JP2000207678A - Vehicle detector - Google Patents

Abstract

PROBLEM TO BE SOLVED: To detect the approach and passing of a vehicle without marking the surface of a road by picking up images within an almost common prescribed range watched from plural positions separated at a prescribed interval and detecting the presence/absence of the vehicle on the basis of a parallax. SOLUTION: Line sensors 1a and 1b are arranged separately at an interval and respective visual fields 3a and 3b are adjusted so as to pickup images within almost the common image pickup range. When the vehicle passes inside the visual fields of the line sensors 1a and 1b, an image density is changed and further, a distance between the line sensors 1a and 1b and an image pickup object is changed by the height of the vehicle (image pickup object) so that the position of the edge is deviated (parallax is generated). Therefore, by identifying whether this parallax is more than a prescribed parallax corresponding to the height of the vehicle, it can be detected that the image pickup object is the vehicle. Further, even when the vehicle is detected, the height of the vehicle can be detected from the level of the parallax as well.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a vehicle detecting device for detecting entry or passage of a vehicle traveling on a road.

[0002]

2. Description of the Related Art In a tollgate or a parking lot on a toll road, when automating the operation of issuing a toll ticket or a parking ticket, it is necessary to automatically detect a vehicle entering or passing. For this purpose, a vehicle detection device as disclosed in Japanese Patent Application Laid-Open No. 07-244795 has been conventionally proposed.

In this conventional vehicle detection device, as shown in FIG. 10, the condition of a periodic mark 18 drawn on a road surface is changed by line sensors 1a, 1b, 1c captures an image. Now, when there is no vehicle, the images captured by the line sensors 1a to 1c are as shown in FIG.
As shown in FIG. 1 (a), only the signal of the mark 18 exists.

Here, when the vehicle 21 enters the field of view of the line sensors 1a to 1c, the picked-up image is displayed as shown in FIG.
As shown in (b), the vehicle 2 is located at a location corresponding to the vehicle 21.
A change in the image density corresponding to the luminance of 1 appears.
When the vehicle 21 moves out of the field of view of the line sensors 1a to 1c due to the traveling of the vehicle 21, the captured image returns to the original image including only the mark 18. As described above, the entry and the passage of the vehicle can be detected based on the density change of the captured images of the line sensors 1a to 1c.

[0005]

However, in the vehicle detection device according to the prior art, it is necessary to draw the mark 18 on the road surface. Therefore, when the mark 18 is worn due to the passage of the vehicle 21, the mark 18 needs to be repaired. Further, when the soil 18 covers the mark 18, it is necessary to remove the mark 18. As described above, maintaining the display of the mark 18 requires a great deal of labor and cost.

Accordingly, it is an object of the present invention to provide a vehicle detection device capable of detecting entry and passage of a vehicle without drawing a mark on the road surface 18.

[0007]

According to a first aspect of the present invention, there is provided a vehicle detection apparatus for capturing an image of a substantially common predetermined range viewed from a plurality of positions separated by a predetermined distance. It is characterized by comprising imaging means for outputting an image signal, parallax detecting means for detecting parallax of the image signal, and detecting means for detecting the presence or absence of a vehicle based on the parallax.

[0008] In the vehicle detecting device having the first configuration, the imaging means includes, for example, a plurality of imaging elements arranged at a plurality of positions separated by a predetermined interval, and each of which picks up an image in a predetermined range. Further, the imaging means includes, for example, a plurality of mirrors respectively arranged at a plurality of positions separated by a predetermined interval, and an image sensor for imaging an image in a predetermined range reflected by the plurality of mirrors.

Further, the vehicle detecting device of the first configuration is further provided with height detecting means for detecting the height of the vehicle based on parallax.

According to a second aspect of the present invention, there is provided a vehicle detecting apparatus for capturing an image of a predetermined range and outputting an image signal, and an edge detecting an edge gradient of the image signal. It is characterized by comprising a gradient detecting means and a detecting means for detecting the presence or absence of a vehicle based on the gradient.

Further, the vehicle detecting device of the second configuration is further provided with height detecting means for detecting the height of the vehicle based on the gradient.

[0012]

Embodiments of the present invention will be described below. However, the technical scope of the present invention is not limited to the present embodiment.

FIG. 1 is a configuration diagram of a vehicle detection device according to a first embodiment of the present invention. In the vehicle detection device shown in FIG. 1, a stereo line sensor 1 including a line sensor 1a and a line sensor 1b, which are image pickup devices, is installed above a road and images vehicles 21 and 22 passing below. The line sensor 1a and the line sensor 1b are arranged at an interval as shown in FIG.
a and 3b are adjusted so as to capture a substantially common imaging range from the respective positions where the line sensors 1a and 1b are arranged.

As described above, the two line sensors 1a, 1
b, by capturing an image of a substantially common range from different positions, as described in detail below, based on the principle of triangulation, the captured object and the stereo line sensor 1 are used.
Can be detected.

That is, the line sensor 1a and the line sensor 1
b, if the distance between the stereo line sensor 1 and the object to be imaged is short, the line sensors 1a and 1b
Causes a large deviation in the imaging range of each captured image, and when the distance to the imaging target is long, the deviation becomes small. This shift is called parallax.

More specifically, when the line sensors 1a and 1b are capturing an image of a road surface on which a vehicle is not passing, the line sensors 1a and 1b capture an image of a substantially common imaging range. Deviation (parallax) of the position of the part (edge) where the image density changes due to the difference in luminance of the object
Is slight or rare.

Here, when the vehicle passes within the field of view of the line sensors 1a and 1b, a change occurs in the image density.
Depending on the height of the vehicle (the object to be imaged), the line sensor 1a,
Since the distance between 1b and the imaging target changes, a displacement (parallax) occurs at the edge position. Therefore, this parallax is
By identifying whether or not the parallax is equal to or greater than a predetermined parallax corresponding to the height of the vehicle, it is possible to detect that the imaging target is a vehicle.

Further, even when a vehicle is detected, the height of the vehicle can be detected from the magnitude of the parallax. That is, as the height of the vehicle increases, the distance between the stereo line sensor 1 and the vehicle decreases, and the parallax also increases. Therefore, relatively low vehicles such as ordinary passenger cars,
When relatively high vehicles such as buses and trucks pass, the magnitude of parallax differs. Therefore, it is possible to determine the type of the vehicle based on the degree of the magnitude of the parallax.

FIGS. 2A and 2B are diagrams showing examples of the density change of the picked-up images of the line sensors 1a and 1b, respectively. As shown in FIGS. 2A and 2B, when the deviation (parallax) of the position of the density change portion (edge) at the imaging position in the captured image is relatively small, a low vehicle (such as an ordinary passenger car) 21 is used. If the parallax is relatively large,
It is determined that the vehicle is expensive (eg, a bus or a truck) 22.

FIG. 3 is a functional block diagram of the vehicle detection device according to the first embodiment. In FIG. 3, analog image signals S1a and S1b output from two line sensors 1a and 1b in the stereo line sensor 1 are A
/ D converters 4a and 4b input to the digital image signal S
4a and s4b.

The reference numerals 5 to 1 described below
Each unit of 0 is realized by individual hardware, or each function is realized by an arithmetic unit such as a CPU executing a predetermined program stored in a memory such as a ROM.

The digital image signals S4a and S4b of the line sensors 1a and 1b output from the A / D converter 4 are
Edge detection means 5a and 5b and density detection means 6
a and 6b. The edge detection means 5a, 5b
A position (edge) where the density (luminance) of the digital image signal changes is detected based on a difference between the luminance data of adjacent pixels, and the density detecting means 6a and 6b determine the density (luminance) of the digital image signal S2. To detect.

The edge signals S5a and S5b output from the edge detecting means 5a and 5b respectively and the density detecting means 6
density signals S6a, S6 respectively output from a, 6b
b is input to the corresponding edge detection means 7.

The corresponding edge detecting means 7 detects at least one of the edge signals S5a and S5b
The densities at both ends of the two edge data are compared based on the density data included in the density signals S6a and S6b. Then, edges having the same density change at both ends are detected as the same edge in the visual field, and the edges are associated with each other.

The corresponding edge signal S7 output from the corresponding edge detecting means 7 is input to the parallax detecting means 8. The parallax detection means 8 detects an interval between edges (parallax) based on at least one set of edge data included in the corresponding edge signal S7.

The parallax signal S output from the parallax detecting means 8
8 is input to the height measuring means 9. Height measuring means 9
Measures the height of the vehicle using the disparity data included in the disparity signal S8, and information such as the interval and the inclination between the line sensors 1a and 1b.

The height signal S output from the height measuring means 9
9 is input to the vehicle detection means 10. Vehicle detection means 1
In the case of 0, the minimum height for judging a vehicle is set in advance as a threshold value. , Judge as the road surface.

Further, by obtaining a predetermined parallax for a vehicle having a predetermined height in advance, the vehicle can be detected by comparing the predetermined parallax with the detected parallax without measuring the height. It is. In this case, the parallax signal S8 is directly input to the vehicle detection means 10 (not shown).

In FIG. 2, only the edge at the upper end of the vehicle is displayed. However, the edge generated by the lower end, the window, the pattern, the dirt, etc. can be measured. , That is, the vehicle area and the road area.

When it is not possible to image the entire lane in the field of view of one stereo line sensor 1, FIG.
As shown in (1), the entire lane may be imaged by a plurality of stereo line sensors 1. In this case, the parallax and the height of the edge are detected by the parallax detecting means 8 or the height measuring means 9 provided for each stereo line sensor, and the parallax data or the height data is input to the vehicle detecting means 10. The vehicle detection unit 10 integrates a plurality of pieces of parallax data or height data corresponding to the respective stereo line sensors 1 and detects a vehicle in the lane.

When the vehicle detection means 10 detects a vehicle, the vehicle detection means 10 transmits a vehicle detection signal S10 to the host system 11. The host system 11, for example,
It is a pass ticket issuing device or a parking ticket issuing device, and issues a pass ticket or a parking ticket based on the vehicle detection signal S10.

According to the first embodiment, the vehicle can be detected by detecting the parallax without providing a mark on the road surface, and the height of the vehicle can be detected. Therefore, the maintenance and inspection work of the mark on the road surface becomes unnecessary, and the cost and labor for the maintenance and inspection are reduced.

FIG. 4 is a configuration diagram of a vehicle detection device according to a second embodiment of the present invention. According to FIG. 4, the vehicle detection device according to the second embodiment has a single line sensor 1a and a pair of mirrors 13a instead of the stereo line sensor 1 incorporating the two line sensors in FIG. , 13b and a triangular mirror 14. The line sensor 1a is installed above a road and captures images of vehicles 21 and 22 passing below. The triangular mirror 14 is arranged so that its vertex is located on the optical axis of the line sensor 1a, and divides the field of view of the line sensor 1a into two. The divided left and right fields of view 3a, 3b are mirrors 13a, 13b, respectively.
The light is reflected again at b and adjusted so as to capture an image of a substantially common imaging range on the road surface. Therefore, the images captured in the visual fields 3a and 3b are combined by the triangular mirror 14 via the mirrors 13a and 13b, and are simultaneously captured by the line sensor 1a.

Since the mirrors 13a and 13b are arranged at an interval on the left and right, an image of a substantially common range is taken from different positions in the same manner as in the first embodiment, thereby achieving the principle of triangulation. Based on this, the parallax can be detected.

FIG. 5A is a diagram showing a change in density of a picked-up image picked up by the line sensor 1a. FIG.
In (a), the left side of the dotted line in the figure is the captured image corresponding to the visual field 3a (FIG. 5 (b)), and the right side of the dotted line is the captured image corresponding to the visual field 3b (FIG. 5 (c)). It is.

Therefore, as in the first embodiment,
The vehicle can be detected based on the magnitude of the parallax, which is the difference between the positions of the edges in FIGS. 5B and 5C. FIG. 6 is a functional block diagram of the vehicle detection device according to the second embodiment. In FIG. 6, the line sensor 1
The analog image signal S1 output from a is input to the A / D converter 4 and converted into a digital image signal S4.

The digital image signal S4 output from the A / D converter 4 is input to the edge detecting means 5 and the density detecting means 6. In the second embodiment, only one line sensor 1a is used to detect parallax. Therefore, unlike the first embodiment, a plurality of edge detecting means 5 and density detecting means 6 are not required. .

Then, as in the first embodiment,
The edge detecting means 5 detects a position (edge) where the density (luminance) of the digital image signal changes based on the difference between the luminance data of the adjacent pixels, and the density detecting means 6 detects the density (density) of the digital image signal. Luminance).

The edge signal S5 output from the edge detecting means 5 and the density signal S6 output from the density detecting means 6
Is input to the left / right dividing means 15. Right and left division means 15
Divides the input edge signal S5 and the density signal S6 at the position of the vertex of the triangular mirror 14, and outputs the right edge signal S5
R, the left edge signal S5L, the right density signal S6R, and the left density signal S6L are output to the corresponding edge detection means 7. Note that the left and right dividing means 15 may be configured as individual hardware, similarly to the above-described means, or may have a corresponding function according to a predetermined program stored in a memory such as a ROM by an arithmetic device such as a CPU. Can be realized.

The operation after the corresponding edge detecting means 7, that is, the function from the corresponding edge detecting means 7 to the host system 12 is the same as that of the first embodiment, and the description is omitted.

When all the lanes do not fall within the field of view with one imaging means, as in the first embodiment, a plurality of imaging means are used to detect the vehicle over all the lanes. Will be installed.

According to the second embodiment, similarly to the first embodiment, the vehicle can be detected by detecting the parallax without providing a mark on the road surface. The height can also be detected.

Further, in the second embodiment, in order to detect parallax, imaging means for imaging a substantially common imaging range from different positions is constituted by one line sensor. Therefore, compared to the first embodiment using a plurality of line sensors, there is no density difference due to the sensitivity difference between the plurality of line sensors, so that the reliability of the corresponding edge detection by the corresponding edge detecting means 7 is improved. . In addition, by taking an image with one line sensor, a part of a processing circuit such as a synchronization circuit for synchronizing a plurality of line sensors and an A / D converter can be reduced, thereby reducing costs. Can be achieved.

FIG. 7 is a configuration diagram of a vehicle detection device according to a third embodiment of the present invention. According to FIG. 7, the vehicle detection device according to the third embodiment includes one line sensor 1a and an attenuation filter 16 as imaging means. The line sensor 1a is installed above a road and captures images of vehicles 21 and 22 passing below. The attenuation filter 16 is arranged on the front surface of the lens of the line sensor 1a.

In the third embodiment, as will be described in detail below, the distance between the line sensor 1a and the imaged object is changed using the depth of field of the lens of the line sensor 1a. Is detected.

In general, at a distance outside the focal length of the lens, the image becomes out of focus and the edge of the image becomes dull. Then, as the aperture of the lens is opened and the depth of field becomes shallower, even a slight shift from the focal length causes image blur, and the degree of blur corresponding to the shift from the focal length increases.

Therefore, in the third embodiment, first, the focal length of the line sensor 1a is adjusted to the distance to the road surface,
Open the lens aperture. Since the aperture of the lens is opened, the adjustment of the incident light is performed by adjusting the attenuation rate of the attenuation filter 16 arranged on the front surface of the lens.

When the vehicle does not pass through the field of view of the line sensor 1a, since the line sensor 1a captures an image of the road surface, the edge gradient of the image is steep. Here, when the vehicle passes within the field of view of the line sensor 1a, a change occurs in the image density, and the gradient of the edge becomes dull. Then, by identifying whether the gradient of this edge is equal to or greater than a predetermined gradient corresponding to the height of the vehicle,
The vehicle can be detected.

Further, even when a vehicle is detected, the height of the vehicle can be detected from the magnitude of the gradient. That is, as the height of the vehicle increases, the line sensor 1
Since the distance between a and the vehicle is short, the gradient of the edge becomes dull. Therefore, relatively low vehicles such as ordinary passenger cars,
When passing a relatively high vehicle such as a bus or a truck, the gradient of the edge is different. Therefore, it is also possible to determine the type of the vehicle based on the magnitude of the gradient.

FIG. 8 is a diagram showing an example of a change in density of an image picked up by the line sensor 1a. As shown in FIG. 8, when the gradient of the density change portion (edge) is relatively small at the imaging position in the captured image, the vehicle is determined to be a low vehicle (such as a normal passenger car) 21 and the gradient of the edge is relatively large. In this case, it is determined that the tall vehicle (bus, truck, etc.) 22.

FIG. 9 is a functional block diagram of the vehicle detection device according to the third embodiment. 9, an analog image signal S1 output from a line sensor 1a is input to an A / D converter 4 and converted into a digital image signal S4. The digital image signal S2 is input to the edge detecting means 5 as in the second embodiment. Then, the edge is detected by the edge detecting means 5. The data of the detected edge is input to the edge gradient determining means 17 as an edge signal S5. The edge gradient determining means 17 determines an edge gradient from the edge strength (density difference between adjacent pixels) and its continuity. The edge gradient determining means 17 may be configured as individual hardware, similarly to the above-described means, or may be operated by an arithmetic device such as a CPU according to a predetermined program stored in a memory such as a ROM. Can be realized.

The edge gradient signal S17 output from the edge gradient means 17 is input to the height measuring means 9. The height measuring means 9 calculates the height of the imaging target from the edge gradient data included in the edge gradient signal S17, that is, the blur amount.

The height signal S output from the height measuring means 9
9 is input to the vehicle detection means 10. Vehicle detection means 1
In the case of 0, the minimum height determined as a vehicle is set in advance as a threshold, and when the input height data is equal to or greater than the threshold, it is determined that the vehicle is a vehicle, and when the height data is less than the threshold Is determined to be a road surface.

Further, by determining a predetermined edge gradient for a vehicle having a predetermined height in advance, the vehicle is detected by comparing the determined edge gradient with the determined edge gradient without calculating the height. It is possible. In this case, the edge gradient signal S17 is directly input to the vehicle detection means 10 (not shown).

When the vehicle detection means 10 detects a vehicle, the vehicle detection means 10 transmits a vehicle detection signal S10 to the host system 11. The host system 11, for example,
It is a pass ticket issuing device or a parking ticket issuing device, and issues a pass ticket or a parking ticket based on the vehicle detection signal S10.

When all the lanes do not fall within the field of view with one image pickup means, a plurality of image pickup means are provided to perform vehicle detection over all the lanes as in the first embodiment. Will be installed.

According to the third embodiment, the vehicle can be detected by detecting the edge gradient of the image without providing a mark on the road surface, and the height of the vehicle can be detected. Becomes

Further, in the third embodiment, 1
Vehicle detection is possible with only one line sensor,
As in the second embodiment, the cost can be reduced as compared with the first embodiment using a plurality of line sensors.

In each of the above embodiments, the analog image signal output from the line sensor is converted into a digital signal, and vehicle detection is performed by digital processing. Vehicle detection may be performed. For example, the edge and its gradient may be detected by a known differentiation circuit, and vehicle detection may be performed by a known comparison circuit.

[0060]

As described above, according to the present invention, an image pickup means for picking up an image of a substantially common predetermined range viewed from a plurality of positions separated by a predetermined distance and outputting an image signal, and detecting a parallax of the image signal. A vehicle detection device is provided that includes a parallax detection unit and a detection unit that detects the presence or absence of a vehicle based on the parallax.

Further, according to the present invention, an image pickup means for picking up an image in a predetermined range and outputting an image signal, an edge gradient detecting means for detecting a gradient of an edge of the image signal, and the presence or absence of a vehicle based on the gradient And a detecting unit for detecting the vehicle detection.

As a result, the vehicle can be detected without providing a mark on the road surface, so that the labor and cost for maintenance and inspection of the mark, which were conventionally required, can be reduced.

In the above vehicle detecting device, there is provided a vehicle detecting device further comprising height detecting means for detecting the height of the vehicle detected by the detecting means based on the parallax or gradient. This makes it possible to determine the type of the detected vehicle.

[Brief description of the drawings]

FIG. 1 is a configuration diagram of a vehicle detection device according to a first embodiment.

FIG. 2 shows a line sensor 1a according to the first embodiment;
FIG. 5 is a diagram illustrating an example of a density change of a captured image according to 1b.

FIG. 3 is a functional block diagram of the vehicle detection device according to the first embodiment.

FIG. 4 is a configuration diagram of a vehicle detection device according to a second embodiment.

FIG. 5 is a diagram illustrating an example of a density change of a captured image by a line sensor 1a according to a second embodiment.

FIG. 6 is a functional block diagram of a vehicle detection device according to a second embodiment.

FIG. 7 is a configuration diagram of a vehicle detection device according to a third embodiment.

FIG. 8 is a diagram illustrating an example of a density change of a captured image by a line sensor 1a according to a third embodiment.

FIG. 9 is a functional block diagram of a vehicle detection device according to a third embodiment.

FIG. 10 is a configuration diagram of a conventional vehicle detection device.

FIG. 11 is a diagram illustrating a change in density of an image captured by a line sensor in a conventional vehicle detection device.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Stereo line sensor 1a Line sensor 1b Line sensor 4 A / D converter 5 Edge detecting means 6 Density detecting means 7 Corresponding edge detecting means 8 Parallax detecting means 9 Height measuring means 10 Vehicle detecting means 11 Upper system 15 Left and right dividing means 16 Attenuation filter 17 Edge gradient judgment means

Claims (6)

[Claims] An image pickup means for picking up an image in a substantially common predetermined range viewed from a plurality of positions separated by a predetermined distance and outputting an image signal; a parallax detection means for detecting a parallax of the image signal; Detecting means for detecting the presence or absence of the vehicle based on the vehicle detection device. 2. A vehicle detection system according to claim 1, wherein said image pickup means includes a plurality of image pickup devices respectively arranged at said plurality of positions separated by said predetermined distance and for picking up images of said predetermined range. apparatus. 3. The imaging device according to claim 1, wherein the imaging unit is configured to capture a plurality of mirrors respectively arranged at the plurality of positions separated by the predetermined interval, and to capture an image of the predetermined range reflected by the plurality of mirrors. A vehicle detection device comprising an element. 4. The vehicle detection device according to claim 1, further comprising a height detection unit that detects a height of the vehicle based on the parallax. 5. An image pickup means for picking up an image in a predetermined range and outputting an image signal, an edge gradient detecting means for detecting a gradient of an edge of the image signal, and detecting presence or absence of a vehicle based on the gradient. A vehicle detection device comprising: a detection unit. 6. The vehicle detecting device according to claim 5, further comprising height detecting means for detecting a height of the vehicle based on the gradient.