JP7525247B2 - Vehicle type determination device, vehicle type determination method, and vehicle type determination program - Google Patents

Description

This invention relates to a technology for measuring the traffic volume of different types of vehicles on a road on which vehicles are traveling.

Conventionally, there has been a device that uses a millimeter wave sensor to identify the vehicle type (see Patent Document 1). This device in Patent Document 1 is configured to measure the vehicle length and identify the vehicle type based on the measured vehicle length.

JP 2003-248893 A

However, there are many different types of vehicles that are classified according to size and shape. The types of vehicles referred to here include light cars, light trucks, compact cars, minivans, sedans, station wagons, buses, small trucks, large trucks, cranes, motorcycles (two-wheeled vehicles), etc. Also, even if the types of vehicles are different, there are many vehicles that are about the same length.

As mentioned above, the device in Patent Document 1 was configured to distinguish vehicle types based on vehicle length, and therefore distinguished vehicle types into two categories: large and small. Therefore, the device in Patent Document 1 was unable to distinguish between types of vehicles that are classified based on size and shape.

The objective of this invention is to provide a technology for determining the type of vehicle classified based on size and shape.

To achieve the above objective, the traffic volume measurement device of this invention is configured as follows:

The detection data input unit receives vehicle detection data including the position of the vehicle detected by the vehicle detection device using a search wave to scan a target area set for the road on which the vehicle is traveling. The image input unit receives a frame image captured by the imaging device. The imaging device is installed in a direction that captures the front of the vehicle from the downstream side of the road.

The type determination unit processes the frame image captured by the imaging device when a vehicle is located at a type determination position set on the road based on the vehicle detection data input to the detection data input unit, and determines the type of vehicle located at the type determination position. This vehicle type is classified based on the size and external shape of the vehicle.

With this configuration, the vehicle whose type is to be determined is imaged by the imaging device when it is located at a type determination position that is a fixed distance away from the imaging device, so that the size of the vehicle in the frame image is reduced at a substantially fixed ratio. Therefore, when determining the type of vehicle captured in the frame image, the effect of fluctuations in the distance between the imaging device and the vehicle when the frame image was captured can be suppressed, and the type of the captured vehicle can be determined.

In addition, if the type determination position is set within the range of the depth of field of the imaging device, it is possible to process a frame image in which the focus is on the subject vehicle and determine the type of the captured vehicle. In other words, it is possible to prevent processing a frame image in which the focus is not on the subject vehicle and the captured vehicle is blurred and the type of the captured vehicle being determined.

This allows for accurate determination of vehicle type based on size and external shape.

The system may also be configured to include a detection unit that detects the type determination timing at which a vehicle was located at a type determination position based on vehicle detection data input to the detection data input unit, and the type determination unit processes frame images input to the image input unit, the image capture timing of which corresponds to the type determination timing, to determine the type of vehicle.

In this configuration, a video camera capable of capturing moving images can be used as the imaging device. This allows the determination of vehicle types classified according to size and shape to be performed not only in real time but also in batch processing.

The system may also be configured to include a determination unit that determines whether a vehicle is located at a type determination position based on the vehicle detection data input to the detection data input unit, and an image capture instruction unit that instructs the image capture device to capture an image when the determination unit determines that the vehicle is located at the type determination position, and the type determination unit processes the frame images captured in response to the image capture instruction from the image capture instruction unit to determine the type of vehicle. With this configuration, a digital still camera that captures still images can be used as the image capture device, and the image capture device can capture frame images at the required timing.

The system may also be configured to include a traffic volume data generation unit that generates traffic volume data by tallying the traffic volume for each vehicle type determined by the type determination unit.

With this configuration, the generated traffic volume data can be used to accurately estimate the condition of the road surface (such as the degree of damage).

The type determination unit may also be configured to determine the name given to the vehicle.

This invention makes it possible to determine the type of vehicle classified based on size and external shape.

FIG. 1A is a plan view seen in the width direction of a vehicle traveling on a road, and FIG. 1B is a plan view seen in a direction perpendicular to the road surface of the road on which the vehicle travels. 2 is a diagram showing the connection between a vehicle type determination device, a radio wave radar device, and an imaging device according to this example. FIG. 13 is a diagram showing a frame image in which a target vehicle is captured. FIG. 2 is a diagram showing a configuration of a main part of a vehicle type determination device according to this example. FIG. 2 is a diagram illustrating data registered in a feature amount DB. FIG. 4 is a diagram showing an example of traffic volume data generated by a traffic volume data generating unit; 4 is a flowchart showing the operation of the radio radar device. FIG. 2 is a diagram illustrating tracking data. 4 is a flowchart showing the operation of the vehicle type determination device according to this example. FIG. 13 is a diagram showing a configuration of a main part of a vehicle type determination device according to a modified example. 10 is a flowchart showing the operation of the vehicle type determination device according to the modified example.

The vehicle type determination device according to an embodiment of the present invention is described below.

＜1. Application examples＞
Fig. 1 is a schematic diagram showing a road on which a vehicle to be determined as a type of vehicle travels using a vehicle type determination device according to this embodiment. Fig. 1(A) is a plan view seen in the width direction of the vehicle traveling on the road, and Fig. 1(B) is a plan view seen in a direction perpendicular to the road surface of the road on which the vehicle travels. Fig. 2 is a diagram showing the connection between the vehicle type determination device, the radio radar device, and the imaging device according to this embodiment.

As shown in FIG. 2, the vehicle type determination device 1 in this example is connected to a radio wave radar device 5 and an image capture device 6. The vehicle type determination device 1 may be connected to the radio wave radar device 5 and the image capture device 6 by wire or wirelessly. As will be described later, the radio wave radar device 5 is installed in a position where it can scan the target area set for the road with radio waves, which are search waves. The image capture device 6 is installed so as to focus on a subject located at a vehicle type determination position 101 set within the target area. In this way, the radio wave radar device 5 and the image capture device 6 are installed near the target area 100, but the vehicle type determination device 1 may be installed near the target area 100, or may be installed in a control center, data collection center, or the like, away from the target area 100.

Figure 1 shows a three-lane road. The radio radar device 5 and the imaging device 6 are attached, for example, to a pole installed on the roadside of the road, or to an arm attached to the pole and extending in the width direction (vehicle width direction) of the vehicle 110 traveling on the road. The radio radar device 5 and the imaging device 6 may also be attached to a gantry installed to straddle the road.

The radio radar device 5 scans the target area 100 shown in FIG. 1 with radio waves (corresponding to the search waves in this invention) and detects the position, speed, and size of the traveling vehicle 110. The radio radar device 5 repeats scanning the target area 100 with radio waves at set time intervals (e.g., 100 msec intervals).

The radio radar device 5 may also be configured to generate tracking data indicating the travel trajectory of the vehicle 110 within the target area 100. In this example, the radio radar device 5 is described as having the configuration for generating this tracking data, but the configuration for generating this tracking data may also be provided in the vehicle type determination device 1. The configuration for generating this tracking data is configured to associate the vehicle 110 detected in the previous scan of the target area 100 by the radio radar device 5 with the vehicle 110 detected in the current scan of the target area 100 by radio waves. Each time the radio radar device 5 scans the target area 100, it outputs vehicle detection data that associates the ID, position, speed, and size of the vehicle 110 detected in the current scan.

The imaging device 6 is a known video camera. The frame rate of the imaging device 6 is several tens of frames/sec (for example, 30 frames/sec). As described above, the imaging device 6 focuses on a subject located at the vehicle type determination position 101. This vehicle type determination position 101 corresponds to the type determination position referred to in this invention. This vehicle type determination position 101 is a position several tens of meters (approximately 10 m to 30 m) away from the installation position of the imaging device 6. The imaging device 6 is attached at an angle that captures an image of the front of the vehicle 110a located at the vehicle type determination position 101. The imaging device 6 outputs the captured video image.

Note that vehicle 110a indicates a vehicle 110 located at the vehicle type determination position 101 (to be distinguished from a vehicle 110 that is not located at the vehicle type determination position 101).

As shown in FIG. 2, the vehicle type determination device 1 in this example receives the above-mentioned vehicle detection data from the radio radar device 5 and also receives video images from the imaging device 6.

The vehicle type determination device 1 detects the type determination timing when the vehicle 110 is located at the vehicle type determination position 101 based on the input vehicle detection data. The vehicle type determination device 1 processes the frame image whose image capture timing corresponds to the type determination timing detected here, and determines the type of the vehicle 110a located at the vehicle type determination position 101. The image capture timing corresponding to the type determination timing is the image capture timing of the image capture device 6 that is closest in time to this type determination timing. The type of the vehicle 110a determined here is a type classified according to size and shape, and includes light cars, light trucks, compact cars, minivans, sedans, station wagons, buses, small trucks, large trucks, cranes, motorcycles (two-wheeled vehicles), etc.

As described above, the imaging device 6 focuses on the subject located at the vehicle type determination position 101. Therefore, the frame image captured by the imaging device 6 at the imaging timing corresponding to the type determination timing is a frame image captured when the vehicle 110a to be determined (hereinafter, the vehicle 110a may be referred to as the determination target vehicle 110a) is located within the depth of field. Therefore, in the frame image captured by the imaging device 6 at the imaging timing corresponding to the type determination timing, the determination target vehicle 110a is not blurred due to focus shift, and the front is captured sharply. In addition, since the determination target vehicle 110a is captured by the imaging device 6 when it is located at the vehicle type determination position 101 at a certain distance from the imaging device 6, the size on the frame image is reduced at a substantially constant ratio. Therefore, in determining the type of the determination target vehicle 110a captured in the frame image, the influence of the change in the distance between the imaging device 6 and the determination target vehicle 110a at the time the frame image is captured can be suppressed, and the type of the captured determination target vehicle 110a can be determined. FIG. 3 shows a frame image in which the vehicle to be judged is captured.

The vehicle type determination device 1 processes the frame image to extract the features of the vehicle 110a to be determined, and uses the extracted features to determine the type of the vehicle 110a to be determined, which has been classified according to size and shape. The type of the vehicle 110a to be determined may be determined by registering the type of the vehicle 110 and the features of the front of the vehicle 110 in a feature DB in association with each other, processing the frame image to extract the features, and comparing them with the features of the vehicle 110a to determine the type of the vehicle 110. Alternatively, the type of the vehicle 110 may be determined by AI (Artificial Intelligence) (AI recognizes the vehicle type using a recognition model).

This vehicle type determination device 1 determines the type of vehicle 110 classified according to size and shape. In addition, by using the vehicle detection data input from the radio radar device 5, it is possible to generate traffic volume data that tally up the traffic volume for each type of vehicle 110 classified according to size and shape.

2. Configuration example
4 is a diagram showing the configuration of the main parts of the vehicle type determination device 1 according to this example. The vehicle type determination device 1 according to this example includes a control unit 11, a detection data input unit 12, an image input unit 13, a feature database 14 (feature DB 14), and an output unit 15.

The control unit 11 controls the operation of each part of the vehicle type determination device 1. The control unit 11 also has a detection unit 21, a type determination unit 22, and a traffic volume data generation unit 23. The detection unit 21, type determination unit 22, and traffic volume data generation unit 23 of the control unit 11 will be described later.

The radio radar device 5 is connected to the detection data input unit 12. Vehicle detection data is input to the detection data input unit 12 from the radio radar device 5. The vehicle detection data is data that associates the position, speed, and size of the detected vehicle 110. Each time the radio radar device 5 scans the target area 100 with radio waves, it outputs the vehicle detection data of each vehicle 110 detected during that scan to the vehicle type determination device 1.

The image input unit 13 is connected to the imaging device 6. A moving image captured by the imaging device 6 is input to the image input unit 13.

As shown in FIG. 5, the feature DB 14 is a database in which the name of the vehicle 110, the type of the vehicle 110, and the feature of the vehicle 110 are registered in association with each other for each vehicle 110. The feature of each vehicle 110 registered in the feature DB 14 is the feature of the front of the vehicle 110.

The feature DB 14 may be a database in which the type of vehicle 110 and the feature of the vehicle 110 are registered in association with each other (the name of the vehicle 110 does not have to be associated with the type). The name of the vehicle 110 referred to here is a name given by the automobile manufacturer to distinguish the vehicle 110 from other vehicles 110.

The output unit 15 outputs traffic volume data, which will be described later, to a higher-level device (not shown) such as a control center.

Next, we will explain the detection unit 21, type determination unit 22, and traffic volume data generation unit 23 of the control unit 11.

The detection unit 21 determines whether a vehicle 110 is located at the vehicle type determination position 101 based on the vehicle detection data input to the detection data input unit 12. The detection unit 21 detects the vehicle 110 located at the vehicle type determination position 101 as a determination target vehicle 110a.

The type determination unit 22 selects a frame image whose capture timing corresponds to the type determination timing from among the frame images captured by the imaging device 6.

The type determination timing may be, for example, the timing when the radio radar device 5 starts scanning the target area 100, the timing when the scanning ends, or the timing when the detection unit 21 determines that a vehicle (the vehicle to be determined 110a) is located at the vehicle type determination position 101. In addition, a frame image whose imaging timing corresponds to the type determination timing is a frame image whose imaging timing is closest in time to the type determination timing.

The type determination unit 22 processes the selected frame image and extracts features of the front of the vehicle 110a being determined. The type determination unit 22 compares the extracted features of the vehicle 110a being determined with the features of each vehicle 110 registered in the feature DB 14, and determines the model of the vehicle 110a being determined. In this example, the type determination unit 22 can also determine the name of the vehicle 110a being determined.

The traffic volume data generating unit 23 generates traffic volume data for the target area 100. The traffic volume data generated by the traffic volume data generating unit 23 is data that associates the number of vehicles 110 by type and average speed that traveled in the target area 100 during a certain period (in FIG. 6, one hour from 7:00 AM to 8:00 AM on January 15, 2019), as shown in FIG. 6, for example. The traffic volume data generated by the traffic volume data generating unit 23 is output to a higher-level device by the output unit 15.

The traffic volume data generated by the traffic volume data generating unit 23 may be in a format that does not include the average speed described above, and may include other items such as the average inter-vehicle distance, as long as the traffic volume by type of vehicle 110 can be obtained.

The control unit 11 of the vehicle type determination device 1 is composed of a hardware CPU, memory, and other electronic circuits. When the hardware CPU executes the vehicle type determination program of the present invention, it operates as a detection unit 21, a type determination unit 22, and a traffic volume data generation unit 23. The memory also has an area for expanding the vehicle type determination program of the present invention, and an area for temporarily storing data generated when the vehicle type determination program is executed. The control unit 11 may be an LSI that integrates the hardware CPU, memory, and the like. The hardware CPU is also a computer that executes the vehicle type determination method of the present invention.

<3. Operation example>
7 is a flowchart showing the operation of the radio wave radar device. When it is time to scan the target area 100 (s1), the radio wave radar device 5 scans the target area 100 with radio waves, which are search waves, and detects the reflected waves to perform a detection process to detect vehicles 110 located within the target area 100 (s2). In s2, the radio wave radar device 5 detects the time it takes to detect the reflected waves (time of flight of the radio waves) and the frequency of the reflected waves for each direction of the radio waves, and detects the position, size, and speed of the object (vehicle 110, road surface, etc.) that reflected the radio waves.

When the detection process in s2 is completed, the radio radar device 5 performs a matching process to match the vehicles 110 detected in the previous scan of the target area 100 with the vehicles 110 detected in the current scan of the target area 100 (s3). The matching process in s3 is performed using the size, position, and speed of the vehicles 110. For the vehicles 110 that were able to be matched with the vehicles 110 detected in the previous scan (the vehicles 110 detected in the current scan), the radio radar device 5 assigns the ID assigned to the matched vehicles 110. On the other hand, for the vehicles 110 that were not able to be matched with the vehicles 110 detected in the previous scan (the vehicles 110 detected in the current scan), the radio radar device 5 generates and assigns a new ID.

The radio radar device 5 generates vehicle detection data that associates the ID, position, speed, and size of each vehicle 110 detected in this scan, outputs the vehicle detection data generated here to the vehicle type determination device 1 (s4), and returns to s1.

In this example, the radio radar device 5 scans the target area 100 with radio waves at 100 msec intervals and outputs vehicle detection data related to the detected vehicle 110 to the vehicle type determination device 1. In other words, the vehicle detection data is input to the vehicle type determination device 1 at 100 msec intervals.

By performing the above processing, the radio radar device 5 can obtain the tracking data shown in FIG. 8 for each vehicle 110 that has traveled through the target area 100. This tracking data is data that associates an ID that identifies the vehicle 110, the size of the vehicle 110, and the speed and position of the vehicle 110 for each detection time. In this example, the position of the vehicle 110 is the distance from a first reference position in the traveling direction of the vehicle 110 and the distance from a second reference position in the width direction of the traveling road.

The vehicle type determination device 1 can also obtain the tracking data shown in Figure 8 by classifying and aggregating the vehicle detection data input from the radio radar device 5 by ID.

The radio radar device 5 may also be configured not to generate tracking data. In this case, the radio radar device 5 does not perform the process in s3 described above, but generates vehicle detection data in s4 that associates the position, speed, and size (vehicle detection data that does not associate an ID), and outputs this to the vehicle type determination device 1. The vehicle type determination device 1 performs the same process as in s3 described above to generate tracking data.

Figure 8 is a diagram showing tracking data for a certain vehicle. As shown in Figure 8, the tracking data is data that associates an ID that identifies the vehicle 110, the size of the vehicle 110, and the speed and position of the vehicle 110 for each detection time. In this example, the position of the vehicle 110 is the distance from a first reference position in the traveling direction of the vehicle 110 (e.g., the distance from the installation position of the radio radar device 5 in the traveling direction) and the distance from a second reference position in the width direction of the traveling path (e.g., the distance from the center position in the width direction of the traveling path, with the right side being a positive value and the left side being a negative value).

Figure 9 is a flowchart showing the operation of the vehicle type determination device in this example. When vehicle detection data is input to the detection data input unit 12 (s11), the vehicle type determination device 1 determines whether or not there is a vehicle 110a to be determined located at the vehicle type determination position 101 (s12). The process in s12 is executed by the detection unit 21. When the vehicle type determination device 1 determines in s12 that there is no vehicle 110a to be determined, it returns to s11.

When the vehicle type determination device 1 determines in s12 that a vehicle 110a is present, it selects a frame image to be used to determine the vehicle type of the vehicle 110a from among the moving images of the imaging device 6 input to the image input unit 13 (s13).

In s13, a frame image is selected according to the type determination timing. This type determination timing may be the timing when scanning of the target area 100 related to the vehicle detection data in which it was determined in s12 that the target vehicle 110a is present is started, or the timing when scanning of the target area 100 is completed, or even the timing when the detection unit 21 determines that a vehicle (target vehicle 110a) is located at the vehicle type determination position 101. In other words, the type determination timing may be the timing when the vehicle 110 (target vehicle 110a) is located near the vehicle type determination position 101.

In addition, in s13, a frame image captured at an imaging timing corresponding to the type determination timing is selected. The imaging timing corresponding to the type determination timing is the imaging timing closest in time to this type determination timing. Therefore, the frame image selected in s13 is a frame image captured when the vehicle 110 (the vehicle to be determined 110a) is located near the vehicle type determination position 101. As described above, the imaging device 6 focuses on the subject located at the vehicle type determination position 101, so the frame image selected in s13 does not have blur due to focus deviation, and the front of the vehicle to be determined 110a is captured sharply. The frame image selected in s13 is a frame image captured when the vehicle to be determined 110a is located within the depth of field.

The vehicle type determination device 1 extracts features of the front of the vehicle 110a to be determined that are captured in the frame image selected in s13 (s14), and compares the extracted features with the features of the vehicle 110 registered in the feature DB 14 (s15). The vehicle type determination device 1 determines the vehicle type of the vehicle 110a to be determined based on the result of the feature comparison in s15 (s16). In s16, the vehicle type determination device 1 determines the type of the vehicle 110a to be determined that has the highest similarity in the features. At this time, the vehicle type determination device 1 also determines the name of the vehicle 110a to be determined. The type determination unit 22 executes the processes from s13 to s16.

The vehicle type determination device 1 generates a determination result that associates the vehicle type, speed (speed at the vehicle type determination position 101), and passing time (time of passing the vehicle type determination position 101), and stores this determination result in a storage medium such as a memory, hard disk, or SSD (Solid State Drive) (not shown) (s17), and returns to s11.

The traffic volume data generation unit 23 generates the traffic volume data shown in FIG. 6 by aggregating the judgment results generated in s17.

In this way, the vehicle type determination device 1 in this example can determine the type of vehicle classified according to the size and external shape of the vehicle 110. Therefore, it is possible to obtain traffic volume data that tally up the number of vehicles 110 that have traveled on a road for each vehicle type classified according to size and external shape, which can be usefully used for the maintenance management of the road surface, etc.

In the above explanation, it is assumed that the radio radar device 5 outputs vehicle detection data to the vehicle type determination device 1 in almost real time, and the imaging device 6 outputs captured video images to the vehicle type determination device 1 in almost real time. However, the vehicle detection data output by the radio radar device 5 and the video images output by the imaging device 6 may be stored in a storage medium such as a hard disk or SSD (Solid State Drive), and the process shown in FIG. 9 may be performed as a batch process. In this case, a timestamp or the like may be added to the vehicle detection data output by the radio radar device 5 and the video images output by the imaging device 6 in order to associate these data.

The vehicle type determination device 1 may be configured not to have the traffic volume data generation unit 23. In this case, the determination result stored in s17 may be transmitted to a higher-level device via the output unit 15, and the higher-level device may generate traffic volume data.

4. Modifications
Next, a modified example of the present invention will be described. Fig. 10 is a diagram showing the configuration of the main parts of a vehicle type determination device according to this modified example. In Fig. 10, the same components as those shown in Fig. 4 are given the same reference numerals. A vehicle type determination device 1A according to this modified example differs in that a control unit 11A includes a release signal generation unit 24 in addition to the detection unit 21, type determination unit 22, and traffic volume data generation unit described in the above example.

The vehicle type determination device 1A of this modified example also differs in that it includes a release signal output unit 16 in addition to the detection data input unit 12, image input unit 13, feature DB 14, and output unit 15 described in the above example.

The release signal generation unit 24 generates a release signal that instructs the imaging device 6 to capture a still image. The release signal output unit 16 outputs the release signal generated by the release signal generation unit 24 to the imaging device 6.

In this modified example, the imaging device 6 may be a digital still camera that does not have the function of capturing moving images. Furthermore, the imaging device 6 only needs to capture a still image when a release signal is input from the vehicle type determination device 1A as a trigger, and does not need to capture still images periodically.

Figure 11 is a flowchart showing the operation of the vehicle type determination device according to this modified example. In Figure 11, the same steps as those shown in Figure 9 are given the same step numbers.

The vehicle type determination device 1A according to this modified example performs the processes of s11 and s12 described above. When the vehicle type determination device 1A determines in s12 that the determination target vehicle 110a is present, the release signal generation unit 24 generates a release signal (s21). The vehicle type determination device 1A outputs the release signal generated in s21 to the imaging device 6 in the release signal output unit 16 (s22). In response to the release signal output in s22, the vehicle type determination device 1A waits for a frame image (still image) captured by the imaging device 6 to be input (s23), and when the frame image is input, performs the processes of s14 to s17 described above.

In this example, the imaging device 6 also focuses on the subject located at the vehicle type determination position 101. Therefore, in the frame image captured by the imaging device 6 in response to the release signal input from the vehicle type determination device 1A, the vehicle to be determined 110a is not blurred due to focus deviation, and the front of the vehicle is captured sharply.

Therefore, the vehicle type determination device 1A according to this modified example can determine the type of vehicle classified according to the size and external shape of the vehicle 110, as in the above example. Therefore, it is possible to obtain traffic volume data that tally up the number of vehicles 110 that have traveled on a road for each type of vehicle classified according to size and external shape, which can be usefully used for the maintenance and management of the road surface, etc.

The radio radar device 5 described in the above example may be replaced with a laser radar device that uses laser light as a search wave.

Note that this invention is not limited to the above-described embodiment as it is, and in the implementation stage, the components can be modified and embodied without departing from the gist of the invention. In addition, various inventions can be formed by appropriately combining multiple components disclosed in the above-described embodiment. For example, some components may be deleted from all the components shown in the embodiment. Furthermore, components from different embodiments may be appropriately combined.

Furthermore, the correspondence between the configuration according to the present invention and the configuration according to the above-described embodiment can be described as follows.
<Additional Notes>
a detection data input unit (12) into which vehicle detection data including the position of the vehicle (110) detected by scanning a target area (100) set for a road on which the vehicle (110) runs using a search wave is input from a vehicle detection device (5);
an image input unit (13) to which a frame image captured by an imaging device (6) installed in a direction to capture an image of the front of the vehicle (110) from the downstream side of the road is input;
a type determination unit (22) that processes a frame image captured by the imaging device (6) when the vehicle (110) is located at a type determination position (101) set on the road based on the vehicle detection data input to the detection data input unit (12), and determines the type of the vehicle (110) located at the type determination position (101),
The type of the vehicle (110) is a type classified based on the size and external shape of the vehicle, in the vehicle type determination device (1).

1, 1A... vehicle type determination device 5... radio wave radar device 6... imaging device 11, 11A... control unit 12... detection data input section 13... image input section 14... feature amount database (feature amount DB)
15: Output section 16: Release signal output section 21: Detection section 22: Type determination section 23: Traffic volume data generation section 24: Release signal generation section 100: Target area 101: Type of vehicle determination position 110: Vehicle 110a: Vehicle to be determined

Claims (9)

A vehicle detection device that detects vehicle detection data including the position of the vehicle by scanning a target area set for a road on which the vehicle travels with a search wave;
a detection data input unit to which the vehicle detection data detected by the vehicle detection device is input;
an imaging device installed in a direction to image the front of the vehicle from the downstream side of the road;
an image input unit to which a frame image captured by the imaging device is input;
A feature database in which front feature values for each type of vehicle are registered;
a type determination unit that, when the vehicle is located at a type determination position set on the road, processes a frame image captured by the imaging device to extract a feature amount of the front of the vehicle located at the type determination position based on the vehicle detection data input to the detection data input unit, and compares the extracted feature amount with the front feature amounts of the vehicle registered for each type in the feature amount database to determine the type of the vehicle ,
The vehicle type determination device, wherein the vehicle type is a type classified according to the size and shape of the vehicle. a detection unit that detects a type determination timing at which the vehicle was located at the type determination position based on the vehicle detection data input to the detection data input unit,
2 . The vehicle type determination device according to claim 1 , wherein the type determination unit determines the type of the vehicle by processing the frame images input to the image input unit, the frame images being captured at timings corresponding to the type determination timings. a determination unit that determines whether the vehicle is located at the type determination position based on the vehicle detection data input to the detection data input unit;
an imaging instruction unit that instructs the imaging device to capture an image when the determination unit determines that the vehicle is located at the type determination position,
The vehicle type determination device according to claim 1 , wherein the type determination unit determines the type of the vehicle by processing the frame images captured in response to an image capture instruction from the image capture instruction unit. the vehicle detection device repeats scanning within the target area at a set period and outputs the vehicle detection data at each period;
the imaging device is a video camera that captures frame images at a set frame rate,
a detection unit that detects a type determination timing at which the vehicle was located at the type determination position based on the vehicle detection data input to the detection data input unit,
2. The vehicle type determination device according to claim 1, wherein the type determination unit determines the type of the vehicle by processing the frame image input to the image input unit, the frame image having an image capture timing that is closest in time to the type determination timing detected by the detection unit. A vehicle type determination device according to any one of claims 1 to 4, comprising a traffic volume data generation unit that generates traffic volume data that aggregates traffic volume for each vehicle type determined by the type determination unit. the feature database is a database in which the vehicle type, the name of the vehicle type, and the feature of the front surface of the vehicle type are registered in association with each other,
The vehicle type determination device according to any one of claims 1 to 5, wherein the type determination unit compares the front feature values of the vehicle located at the type determination position with the front feature values of the vehicle type registered in the feature database, and also determines the name given to the vehicle. A vehicle type determination device according to any one of claims 1 to 6, wherein the type determination position is set within the range of the depth of field of the imaging device. a vehicle detection device that detects vehicle detection data including the position of a vehicle detected by scanning a target area set for a road on which the vehicle travels with a search wave, a detection data input unit to which the vehicle detection data detected by the vehicle detection device is input , an imaging device that is installed in a direction to capture an image of the front of the vehicle from the downstream side of the road, an image input unit to which a frame image captured by the imaging device is input , and a feature database in which feature values of the front of the vehicle for each type of vehicle are registered ,
a vehicle type determination method which executes a step of extracting a feature amount of a front side of the vehicle which is located at a type determination position set on the road based on the vehicle detection data inputted to the detection data input unit by processing a frame image captured by the imaging device when the vehicle is located at the type determination position , and collating the extracted feature amount with a feature amount of the front side of the vehicle which is registered for each type of the vehicle in the feature amount database to determine the type of the vehicle,
The vehicle type determination method, wherein the vehicle type is a type classified according to the size and shape of the vehicle. a vehicle detection device that detects vehicle detection data including the position of a vehicle detected by scanning a target area set for a road on which the vehicle travels with a search wave, a detection data input unit to which the vehicle detection data detected by the vehicle detection device is input , an imaging device that is installed in a direction to capture an image of the front of the vehicle from the downstream side of the road, an image input unit to which a frame image captured by the imaging device is input , and a feature database in which feature values of the front of the vehicle for each type of vehicle are registered;
a vehicle type determination program that executes a step of processing a frame image captured by the imaging device to extract a feature amount of a front side of the vehicle that was located at the type determination position set on the road based on the vehicle detection data input to the detection data input unit, and collating the extracted feature amount with a feature amount of the front side of the vehicle registered for each type of the vehicle in the feature amount database to determine the type of the vehicle ,
A vehicle type determination program, wherein the vehicle type is a type classified according to the size and shape of the vehicle.

Images