to the present invention, the object detection device is a passive type detection device, and the distance is measured based on the amount of deviation between the two images received by the light receiving means. The presence or absence of the detected object can be detected even with such a distribution. Further, according to the present invention, by setting the background object as a mark installed in a parking space, the distance to the mark is detected, and the presence or absence of a parked vehicle is determined by this, so that the surface of the mark may be clean or dirty. The presence or absence of a vehicle can be stably detected even if it is scratched or the shadow of an adjacent vehicle is applied. Also,
According to the present invention, by using the parking lot management system including the above-mentioned object detection device, it is possible to accurately determine the presence / absence of a parked vehicle, so that more appropriate parking lot management becomes possible.

[Brief description of drawings]

FIG. 1 is a diagram showing a configuration of a parking detection sensor according to a first embodiment of the present invention and a light intensity pattern of an image obtained by a light receiving section of the sensor.

FIG. 2 is a block diagram of a parking detection sensor according to the first embodiment.

FIG. 3 is a diagram showing a positional relationship of marks, sensors and the like according to the first embodiment.

4A and 4B are diagrams for explaining the principle of obtaining the mutual shift amount of two images according to the first embodiment, in which FIG. 4A is a configuration diagram in the vicinity of an image detector, and FIGS. 4B to 4D are image detectors. It is a figure which shows the light intensity of the image obtained by.

5A is a perspective view around a sensor when a parked vehicle is present in the parking space according to the first embodiment, and FIG. 5B is a perspective view around a sensor when there is no parked vehicle in the parking space.

FIG. 6 is a flowchart for determining whether or not a parked vehicle is present by a sensor.

FIG. 7A is a diagram showing a state when there is no vehicle in the parking space according to the modified example of the first embodiment, and FIG. 7B is a diagram showing a state when there is a vehicle in the parking space.

FIG. 8 is a diagram showing a mark configuration according to a second embodiment and an image of the mark detected by a parking detection sensor.

FIG. 9 is a diagram showing an image of a mark when the mark and the ground have the same color and an image of the mark detected by the parking detection sensor.

FIG. 10 is a configuration diagram of marks according to a modification of the second embodiment.

FIG. 11 is a diagram showing a positional relationship between a mark and a light receiving field when a small mark is arranged.

FIG. 12 is a configuration diagram of marks according to still another modification of the second embodiment.

FIG. 13 is a configuration diagram of marks according to still another modification of the second embodiment.

FIG. 14 is a diagram showing a positional relationship between a mark and a light receiving visual field when a small mark is arranged.

FIG. 15 is a diagram showing marks according to still another modification of the second embodiment.

FIG. 16 is a diagram showing a mark according to still another modification of the second embodiment, (a) is a diagram showing a state of a parking space when a vehicle is present, and (b) is a vehicle. It is a figure which shows the appearance of the parking space when is present.

FIG. 17 is a sectional view of a mark structure according to still another modification of the second embodiment.

FIG. 18 is a block diagram of a parking management system according to a third embodiment.

FIG. 19 is a perspective view of a conventional parking detection device, (a)
FIG. 4 is a diagram showing a state of a parking space when a vehicle is not present, and FIG. 7B is a diagram showing a state of a parking space when a vehicle is present.

FIG. 20 is a diagram showing an image around the mark when the mark is a bright color and the ground is a dark color, and an image of the mark detected by the TV camera.

FIG. 21 is a diagram showing an area around the mark when the mark is dirty or scratched and an image of the mark detected by the TV camera.

FIG. 22 is a diagram showing an image around a mark when a shadow of an adjacent vehicle approaches the mark and an image of the mark detected by the TV camera.

[Explanation of symbols]

 1 parking detection sensor (object detection device) 11 mark (background object) 13a, 13b image detector (light receiving means) 15a, 15b image (image) 18 distance calculation unit (distance measuring unit) 19 comparison unit (determination unit) 25 parking Space (road surface of parking place) 26 Vehicle (object to be detected) 50 Parking management system

Claims (4)

[Claims] 1. A light receiving means for receiving light from an object to be detected or a background object existing within a light receiving field and outputting a light receiving signal based on the amount of light received, the distance between the two light receiving means being set. Distance measuring means for measuring the distance between the light receiving means and the detected object or the background object based on the triangulation method with the base line length, and the background object based on the distance measurement output from the distance measuring means. An object detection device, comprising: a determination unit that determines whether or not the detected object is present between the light receiving unit and the light receiving unit. 2. The object detecting device is a passive type detecting device for detecting the presence or absence of the detected object by receiving light from the detected object or a background object, and the light received by the light receiving means. The object detection device according to claim 1, wherein distance measurement is performed based on a shift amount of two images. 3. The object detection device according to claim 1, wherein the background object is a mark installed on a road surface of a parking space. 4. The object detection device according to claim 3, comprising:
A parking lot management system, which detects whether or not a parked vehicle is present in a parking space based on a discrimination output by the device.