JP7439314B2 - Vehicle type identification device and vehicle type identification method - Google Patents

Description

Embodiments of the present invention relate to a vehicle type discrimination device and a vehicle type discrimination method.

Toll road tolls are determined by vehicle type. Therefore, a vehicle type discrimination device is installed at a toll gate of a toll road, and tolls are collected according to the result of vehicle type discrimination. In the current vehicle type determination, the vehicle type is determined by detecting various information using various devices as described in (1) to (5) below. (1) A vehicle detector detects vehicles and separates the vehicles. (2) Vehicle number information is photographed using a number photographing camera. (3) Measure the number of axles using an axle detector. (4) Measure the vehicle length using a vehicle length meter. (5) Measure the vehicle height using a vehicle height meter.

Various methods are used to detect various types of information, and the following methods are typical ones. The vehicle detector, vehicle height meter, and vehicle length meter use transmission-type sensors that use infrared light (vehicle detectors are line sensors), and light emitting/light receiving sensors are placed on the roadside on both sides of the lane across the vehicle. be done. The axle detector uses a pressure sensor buried in the roadway to detect the pressure exerted by the axle (tire) of the vehicle. Additionally, the license plate camera is placed on the roadside so that it can capture a panoramic view of the vehicle. The vehicle number is detected by image processing from the image obtained by the number image capturing camera.

ETC (Electronic Toll Collection System), which is widely used on toll roads, communicates between an on-board wireless communication device installed in a vehicle (hereinafter referred to as an on-vehicle device) and a roadside wireless device installed on the roadside. In some cases, vehicle type information recorded on the on-vehicle device is received, and a toll fee corresponding to the received vehicle type information is collected. However, for vehicles equipped with a towing device, the vehicle type (toll) may change depending on whether or not there is a towing vehicle. Yes, tolls are being collected.
As described above, the vehicle type identification device functions as an important device for toll collection on toll roads.

Japanese Patent Application Publication No. 2016-170508 Japanese Patent Application Publication No. 8-233525 JP2003-281581A JP2008-129920A

In order to determine the vehicle type, it is necessary to detect various vehicle information. As described above, the axle detector uses a pressure sensor buried in the roadway to detect the pressure applied by the axle (tire) of the vehicle. Therefore, the measurement of the number of axles of one vehicle cannot be completed unless all the axles of one vehicle have completed passing. In the case of large vehicles, the length of the vehicle is 12m or more, so axle measurement cannot be completed until the vehicle has traveled more than 12m, so the lane length at the toll plaza must be at least 12m or more. Particularly at small tollgates in urban areas, it is difficult to secure tollgate spaces that meet these conditions, and improvements in vehicle type identification devices are desired.

For example, a road operator that can secure toll plaza space (lane length) that satisfies the above conditions can determine the type of vehicle (toll) before the vehicle completes passing through the toll plaza. However, at narrow tollgates where the tollgate space (lane length) that meets the above conditions cannot be secured, it is not possible to determine the number of axles (vehicle type) before the vehicle passes, and ), the vehicle type (toll) will be determined in the backward processing. In the case of deferred payment collection using credit, such as the ETC system, back-order processing becomes possible. However, if cash is used, the number of axles (vehicle type) must be visually determined by a collector or the like, leading to concerns about miscalculation and increased service time.

Furthermore, the number of vehicle classifications may increase due to various circumstances. For example, when attempting to integrate vehicle classifications by seamlessly connecting multiple road operators, a road operator with fewer vehicle classifications may match the vehicle classifications of a road operator with many vehicle classifications, and a road operator with fewer vehicle classifications may Viewed from the side, the number of vehicle types increases. An increase in the number of vehicle classifications will lead to an increase in the burden of visually identifying vehicle types on collection staff, etc., and there is a need for technology that reduces the burden of identifying vehicle types.

An object of the present invention is to provide a vehicle type discrimination device and a vehicle type discrimination method that are capable of discriminating vehicle types in a limited toll gate space.

The vehicle type discrimination device according to the embodiment includes a first vehicle detector, a second vehicle detector, an imaging section, a control section, and a discrimination section. The first vehicle detector is provided along a lane in which a vehicle travels, and detects a vehicle reaching a first point on the lane. The second vehicle detector is provided along the lane and detects the vehicle arriving at a second point on the lane downstream of the first point and upstream of the launch controller. The photographing unit is provided at a position closer to the first vehicle detector than the second vehicle detector, and photographs a video of a predetermined range of the lane including the first and second points using a wide-angle lens. . The control unit sets the photographing timing of the photographing unit to start when the vehicle is detected by the first vehicle detector, and when the vehicle is no longer detected by the first vehicle detector. The video is controlled to be shot during the period ending with . The determining unit repeatedly measures the number of axles and vehicle length of the vehicle over the period from the video shot by the shooting unit with the wide-angle lens at the shooting timing controlled by the control unit, and Based on the result of the number of axles and the vehicle length, the type of vehicle is determined.

FIG. 1 is a block diagram showing an example of a schematic configuration of an ETC system according to first and second embodiments. FIG. 2 is a block diagram showing an example of a schematic configuration of lane equipment of the ETC system according to the first and second embodiments. FIG. 2 is a top view and a side view showing an example of installing lane equipment in a lane of the ETC system according to the first and second embodiments. 7 is a flowchart illustrating an example of vehicle type determination processing based on a moving image according to the first embodiment. 5 is a timing chart showing an example of the timing of vehicle detection by a first vehicle detector. FIG. 3 is a diagram illustrating an example of a transition in the positional relationship between the first and second vehicle detectors and a large vehicle that moves by running. FIG. 6 is a diagram illustrating an example of a transition in the positional relationship between the first and second vehicle detectors and a moving ordinary vehicle. 12 is a flowchart illustrating an example of vehicle type determination processing based on still images according to the second embodiment. 5 is a timing chart showing an example of the timing of vehicle detection by the first and second vehicle detectors.

Hereinafter, an example of the ETC system according to the first and second embodiments will be described with reference to the drawings.
FIG. 1 is a block diagram showing an example of a schematic configuration of an ETC system according to the first and second embodiments.
The ETC system is a system that controls vehicle traffic and collects tolls from passing vehicles. For example, an ETC system collects tolls by wirelessly communicating with an on-vehicle device mounted on a vehicle passing through a lane at a toll gate on a toll road. In addition to collecting tolls, the ETC system also controls vehicle traffic by opening and closing bars installed in the lanes in which vehicles are passing, and by displaying guidance on roadside indicators installed on the sides of the lanes. For example, toll gates are provided at the entrances and exits of toll roads to which the ETC system is applied, and one or more lanes are provided at these entrances and exits.

As shown in FIG. 1, the ETC system includes a central device 1 at the center. Further, the ETC system includes a toll plaza server 2 and a lane server 3 at each toll plaza (toll plaza machine room). Further, the ETC system includes lane equipment 4 for each lane in which vehicles travel at the toll gate. The central device 1 is connected to the toll plaza server 2 of each toll plaza. The toll plaza server 2 of a predetermined toll plaza is connected to the lane server 3 of the same predetermined toll plaza, and the lane server 3 of the predetermined toll plaza is connected to lane equipment 4 provided in each lane of the same predetermined toll plaza.

The central device 1 exchanges various data with each toll plaza server 2, and the toll plaza server 2 exchanges various data with a lane server 3. The lane server 3 controls the lane equipment 4 based on sensor data and the like transmitted from the lane equipment 4.

FIG. 2 is a block diagram showing an example of a schematic configuration of lane equipment of the ETC system according to the first and second embodiments.

As shown in FIG. 2, the lane equipment 4 includes an interface aggregation unit 401, a first antenna 411, a second antenna 412, a recovery antenna 413, a first vehicle detector S1, a second vehicle detector S2, and a vehicle detector. S4, a camera equipped with a wide-angle lens (photographing unit) 421, an image processing device 422, a roadside display 423, an in-booth display 4241 in the booth 424, a cash processing terminal 4242 in the booth 424, a starting controller 425, and a lane monitoring camera 426.

FIG. 3 is a top view and a side view showing an example of installing lane equipment in a lane of the ETC system according to the first and second embodiments.
As shown in FIG. 3, islands 40 extending along the lane are arranged facing each other on both sides of the lane in which the vehicle travels. A vehicle travels on a lane between these two islands 40 from upstream to downstream. In addition, on the island 40, in order from upstream to downstream of the lane, a wide-angle lens equipped camera 421, an image processing device 422, a first vehicle detector S1, a second vehicle detector S2, a booth 424, and a roadside display device are installed. 423, a start control device 425, a start control bar 4251, a vehicle detector S4, and a lane monitoring camera 426 are provided.
Alternatively, the positions of the wide-angle lens equipped camera 421 and the first vehicle detector S1 may be reversed. That is, in order from upstream to downstream of the lane, the first vehicle detector S1, the wide-angle lens equipped camera 421 and the image processing device 422, the second vehicle detector S2, the booth 424, the roadside display 423, and the start controller. 425, a start control bar 4251, a vehicle detector S4, and a lane monitoring camera 426 may be provided.

Each part of the lane equipment will be described in detail with reference to FIGS. 2 and 3.
The two islands 40 are provided with a pair of first vehicle detector S1, second vehicle detector S2, and vehicle detector S4 in order from upstream to downstream. The first vehicle detector S1 detects a vehicle reaching a first point in the lane, and the second vehicle detector S2 detects a vehicle reaching a second point downstream from the first point. , the third vehicle detector S4 detects a vehicle arriving at a third point downstream from the second point.
These first vehicle detector S1, second vehicle detector S2, and vehicle detector S4 each include a light projecting section and a light receiving section, and the infrared light from the light projecting section is received by the light receiving section. During the period when the vehicle detection signal is OFF, the vehicle detection signal is output to the interface aggregation unit 401, and during the period when the infrared light from the light projecting unit is not received by the light receiving unit (the period when the infrared light is blocked by the vehicle), the vehicle detection signal is output. A signal ON is output to the interface aggregation unit 401.

The interface aggregation unit 401 aggregates information from each unit and outputs it to the lane server 3. Further, the interface aggregation unit 401 outputs information from the lane server 3 to each unit. For example, the interface aggregation unit 401 outputs a vehicle detection signal ON or OFF to the lane server 3. The lane server 3 detects the position (trend) of the vehicle on the lane based on the ON or OFF vehicle detection signals from the first vehicle detector S1, the second vehicle detector S2, and the vehicle detector S4, Controls the operation of lane equipment 4. As will be described in detail later, the lane server 3 controls the timing of photographing by the wide-angle lens equipped camera 421.

Moreover, three antenna installation gates are provided in the two islands 40 in order from upstream to downstream across the lanes. A first antenna 411 is provided at the first antenna installation gate. A recovery antenna 413 is provided at the second antenna installation gate. A second antenna 412 is provided at the second antenna installation gate. These first antenna 411, second antenna 412, and recovery antenna 413 communicate with the on-vehicle device of the vehicle, receive various information transmitted from the on-vehicle device, and transmit various information to the on-vehicle device.

In the lane equipment 4 of the entrance toll plaza, the first antenna 411, the second antenna 412, and the recovery antenna 413 transmit entrance information to the on-vehicle device. The onboard device records entrance information on the loaded ETC card. In addition, in the lane equipment 4 at the exit tollgate, the first antenna 411, the second antenna 412, and the recovery antenna 413 receive onboard device information, ETC card information, entrance information, etc. transmitted from the onboard device. , transmits on-vehicle device information, ETC card information, entrance information, etc. to the interface aggregation unit 401.

The interface aggregation unit 401 transmits onboard device information, ETC card information, entrance information, etc. to the lane server 3.
The lane server 3 transmits onboard device information, ETC card information, entrance information, etc. to the tollgate server 2.
The toll plaza server 2 stores exit information of the exit toll plaza where the toll plaza server 2 is installed, and also stores toll tables for each section and vehicle type. The toll gate server 2 refers to the toll table, calculates the toll based on the onboard device information, ETC card information, entrance information, exit information, and vehicle type information obtained by vehicle type discrimination, and sends the toll to the central device 1 and lane server 3. Send toll fees, vehicle type information, etc.
The lane server 3 transmits tolls, vehicle type information, etc. to the interface aggregation unit 401.

The interface aggregation unit 401 transmits tolls, vehicle type information, etc. to the first antenna 411, the second antenna 412, and the recovery antenna 413. The first antenna 411, the second antenna 412, and the recovery antenna 413 transmit tolls, vehicle type information, etc. to the on-vehicle device.
The onboard device 71 stores tolls, vehicle type information, and the like. The vehicle type information is information indicating the vehicle type, such as a small car, a regular car, or a large car. The lane server 3 receives images of vehicles (including axles) through the interface aggregation unit 401, measures and detects the number of axles and vehicle length from these images, determines the vehicle type from the number of axles and vehicle length, and determines the vehicle type. Generate discrimination results (vehicle type information).
Note that vehicle type discrimination will be explained in detail later.

The wide-angle lens equipped camera 421 is provided near the first vehicle detector S1, and closer to the installation position of the first vehicle detector S1 than the installation position of the second vehicle detector S2. The wide-angle lens-equipped camera 421 is provided within a predetermined length (for example, 2 m) along the direction in which the lane extends, based on the installation position of the first vehicle detector.
In FIG. 3, the wide-angle lens equipped camera 421 is installed on the upstream side near the first vehicle detector S1, but the lane monitoring camera 426 is installed on the downstream side near the first vehicle detector S1. You can also do this.
In addition, the height of the wide-angle lens-equipped camera 421 from the road surface is desirably high enough to photograph the axle (tire) in front, while taking into account mud splashing from the vehicle (for example, about 30 cm to 100 cm from the road surface). Due to the characteristics of a wide-angle lens, the height above the road surface can be set freely.

The wide-angle lens-equipped camera 421 is equipped with a wide-angle lens (lens with a short focal length) that has a wider angle of view than a standard lens, and photographs a predetermined range of the lane.
The predetermined range of the lane to be photographed includes a first point where the arrival of a vehicle is detected by the first vehicle detector S1, and a second point where the arrival of the vehicle is detected by the second vehicle detector S2. including. For example, the wide-angle lens-equipped camera 421 photographs a predetermined range of about 15 m (total of about 30 m) in the upstream and downstream directions of the lane, with the installation position of the wide-angle lens-equipped camera 421 as a reference. It is desirable for a wide-angle lens to be able to capture the entire side of the vehicle of various types of vehicles (light cars, regular cars, medium-sized cars, large cars, extra-large cars, etc.) driving in the lane, and depending on the installation environment, it can capture a range of about 180°. A lens that can take pictures (fisheye lens) is desirable.
Further, the wide-angle lens equipped camera 421 may be a camera that shoots moving images, a camera that shoots still images, or a camera that shoots both moving images and still images.

If the wide-angle lens equipped camera 421 is a camera that shoots a video, it starts shooting based on the first shooting instruction (shooting start instruction) from the lane server 3, and starts shooting based on the second shooting instruction (shooting end instruction). Then, video shooting is continued from the shooting start instruction to the shooting end instruction, and the captured motion is output. If the wide-angle lens-equipped camera 421 is a camera that takes still images, it takes a still image based on a shooting instruction from the lane server 3 and outputs the still image.

Note that vehicle type discrimination processing in which a video is captured by the camera 421 equipped with a wide-angle lens and the vehicle type is determined based on the video will be described as a first embodiment. Furthermore, vehicle type discrimination processing in which a still image is photographed by the wide-angle lens equipped camera 421 and the vehicle type is discriminated based on the still image will be described in detail later as a second embodiment.

The image processing device 422 detects an image of a vehicle (an image of the side of the vehicle) from a moving image or a still image from the wide-angle lens equipped camera 421, and outputs the detected image of the vehicle (including the axle). The vehicle image is transmitted to the lane server 3 via the interface aggregation unit 401. As described above, the lane server 3 detects the axle image from the vehicle image, measures the number of axles, measures the vehicle length from the vehicle image, and determines the vehicle type from the axle detection result and vehicle length detection result. , generates and outputs vehicle type discrimination results (vehicle type information).

The roadside display 423 displays tolls, vehicle type information, etc. output from the interface aggregation unit 401 to the driver of the vehicle.
An in-booth display 4241 is provided in the booth 424 where the toll collector waits, and the in-booth display 4241 also displays tolls, vehicle type information, etc. output from the interface aggregation unit 401. Further, the booth 424 is provided with a cash processing terminal 4242, and the toll collector can receive cash from a user who wishes to pay in cash and execute the payment processing at the cash processing terminal 4242.

The launch controller 425 controls opening and closing of the launch control bar 4251 based on launch control information from the interface aggregation unit 401. The start control bar 4251 physically prevents vehicles from passing (leaving the lane) when it is closed, and allows vehicles to pass when it is open.
The lane monitoring camera 426 photographs a vehicle when it enters or exits a lane.

(First embodiment: Vehicle type discrimination processing based on video)
Next, the first embodiment will be described with reference to FIGS. 4 to 7.
The first embodiment is a vehicle type determination process that determines the vehicle type based on a video captured by a camera 421 equipped with a wide-angle lens. In this embodiment, the lane server 3 controls the photographing timing of the wide-angle lens equipped camera 421 based on the vehicle detection situation by the first vehicle detector S1, and identifies the vehicle type of the vehicle included in the image acquired at this photographing timing. Discern.

FIG. 5 is a timing chart showing an example of the timing of vehicle detection by the first vehicle detector S1.
The timing chart shown in the upper part of FIG. An example of an output period of a vehicle detection signal ON for detecting a first vehicle type is shown. The timing chart shown in the lower part of FIG. 5 is for a regular car (second vehicle type) in which the first vehicle detector S1 is shorter than the distance between the first vehicle detector S1 and the second vehicle detector S2. An example of the output period of the vehicle detection signal ON for detecting is shown.
FIG. 6 is a diagram showing an example of a transition in the positional relationship between the first vehicle detector S1, the second vehicle detector S2, and a large vehicle that moves by running.
FIG. 7 is a diagram illustrating an example of a transition in the positional relationship between the first vehicle detector S1, the second vehicle detector S2, and a regular vehicle that moves by running.

FIG. 4 is a flowchart illustrating an example of vehicle type discrimination processing based on a moving image according to the first embodiment.
As shown in FIG. 4, the lane server 3 monitors the vehicle detection signal from the interface aggregation unit 401 and enters a standby state for photographing a vehicle (step ST101).

A vehicle such as a large vehicle or a regular vehicle travels from the upstream lane to the downstream direction of the toll booth lane. When a vehicle blocks infrared light, etc. between the pair of first vehicle detectors S1 arranged oppositely on the two islands 40, the vehicle detection signal of the first vehicle detector S1 is switched from OFF to vehicle detection signal ON. (Step ST102, YES) (timing t1 in the timing chart of FIG. 5) (step ST6-1 in FIG. 6 showing an example of a large car) (step ST7-1 in FIG. 7 showing an example of a regular car).
When the lane server 3 detects that the vehicle detection signal from the first vehicle detector S1 is turned on, it instructs to start photographing. That is, the lane server 3 uses the vehicle detection signal ON from the first vehicle detector S1 as a trigger to instruct the start of photographing.
The wide-angle lens equipped camera 421 starts shooting a moving image based on the instruction to start shooting (step ST103).

The image processing device 422 detects an image of the vehicle (including the axle) from the video output from the wide-angle lens equipped camera 421, and outputs the image of the vehicle (including the axle).
The lane server 3 detects axles from the detected image, measures the number of axles (step ST104), and primarily records the result of measuring the number of axles (step ST105).
Furthermore, the lane server 3 measures the vehicle length from the image of the detected vehicle (step ST106), and primarily records the vehicle length measurement result (step ST107).
The processing from step ST104 to step ST107 is repeated until the vehicle detection signal ON from the second vehicle detector S2 is switched to the vehicle detection signal OFF.

When the vehicle advances and passes between the pair of first vehicle detectors S1 arranged oppositely on the two islands 40, the vehicle detection signal ON of the first vehicle detector S1 is switched to the vehicle detection signal OFF (step ST108, YES) (timing t2 or t4 in the timing chart of FIG. 5) (step ST6-3 in FIG. 6 showing an example of a large car) (step ST7-3 in FIG. 7 showing an example of a regular car).
When the lane server 3 detects that the vehicle detection signal from the first vehicle detector S1 is turned off, it instructs to end the photographing. That is, the lane server 3 uses the vehicle detection signal OFF from the first vehicle detector S1 as a trigger to instruct the end of photographing.
The wide-angle lens-equipped camera 421 ends the video shooting based on the instruction to end the shooting (step ST109).
The lane server 3 makes a final judgment on the axle number measurement result (step ST110), and makes a final judgment on the vehicle measurement result (step ST111).

For example, as described above, the lane server 3 measures the number of axles and vehicle length multiple times from images taken and output during the period when the vehicle detection signal from the first vehicle detector S1 is ON, and confirms the number of axles and vehicle length. The final judgment is the most likely result.
As shown in FIG. 6, even in the case of a large vehicle, the positions of the vehicle are different in step ST6-1, step ST6-2, and step ST6-3, so the photographing results of the vehicle are also different, and the axle measurement result and vehicle length measurement result are different. may be different.
Similarly, as shown in FIG. 7, even in the example of a small car, the positions of the vehicle are different in steps ST7-1, ST7-2, and ST7-3, so the photographing results of the vehicle are also different, and the axle measurement results and vehicle length are different. The measurement results may differ.

For example, an image taken at the midpoint of the ON period of the vehicle detection signal from the first vehicle detector S1 shown in FIG. In the image corresponding to step ST7-2 in FIG. 7), when the vehicle is viewed from the side, the position approximately at the center of the vehicle length approximately coincides with the position of the wide-angle lens-equipped camera 421. Therefore, the measurement result obtained from the image taken in the middle of the ON period of the vehicle detection signal from the first vehicle detector S1 shown in FIG. 5 is determined as the most probable result as the final determination.
Alternatively, by majority vote, the number of axles with matching measurement results is determined to be the most probable result and the final determination is made. If the number of axles is 3 in the first determination, but the number of axles is 4 in subsequent determinations, the final determination as a probable result will be 4 axles.
Further, by majority vote, the average value of the vehicle lengths with similar measurement results is determined as the most probable result as the final determination.
Alternatively, a probable result may be determined based on the deviation value of the entire measurement results.

The lane server 3 determines the type of vehicle, such as a large vehicle or a regular vehicle, based on the final determination of the number of axles and vehicle length, and outputs the determination (step ST112).
Steps ST101 to ST112 are repeated until the ETC system is stopped (step ST113, NO).

The tollgate server 2 refers to the toll table for each section and vehicle type, calculates the toll based on the entrance information, exit information, and vehicle type information (vehicle type discrimination result by the lane server 3), and calculates the toll toll based on the entrance information, exit information, and vehicle type information (vehicle type discrimination result by the lane server 3). 3. Send toll fees, vehicle type information, etc.
The lane server 3 transmits tolls, vehicle type information, etc. to the interface aggregation unit 401.
The interface aggregation unit 401 transmits tolls, vehicle type information, etc. to the first antenna 411, the second antenna 412, and the recovery antenna 413. The first antenna 411, the second antenna 412, and the recovery antenna 413 transmit tolls, vehicle type information, etc. to the on-vehicle device.
Further, the interface aggregation unit 401 outputs tolls, vehicle type information, etc. to the in-booth display 4241. The in-booth display 4241 displays tolls, vehicle type information, and the like.
Toll collectors can check the toll and vehicle type information displayed on the booth display 4241 without having to identify the vehicle type themselves, receive cash from users who wish to pay in cash, and transfer it to the cash processing terminal. Settlement processing can be executed at 4242.

In the above description, photographing is started using the switching from the vehicle detection signal OFF to the vehicle detection signal ON of the first vehicle detector S1 as a trigger, and the vehicle detection signal is changed from the vehicle detection signal ON of the first vehicle detector S1 to the vehicle detection signal ON. Although a case has been described in which shooting is stopped using the switching to OFF as a trigger, the first embodiment is not limited to this.
The video is always shot and recorded (the video is overwritten and recorded after a certain period of time), and the recording is performed using the switching from the vehicle detection signal OFF to the vehicle detection signal ON of the first vehicle detector S1 as a trigger. The acquisition of the recorded video is started using the switching from the vehicle detection signal ON to the vehicle detection signal OFF of the first vehicle detector S1 as a trigger, and the axle measurement is performed based on the acquired video. and vehicle length measurement.

(Second embodiment: Vehicle type discrimination processing based on still images)
Next, a second embodiment will be described with reference to FIGS. 6, 7, 8, and 9.
The second embodiment is a vehicle type discrimination process that discriminates the vehicle type based on a still image taken by a camera 421 equipped with a wide-angle lens. The photographing timing (shutter-on timing) by the wide-angle lens-equipped camera 421 is controlled based on the vehicle detection situation, and the type of vehicle included in the image acquired at this photographing timing is determined.

FIG. 9 is a timing chart showing an example of the timing of vehicle detection by the first vehicle detectors S1 and S2.
The timing chart shown in the upper part of FIG. An example of the output period of the vehicle detection signal ON detected by the second vehicle detector S1 and the output period of the vehicle detection signal ON detected by the second vehicle detector S2 is shown.
The timing chart shown in the lower part of FIG. An example of the output period of the vehicle detection signal ON detected by the second vehicle detector S2 and the output period of the vehicle detection signal ON detected by the second vehicle detector S2 is shown.

FIG. 8 is a flowchart illustrating an example of vehicle type discrimination processing based on still images according to the second embodiment.
As shown in FIG. 8, the lane server 3 monitors the vehicle detection signal from the interface aggregation unit 401 and enters a standby state for photographing a vehicle (step ST201).

A vehicle such as a large vehicle or a regular vehicle is traveling from the upstream lane to the downstream direction of the tollgate lane. Then, when the vehicle blocks the infrared light, etc. between the pair of first vehicle detectors S1 arranged oppositely on the two islands 40, the vehicle detection signal of the first vehicle detector S1 turns OFF and the vehicle detection signal turns ON. (Step ST202, YES) (timing t1 in the timing chart of FIG. 9) (Step ST6-1 in FIG. 6 showing an example of a large car) (Step ST7-1 in FIG. 7 showing an example of a regular car) ).

Below, we will explain the cases separately for large cars and ordinary cars.
In the case of a large vehicle, while the vehicle is traveling and being detected by the first vehicle detector S1 (during the output period of the vehicle detection signal ON) (step ST202, YES → step ST203, NO), A vehicle reaches the second vehicle detector S2.
Then, when the vehicle blocks infrared light etc. between the second vehicle detector S2, the vehicle detection signal OFF of the second vehicle detector S2 is switched to the vehicle detection signal ON (step ST204, YES) (FIG. 9 (timing t3) in the timing chart of FIG. 6 (step ST6-2 in FIG. 6, which shows an example of a large vehicle).
During the period when the vehicle detection signal of the first vehicle detector S1 is ON (step ST202, YES → step ST203, NO), the lane server 3 detects the vehicle detection signal ON of the second vehicle detector S2 ( Step ST204, YES), instructs to shoot (shutter on).
That is, the lane server 3 instructs photographing using the vehicle detection signal ON of the second vehicle detector S2 as a trigger during the period when the vehicle detection signal of the first vehicle detector S1 is ON, and the wide-angle lens equipped camera 421 performs photographing. A still image is photographed based on the instruction (step ST205).
Furthermore, when the vehicle advances and passes between the first vehicle detectors S1, the vehicle detection signal of the first vehicle detector S1 is switched from OFF to vehicle detection signal ON (timing t4 in the timing chart of FIG. 9). .
Furthermore, when the vehicle advances and passes between the second vehicle detectors S2, the vehicle detection signal of the second vehicle detector S2 is switched from OFF to vehicle detection signal ON.

On the other hand, in the case of a regular car, when the vehicle moves forward, the first vehicle detector S1 first detects the vehicle (the vehicle detection signal of the first vehicle detector S1 turns ON). As the vehicle advances further from there, the vehicle passes between the first vehicle detectors S1 before reaching the second vehicle detector S2, and the vehicle detection signal of the first vehicle detector S1 turns ON to detect the vehicle. The signal is switched to OFF (step ST203, YES) (timing t2 in the timing chart of FIG. 9) (step ST7-3 in FIG. 7 showing an example of a regular car).
When the lane server 3 detects the switching from the vehicle detection signal ON of the first vehicle detector S1 to the vehicle detection signal OFF during the period when the vehicle detection signal of the second vehicle detector S2 is OFF (step ST202, YES) →Step ST203, YES), and instructs to take a picture (shutter on).
That is, the lane server 3 instructs photography using the switching from the vehicle detection signal ON of the first vehicle detector S1 to the vehicle detection signal OFF as a trigger during the period when the vehicle detection signal of the second vehicle detector S2 is OFF. . The wide-angle lens equipped camera 421 photographs a still image based on the photographing instruction (step ST205).
Furthermore, when the vehicle advances and reaches the second vehicle detector S2, the vehicle detection signal of the second vehicle detector S2 is switched from OFF to vehicle detection signal ON (timing t3 in the timing chart of FIG. 9).
Then, when the vehicle passes through the second vehicle detector S2, the vehicle detection signal ON of the second vehicle detector S2 is switched to the vehicle detection signal OFF (timing t5 in the timing chart of FIG. 9).

The image processing device 422 detects a vehicle image (including the axle) from the still image output from the wide-angle lens equipped camera 421, and outputs the vehicle image.
The lane server 3 detects axles from the detected image, measures the number of axles (step ST206), and records the result of measuring the number of axles (step ST207).
Further, the lane server 3 measures the vehicle length from the image of the detected vehicle (step ST208), and records the vehicle length measurement result (step ST209).

The lane server 3 determines and outputs the vehicle type, such as a large vehicle or a regular vehicle, from the measurement results of the number of axles and vehicle length (step ST210).
Steps ST201 to ST210 are repeated until the ETC system is stopped (step ST2113, NO).

The tollgate server 2 refers to the toll table for each section and vehicle type, calculates the toll based on the entrance information, exit information, and vehicle type information (vehicle type discrimination result by the lane server 3), and calculates the toll toll based on the entrance information, exit information, and vehicle type information (vehicle type discrimination result by the lane server 3). 3. Send toll fees, vehicle type information, etc.
The lane server 3 transmits tolls, vehicle type information, etc. to the interface aggregation unit 401.
The subsequent processing is the same as that in the first embodiment, and a description thereof will be omitted.

In the above description, shutter-on control is performed using the detection of the vehicle detection signal ON of the second vehicle detector S2 as a trigger during the period in which the vehicle detection signal ON of the first vehicle detector S1 is detected. We have explained the case of outputting a signal and taking a still image. Furthermore, in the above description, during the period in which the vehicle detection signal OFF of the second vehicle detector S2 is detected, the switching from the vehicle detection signal ON of the first vehicle detector S1 to the vehicle detection signal OFF is detected. We explained a case where a shutter-on control signal is output using this as a trigger and a still image is taken. However, the second embodiment is not limited to this.
A video is always shot and recorded (the video is overwritten after a certain period of time has elapsed), and during the period when the vehicle detection signal of the first vehicle detector S1 is ON, the vehicle of the second vehicle detector S2 is detected. Using the detection of the detection signal ON as a trigger, a still image is acquired from the recorded video, and during the period when the vehicle detection signal OFF of the second vehicle detector S2 is detected, the first vehicle detection The detection of the switching from the vehicle detection signal ON to the vehicle detection signal OFF of the device S1 is used as a trigger to acquire a still image from the recorded video, and the axle measurement and vehicle length are performed based on the acquired still image. Measurements may also be performed.
Further, when acquiring a still image from a recorded moving image in this way, a plurality of still images before and after the above-mentioned trigger may be acquired, and the vehicle type may be determined from the plurality of still images.
In this case, by majority vote, the number of axles whose measurement results match may be determined as the most likely result and the final determination may be made, and the average value of the vehicle lengths of the measurement results may be determined as the most probable result and the final determination is made.

As explained above, according to the first and second embodiments, the number of axles and vehicle length are detected using the camera 421 equipped with a wide-angle lens, so an axle detector such as a pressure sensor embedded in the road is not required. Large-scale construction costs can be reduced.
Additionally, it was not possible to detect the number of axles until all the axles passed over the axle detector using a pressure sensor or the like (although it was not possible to determine the type of vehicle), but this will also be improved.
Furthermore, when a collector in a booth collects tolls according to the type of vehicle, it is necessary to determine the type of vehicle before the driver's seat of the vehicle reaches the booth. In other words, all the axles of the vehicle must pass over an axle detector such as a pressure sensor before the driver's seat of the vehicle reaches the booth. For this reason, a considerable distance and space was required from the installation location of the axle detector, such as a pressure sensor, to the booth.
However, such a considerable distance space is no longer necessary, and vehicle type discrimination can be performed in a limited and narrow toll booth space.

Further, by appropriately controlling the timing of video shooting by the wide-angle lens equipped camera 421 according to the vehicle detection situation by the first vehicle detector S1, highly accurate axle and vehicle length detection is possible.
In addition, by appropriately controlling the timing of still image shooting by the wide-angle lens equipped camera 421 according to the vehicle detection status by the first vehicle detectors S1 and S2, highly accurate axle and vehicle length detection is possible. .
As a result, accurate vehicle identification becomes possible. If the timing of capturing a video or still image is not appropriately controlled, there is a risk that vehicles in other lanes may be included in the captured image. In such a case, the vehicle type of the vehicle traveling in the other lane will be determined instead of the vehicle type of the vehicle traveling in the original lane, and as a result, there is a risk that the vehicle type will be incorrectly determined.
According to the first and second embodiments, it is possible to prevent such erroneous determination due to erroneous photographing of vehicles in other lanes.

Although several embodiments of the invention have been described, these embodiments are presented by way of example and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other forms, and various omissions, substitutions, and changes can be made without departing from the gist of the invention. These embodiments and their modifications are included within the scope and gist of the invention, as well as within the scope of the invention described in the claims and its equivalents.
Hereinafter, the invention described in the original claims of this application will be additionally described.
[C1]
a first vehicle detector provided along a lane in which a vehicle is traveling and detects a vehicle reaching a first point on the lane;
a second vehicle detector provided along the lane and detecting a vehicle arriving at a second point downstream from the first point of the lane;
a photographing unit provided at a position closer to the first vehicle detector than the second vehicle detector, and photographing a predetermined range of the lane including the first and second points with a wide-angle lens;
a control unit that controls timing of photographing by the photographing unit based on a vehicle detection situation by at least one of the first and second vehicle detectors;
a determination unit that determines the model of a vehicle included in an image photographed by the photographing unit at a timing controlled by the control unit;
an output unit that outputs a vehicle type discrimination result;
A vehicle type identification device.
[C2]
The vehicle type determination device according to [C1], wherein the determination unit measures the number of axles and the length of the vehicle included in the image, and determines the type of vehicle based on the number of axles and the length of the vehicle.
[C3]
The control unit instructs photography in response to a vehicle detection period by the first vehicle detector, and the photography unit instructs video recording in response to the vehicle detection period based on the photography instruction from the control unit. Vehicle type identification device for photographing [C1] or [C2].
[C4]
The control unit instructs photographing based on a first timing when a vehicle is detected by the second vehicle detector during a vehicle detection period by the first vehicle detector,
The vehicle type discrimination device [C1] or [C2], wherein the photographing section photographs a still image at the first timing based on a photographing instruction from the control section.
[C5]
The control unit instructs photographing based on a second timing of the end of vehicle detection by the first vehicle detector during a period when the second vehicle detector does not detect a vehicle,
The vehicle type discrimination device according to [C4], wherein the photographing section photographs a still image at the second timing based on a photographing instruction from the control section.
[C6]
The photographing unit is a vehicle type discrimination device according to any one of [C1] to [C5], which is installed within a predetermined length range along the direction in which the lane extends with respect to the installation position of the first vehicle detector. . [C7]
In the vehicle type discrimination device [C6], the photographing unit is installed upstream of the first vehicle detector.
[C8]
The discrimination unit discriminates between a first vehicle type having a predetermined length or more from the first point to the second point, and a second vehicle type having a length less than the predetermined length from the first point to the second point. One vehicle type identification device.
[C9]
a first vehicle detector provided along a lane in which a vehicle travels and detects a vehicle reaching a first point on the lane; and a first vehicle detector provided along the lane and detecting a vehicle reaching a first point on the lane; a second vehicle detector for detecting a vehicle arriving at a downstream second point; and a second vehicle detector provided at a position closer to the first vehicle detector than the second vehicle detector; A vehicle type discrimination method that discriminates the vehicle type by a photographing unit that photographs a predetermined range of the lane including a second point, the method comprising:
controlling the timing of photographing by the photographing unit based on a vehicle detection situation by at least one of the first and second vehicle detectors;
determining the model of the vehicle included in the image photographed by the photographing unit at the photographing timing;
A vehicle type discrimination method that outputs the vehicle type discrimination result.

1...Central device 2...Toll plaza server 3...Lane server 4...Lane equipment 40...Island 71...Onboard equipment 401...Interface aggregation unit 411...First antenna 412...Second antenna 413...Recovery antenna 421...Wide-angle lens mounted camera 422 …Image processing device 423…Roadside display 424…Booth 425…Start control device 426…Lane monitoring camera 4241…In-booth display 4242…Cash processing terminal 4251…Start control bar

Claims (8)

a first vehicle detector provided along a lane in which a vehicle is traveling and detects a vehicle reaching a first point on the lane;
a second vehicle detector provided along the lane and detecting a vehicle arriving at a second point downstream of the first point of the lane and upstream of the launch controller;
The second vehicle detector is provided at a position closer to the first vehicle detector, and a wide-angle lens is used to detect a predetermined range of a predetermined length in the upstream and downstream directions of the lane, including the first and second points. a photography department that takes still images of
The photographing timing of the photographing unit is set to a first timing when a vehicle is detected by the second vehicle detector during a vehicle detection period by the first vehicle detector, or a first timing when a vehicle is detected by the first vehicle detector. a control unit configured to take the still image at a second timing when vehicle detection by the first vehicle detector ends after detection and before vehicle detection by the second vehicle detector;
The number of axles and the vehicle length of the vehicle are measured from the still image photographed by the photographing section with the wide-angle lens at the first timing or the second timing controlled by the control section, and a determination unit that determines the vehicle type based on the vehicle length;
A vehicle type identification device. The predetermined length is about 15 m in the upstream and downstream directions of the lane with reference to the installation position of the imaging unit.
The vehicle type discrimination device according to claim 1. the wide-angle lens is a fisheye lens;
The vehicle type discrimination device according to claim 1. The photographing section constantly photographs and records a moving image using the wide-angle lens, and acquires the still image from the recorded moving image at the first timing or the second timing instructed by the control section. Department,
The vehicle type discrimination device according to claim 1. The imaging unit is installed within a predetermined length range along the direction in which the lane extends, with reference to the installation position of the first vehicle detector.
A vehicle type discrimination device according to any one of claims 1 to 4. The imaging unit is installed upstream of the first vehicle detector,
The vehicle type discrimination device according to claim 5. In the vehicle model discrimination method for determining the vehicle model,
a first vehicle detection procedure in which a first vehicle detector provided along a lane in which a vehicle is traveling detects a vehicle reaching a first point on the lane;
second vehicle detection, in which a second vehicle detector provided along the lane detects a vehicle arriving at a second point on the lane downstream of the first point and upstream of the launch controller; steps and
The control unit sets the photographing timing of the photographing unit to a first timing when the first vehicle detector is during a vehicle detection period and a vehicle is detected by the second vehicle detector, or a first timing when a vehicle is detected by the second vehicle detector. A control procedure for controlling to take a still image at a second timing when the first vehicle detector finishes detecting the vehicle after the vehicle is detected by the device and before the second vehicle detector detects the vehicle;
The photographing unit, which is provided at a position closer to the first vehicle detector than the second vehicle detector, uses a wide-angle lens to capture images using a wide-angle lens at the first timing or the second timing based on the control of the control unit. a photographing procedure of photographing a still image in a predetermined length range in an upstream direction and a downstream direction of the lane including the first and second points;
a vehicle type determination procedure in which the determination unit determines the type of vehicle included in the still image photographed by the photographing unit at the photographing timing;
A method for identifying vehicle types. The predetermined length is about 15 m in the upstream and downstream directions of the lane with reference to the installation position of the imaging unit.
The vehicle type discrimination method according to claim 7.