JP2014215719A - Vehicle model determination device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a vehicle model determination device which determines a vehicle model efficiently.SOLUTION: An inventive vehicle model determination device includes a stereoscopic camera, a first distance measurement part, a coordinate conversion part, a first vehicle height measurement part, a vehicle width measurement part, and a determination part. The first distance measurement part calculates a distance between the stereoscopic camera and a subject from images photographed by the stereoscopic camera. The coordinate conversion part generates a distance image indicating distances each between each point of the images photographed by the stereoscopic camera and a predetermined position on the basis of the distances calculated by the first distance measurement part. The first vehicle height measurement part measures a vehicle height of a vehicle photographed by the stereoscopic camera on the basis of the distance image. The vehicle width measurement part measures a vehicle width of the vehicle photographed by the stereoscopic camera on the basis of the distance image. The determination part determines a vehicle model of the vehicle on the basis of the vehicle width of the vehicle and the vehicle height of the vehicle.

Description

  Embodiments described herein relate generally to a vehicle type determination device.

  A vehicle type determination device that determines a vehicle type of a vehicle entering or leaving is installed at an entrance / exit such as an expressway. For example, the vehicle type determination device includes a sensor such as a camera or a laser scanner, and determines the vehicle type of the vehicle based on information from the camera or the laser scanner. Conventionally, the vehicle type determination device has a problem that it is necessary to use a plurality of expensive laser scanners or the like, or to use a plurality of cameras.

Japanese Patent Laid-Open No. 10-269489 JP 2000-298007 A

  In order to solve the above-described problems, a vehicle type determination device that efficiently determines a vehicle type is provided.

  According to the embodiment, the vehicle type determination device includes a stereo camera, a first distance measurement unit, a coordinate conversion unit, a first vehicle height measurement unit, a vehicle width measurement unit, and a determination unit. The first distance measuring unit calculates a distance between the stereo camera and the subject from an image captured by the stereo camera. The coordinate conversion unit generates a distance image indicating a distance between each part of the image captured by the stereo camera and a predetermined position based on the distance calculated by the first distance measurement unit. The first vehicle height measurement unit measures the vehicle height of the vehicle captured by the stereo camera based on the distance image. The vehicle width measurement unit measures the vehicle width of the vehicle captured by the stereo camera based on the distance image. The determination unit determines a vehicle type of the vehicle based on a vehicle width of the vehicle and a vehicle height of the vehicle.

FIG. 1 is a diagram illustrating a configuration example of a vehicle type determination device according to an embodiment. FIG. 2 is a block diagram illustrating an example of functions of the vehicle type determination device according to the embodiment. FIG. 3 is a diagram for explaining a method in which the first distance measuring unit according to the embodiment measures the distance. FIG. 4 is a diagram for explaining a method in which the first distance measurement unit according to the embodiment measures the distance. FIG. 5 is a diagram illustrating an example of a first distance image generated by the first distance measurement unit according to the embodiment. FIG. 6 is a diagram for explaining a method in which the first vehicle detection unit according to the embodiment determines a vehicle. FIG. 7 is a diagram illustrating a correspondence example according to the embodiment of the presence / absence and state of a vehicle in the area. FIG. 8 is a diagram illustrating an example of state transition according to the embodiment when the vehicle is moving forward. FIG. 9 is a diagram illustrating an example of state transition according to the embodiment when the vehicle is moving backward. FIG. 10 is a diagram for explaining a method by which the top surface coordinate conversion unit according to the embodiment generates a top surface distance image. FIG. 11 is a diagram illustrating an example of a first distance image according to the embodiment. FIG. 12 is a diagram illustrating an example of an upper surface distance image generated by the upper surface coordinate conversion unit according to the embodiment. FIG. 13 is a diagram for explaining a method by which the side coordinate conversion unit according to the embodiment generates a side distance image. FIG. 14 is a diagram illustrating an example of a side surface distance image generated by the side surface coordinate conversion unit according to the embodiment. FIG. 15 is a diagram for explaining an installation example of the laser scanner according to the embodiment. FIG. 16 is a diagram for explaining a method by which the distance conversion unit according to the embodiment calculates distance data. FIG. 17 is a diagram illustrating an example of a second distance image generated by the second distance measurement unit according to the embodiment. FIG. 18 is a diagram for explaining a method of detecting a tire by the circular figure detection unit according to the embodiment. FIG. 19 is a diagram illustrating an example of tire candidates extracted by the circular figure detection unit according to the embodiment. FIG. 20 is a diagram illustrating another example of tire candidates extracted by the circular graphic detection unit according to the embodiment. FIG. 21 is a flowchart for explaining an operation example of the control unit of the stereo camera according to the embodiment. FIG. 22 is a flowchart for explaining an operation example of the control unit of the laser scanner according to the embodiment. FIG. 23 is a flowchart for explaining an operation example of the vehicle type determination unit according to the embodiment. FIG. 24 is a diagram for describing another method in which the first vehicle detection unit according to the embodiment detects a vehicle. FIG. 25 is a block diagram illustrating a basic configuration example of the vehicle type determination device according to the embodiment.

  The vehicle type determination device according to the embodiment determines the vehicle type of the vehicle. The vehicle type is a classification of vehicles used on an expressway or a toll road. For example, the vehicle types are extra large vehicles, large vehicles, medium-sized vehicles, ordinary vehicles, and the like. The vehicle type determination device is installed at an entrance or exit of an expressway or a toll road, and determines the type of vehicle entering or entering the expressway or toll road. The determined vehicle type is used for determining a fee.

  The vehicle type determination device includes a stereo camera and a laser scanner. The vehicle type determination device determines the vehicle type, vehicle height, and number of axles of the vehicle using a stereo camera and a laser scanner. The vehicle type determination device determines the vehicle type of the vehicle from the vehicle type, vehicle height, and number of axles of the vehicle. Details will be described below.

FIG. 1 is a diagram schematically illustrating a configuration example of a vehicle type determination device 1 according to the embodiment.
As shown in FIG. 1, the vehicle type determination apparatus 1 includes a stereo camera 2, a control unit 3, a laser scanner 4, a control unit 5, a vehicle type determination unit 6, and the like. The stereo camera 2 is electrically connected to the control unit 3. The laser scanner 4 is electrically connected to the control unit 5. The vehicle type determination unit 6 is electrically connected to the control unit 3 and the control unit 5.

  The stereo camera 2 includes an upper camera installed at the upper part and a lower camera installed at the lower part. The upper camera and the lower camera are installed at a predetermined distance. For example, the upper camera and the lower camera are CCD cameras or the like.

  The stereo camera 2 is installed beside a road through which a vehicle (target vehicle) whose vehicle type determination device 1 determines the vehicle type passes. The stereo camera 2 captures an upper surface and a side surface of the target vehicle. That is, the upper camera and the lower camera photograph the upper surface and the side surface of the target vehicle. Therefore, the stereo camera 2 is installed so that the road through which the target vehicle passes can be seen obliquely downward. The stereo camera 2 transmits images taken by the upper camera and the lower camera to the control unit 3. Note that the arrangement relationship between the two cameras of the stereo camera 2 is not necessarily limited to a vertical vertical relationship, and may be an oblique vertical relationship.

  The control unit 3 processes an image transmitted from the stereo camera 2. For example, the control unit 3 calculates the distance between each part of the image and the stereo camera 2 based on the images from the upper camera and the lower camera. The control unit 3 generates a first distance image from the calculated distance from each unit. The control unit 3 determines the vehicle height, the vehicle width, and the like of the target vehicle based on the generated first distance image. The control unit 3 transmits information such as the determined vehicle height and the number of axles to the vehicle type determination unit 6.

  The control unit 3 includes a CPU, a memory, a communication unit, and the like. The control unit 3 may be a PC or the like. The CPU controls the entire control unit 3. The CPU executes each function of the control unit 3 by executing an application stored in a memory or the like. Further, some functions of the control unit 3 may be implemented by hardware. The control unit 3 will be described in detail later.

  The laser scanner 4 includes a laser light source and a light receiving unit. The laser light source irradiates a laser. The laser light source can be rotated vertically and can scan the laser one-dimensionally in the vertical direction. The light receiving unit observes the reflected light of the laser emitted from the laser light source.

  The control unit 5 controls the laser scanner 4. For example, the control unit 5 irradiates the target vehicle with laser using the laser scanner 4 and measures the distance from the target vehicle. Moreover, the control part 5 determines the vehicle height of the object vehicle, the number of axles, etc. from the distance data with the obtained object vehicle. The control unit 3 transmits information such as the determined vehicle height and the number of axles to the vehicle type determination unit 6.

  The control unit 5 includes a CPU, a memory, a communication unit, and the like. The control unit 5 may be a PC or the like. The CPU controls the entire control unit 5. The CPU executes each function of the control unit 5 by executing an application stored in a memory or the like. Further, some functions of the control unit 5 may be implemented by hardware. The controller 5 will be described in detail later.

  The vehicle type determination unit 6 receives information transmitted by the control unit 3 and the control unit 5, and determines the vehicle height, the vehicle width, and the number of axles of the target vehicle from the received information. The vehicle type determination unit 6 determines the vehicle type of the target vehicle from the determined vehicle height, vehicle width, and number of axles of the target vehicle.

  The vehicle type determination unit 6 includes a CPU, a memory, a communication unit, and the like. The vehicle type determination unit 6 may be a PC or the like. The CPU controls the entire vehicle type determination unit 6. The CPU executes each function of the vehicle type determination unit 6 by executing an application stored in a memory or the like. Further, some functions of the vehicle type determination unit 6 may be implemented by hardware. The vehicle type determination unit 6 will be described in detail later.

The control unit 3, the control unit 5, and the vehicle type determination unit 6 may be a single PC. The structure of the control part 3, the control part 5, and the vehicle type determination part 6 is not limited to a specific structure.
In addition, a vehicle C exists as a target vehicle on which the vehicle type determination device 1 determines the vehicle type on the road.

Next, functional examples of each part of the vehicle type determination device 1 will be described.
FIG. 2 is a block diagram for illustrating an example of functions of the vehicle type determination device 1.
First, a function example of the control unit 3 of the stereo camera 2 will be described.

  The control unit 3 includes an image conversion unit 21, a first distance measurement unit 22, a first vehicle detection unit 23, an upper surface coordinate conversion unit 24, a vehicle width measurement unit 25, a side coordinate conversion unit 26, a vehicle height measurement unit 27, and the like. .

The image conversion unit 21 is connected to the stereo camera 2 and acquires image data (hereinafter simply referred to as an image) from the upper camera and the lower camera of the stereo camera 2. The image conversion unit 21 performs coordinate alignment of two acquired images, lens distortion correction, and the like. When the data transmitted by the stereo camera 2 is analog data, the image conversion unit 21 converts the acquired analog data into digital data. In this case, the image conversion unit 21 may be hardware such as an A / D converter.
The image conversion unit 21 transmits images from the upper camera and the lower camera to the first distance measurement unit 22.

The first distance measuring unit 22 measures the distance between the subject and the stereo camera 2 based on the image obtained from the upper camera and the image obtained from the lower camera.
FIG. 3 is a diagram illustrating a method in which the first distance measuring unit 22 measures a distance from a part of the target vehicle.
First, the first distance measuring unit 22 specifies a corresponding point from an image (upper image) taken by the upper camera and an image (lower image) taken by the lower camera. For example, the first distance measuring unit 22 specifies a portion of the upper image and a portion of the lower image that show the same point by pattern matching or the like. When both parts that reflect the same point are specified, the first distance measuring unit 22 calculates the parallax d between the specified upper image part and lower image part. When the parallax d is calculated, the first distance measuring unit 22 calculates the distance from the stereo camera 2 to the point based on the parallax d.

FIG. 4 is a diagram illustrating a relationship between the distance calculated by the first distance measuring unit 22 and the parallax d.
As shown in FIG. 4, the first distance measuring unit 22 determines that the distance is closer as the parallax d is larger, and the distance is longer as the parallax d is smaller.
In addition, the first distance measurement unit 22 generates a first distance image that is color-coded based on the distance. The distance image is an image indicating the distance between the stereo camera 2 and the subject by color.
That is, the first distance measurement unit 22 sets the color corresponding to the calculated distance as the color of the pixel where the point is located. The first distance measuring unit 22 calculates distances at all points of the image captured by the stereo camera 2, and sets the color corresponding to the calculated distance as the color of the pixel at the point. The first distance measuring unit 22 generates a first distance image from the color of each point. Note that the first distance measurement unit 22 may generate a first distance image that indicates the distance by color shading. For example, the first distance measuring unit 22 may darken a point with a long distance and lighten a point with a short distance.

  FIG. 5 shows an example of a first distance image generated by the first distance measuring unit 22.

  In the example shown in FIG. 5, the stereo camera 2 captures a front part of a normal vehicle and a rear part of a truck. The stereo camera 2 captures the road surface between the front part of the ordinary vehicle and the rear part of the truck.

  The image 51 is an original image taken by the stereo camera 2. The distance image 52 is a distance image generated by the first distance measuring unit 22 based on the image 51. The distance image 52 is darker as the distance is longer, and lighter as the distance is shorter.

Since the stereo camera 2 is photographing the road obliquely from above, the distance between the stereo camera 2 and the road surface becomes farther away as it goes farther (upward in the image). For this reason, the color of the central portion of the distance image 52 (that is, the road surface) becomes darker as it goes farther. Further, a portion where the vehicle is present (that is, a front portion of a normal vehicle and a rear portion of a truck) is close to the stereo camera 2. For this reason, the portion of the distance image 52 where the vehicle is present is light in color.
The first distance measurement unit 22 transmits the generated first distance image to the first vehicle detection unit 23, the upper surface coordinate conversion unit 24, and the side surface coordinate conversion unit 26.

The first vehicle detection unit 23 detects the vehicle from the first distance image transmitted by the first distance measurement unit 22.
FIG. 6 is a diagram for explaining a method in which the first vehicle detection unit 23 detects a vehicle.

  For example, the first vehicle detection unit 23 divides the distance image 52 into an area A1 (distance image 52a), an area A2 (distance image 52b), and an area A3 (distance image 52c). The first vehicle detection unit 23 determines whether there is an object closer to each area than the distance to the road surface.

  That is, the 1st vehicle detection part 23 acquires the distance image of the road in a state without a vehicle previously. For example, when the vehicle type determination device 1 is installed, the operator may set a road distance image (road distance image) in a state where there is no vehicle in the vehicle detection unit 23. The 1st vehicle detection part 23 compares the 1st distance image which the 1st distance measurement part 22 transmits, and a road distance image, and detects a lighter color than a road distance image from a 1st distance image. A point lighter than the road distance image indicates a point closer to the stereo camera 2 than the road. Therefore, the first vehicle detection unit 23 determines that the vehicle is in an area that is lighter in color than the road distance image in the first distance image.

  As described above, the first vehicle detection unit 23 divides the first distance image into three regions. The first vehicle detection unit 23 determines whether a vehicle is included in each region. In the example illustrated in FIG. 6, the first vehicle detection unit 23 determines that there is an object closer to the distance image 52a and the distance image 52c than the road. Therefore, the first vehicle detection unit 23 determines that there is a vehicle in the distance image 52a and the distance image 52c.

The 1st vehicle detection part 23 determines the state of a road from the presence or absence of the vehicle of each area.
FIG. 7 is a diagram illustrating a correspondence example between the presence / absence of vehicles in each area and the state of the road.
In the example illustrated in FIG. 7, the first vehicle detection unit 23 determines that the state of the road is “no vehicle” (that is, “S0”) in which there is no vehicle on the road when there is no vehicle in the areas A1 to A3. judge. The first vehicle detection unit 23 is an “entrance” (ie, “S1”) in which the vehicle enters the road state when there is a vehicle in the area A1 and no vehicles in the areas A2 and A3. Is determined. Further, when there is no vehicle in the areas A1 and 2 and there is a vehicle in the area A3, the first vehicle detection unit 23 determines that the road state is “exit” (ie, “S3”) in which the vehicle exits the road. To do. Further, the first vehicle detection unit 23 performs “two approach” (that is, “S4”) in which the two vehicles approach each other when the area A1 and 3 have vehicles and the area A2 has no vehicles. Judge that there is. In other cases than the above, the first vehicle detection unit 23 determines that the state of the road is “staying” (that is, “S2”) where the vehicle is staying on the road.

The first vehicle detection unit 23 determines the progress of the vehicle on the road from the transition of the road state.
FIG. 8 shows an example of the transition of the road state when the vehicle is traveling on the road.
As shown in FIG. 8, when the vehicle is traveling on the road, the state of the road is that the vehicle enters the road from “no vehicle” (that is, “S0”) where there is no vehicle on the road. The process proceeds to “Enter” (ie, “S1”). Furthermore, the state of the road shifts from “stay” (ie, “S2”) where the vehicle stays on the road to “leave” (ie, “S3”) where the vehicle leaves. Subsequently, the state of the road returns to “no vehicle” (that is, “S0”) where there is no vehicle on the road.

When vehicles on the road are congested and the inter-vehicle distance is short, the state of the road is “retention” in which the vehicle stays on the road from “entrance” (ie, “S1”) where the vehicle enters the road. That is, after shifting to “S2”), “stay” and “two approaches” (ie, “S4”) in which two vehicles approach each other are repeated.
When the state of the road shifts from “exit” (ie, “S3”) to “no vehicle” (ie, “S0”), the first vehicle detection unit 23 determines that the vehicle on the road has advanced. Further, the state of the road is “no vehicle” (ie, “S0”), “approach” (ie, “S1”), “stay” (ie, “S2”), “two approaching” (ie, “S4”). ”), And then the transition from“ approaching two vehicles ”(ie,“ S4 ”) to“ staying ”(ie,“ S2 ”), the first vehicle detection unit 23 causes the vehicle on the road to travel. Is determined. When it is determined that the vehicle on the road has traveled, the first vehicle detection unit 23 transmits detection information indicating that the vehicle has traveled to the vehicle detection integration unit 42.

FIG. 9 shows an example of the transition of the road situation when the vehicle is moving backward on the road.
As shown in FIG. 9, when the vehicle is moving back on the road, the road condition shifts from “no vehicle” (ie, “S0”) to “exit” (ie, “S3”). Furthermore, the state of the road shifts from “Stay” (ie, “S2”) to “Enter” (ie, “S1”). Subsequently, the state of the road returns to “no vehicle” (ie, “S0”).

When vehicles on the road are congested and the inter-vehicle distance is short, the state of the road changes from “exit” (ie, “S3”) to “stay” (ie, “S2”), and then “stay”. Repeat “2 approaches” (ie, “S4”).
When the state of the road shifts from “entry” (ie, “S1”) to “no vehicle” (ie, “S0”), the first vehicle detection unit 23 determines that the vehicle on the road has moved backward. In addition, the state of the road is “no vehicle” (ie, “S0”), “exit” (ie, “S3”), “stay” (ie, “S2”), “two approaching” (ie, “S4”). )) And then, when the vehicle moves from “approaching two vehicles” (ie, “S4”) to “staying” (ie, “S2”), the first vehicle detection unit 23 moves the vehicle on the road backward. Is determined. When it is determined that the vehicle on the road has moved backward, the first vehicle detection unit 23 transmits detection information indicating that the vehicle has moved back to the vehicle detection integration unit 42.

  The upper surface coordinate conversion unit 24 generates a distance image (upper surface distance image) when the road is viewed from directly above, from the first distance image generated by the first distance measurement unit 22. The upper surface distance image is an image that indicates the distance between a predetermined position directly above the road and each part of the image by color.

FIG. 10 is a diagram illustrating a method in which the upper surface coordinate conversion unit 24 generates an upper surface distance image.
As shown in FIG. 10, the angle between the stereo camera 2 viewing the road and the line extending vertically from the road is ω1. The upper surface coordinate conversion unit 24 acquires ω1 in advance. For example, when the vehicle type determination device 1 is installed, the operator may input the value of ω1 to the upper surface coordinate conversion unit 24.

  The upper surface coordinate conversion unit 24 generates an upper surface distance image from the first distance image generated by the first distance measurement unit 22 using the value of ω1. For example, the upper surface coordinate conversion unit 24 generates an upper surface distance image by affine transforming the first distance image using the value of ω1. The method by which the upper surface coordinate conversion unit 24 generates the upper surface distance image is not limited to a specific method.

FIG. 11 is an example of a distance image generated by the first distance measuring unit 22.
As shown in FIG. 11, the first distance image shows the upper surface 71, the side surface 72, and the wheels 73 of the vehicle. Since the stereo camera 2 looks at the road obliquely from above, the first distance image reflects the upper and side surfaces of the vehicle C traveling on the road. As a result, the upper surface 71, the side surface 72, and the wheels 73 of the vehicle C are reflected in the first distance image.

FIG. 12 is an example of an upper surface distance image generated by the upper surface coordinate conversion unit 24.
The upper surface distance image is an image indicating the distance between a predetermined position directly on the road and each part. Therefore, in the upper surface distance image, the upper surface 81 and the wheels 83 of the vehicle C are reflected without the side surface portion of the vehicle C being reflected. Further, as shown in FIG. 11, no distance data exists for the portion behind the vehicle C. Accordingly, as shown in FIG. 12, the top image includes a region 84 without an image. That is, the region 84 is a portion behind the vehicle C.
The upper surface coordinate conversion unit 24 transmits the generated upper surface distance image to the vehicle width measurement unit 25.

The vehicle width measurement unit 25 measures the vehicle width of the vehicle based on the upper surface distance image generated by the upper surface coordinate conversion unit 24. For example, the vehicle width measurement unit 25 determines that a region (a region where the distance is close) that is lighter than the upper surface distance image of the road surface is a region where the vehicle is present (a vehicle region). That is, the vehicle width measurement unit 25 determines that the region surrounded by the road surface and the region without an image (for example, the region 84) is a vehicle region. When determining the area where the vehicle is, the vehicle width measuring unit 25 measures the number of pixels of the height of the vehicle area (that is, the distance from the road in front to the area where there is no image), and the vehicle width is calculated from the measured number of pixels. Is calculated. The method by which the vehicle width measuring unit 25 measures the width of the vehicle is not limited to a specific method.
When the vehicle width is calculated, the vehicle width measurement unit 25 transmits vehicle width information indicating the measured vehicle width to the vehicle width measurement integration unit 43. The vehicle width information may include a distance from the laser scanner 4 to the vehicle (a distance from a predetermined position on the road to the vehicle area).

  The side coordinate conversion unit 26 generates a distance image (side distance image) obtained by viewing the road from the side from the first distance image generated by the first distance measurement unit 22. The side distance image is an image indicating the distance between a predetermined position directly beside the road and each part of the image by color.

FIG. 13 is a diagram illustrating a method in which the side surface coordinate conversion unit 26 generates a side surface distance image.
As shown in FIG. 13, the angle between the stereo camera 2 looking at the road and the line extending horizontally with the road is ω2. The side coordinate conversion unit 26 acquires ω2 in advance. For example, when the vehicle type determination device 1 is installed, the operator may input the value of ω <b> 2 to the side surface coordinate conversion unit 26. Further, the side surface coordinate conversion unit 26 may calculate ω2 (90 ° −ω1) based on ω1.

  The side coordinate conversion unit 26 generates a side distance image from the first distance image generated by the first distance measurement unit 22 using the value of ω2. For example, the side surface coordinate conversion unit 26 generates a side surface distance image by performing affine transformation of the first distance image using the value of ω2. The method by which the side coordinate conversion unit 26 generates the side distance image is not limited to a specific method.

FIG. 14 is an example of a side surface distance image generated by the side surface coordinate conversion unit 26.
The side surface distance image shown in FIG. 14 is an image generated based on the first distance image shown in FIG.
The side distance image is an image indicating the distance between a predetermined position directly on the road and each part. Therefore, in the side distance image, the upper surface portion of the vehicle C is not reflected, and the side surface 92 and the wheels 93 of the vehicle C are reflected. In the vicinity of the vehicle area, the distance data may or may not be present. Whether there is distance data around the vehicle area depends on the situation around the road.
The side coordinate conversion unit 26 transmits the generated side distance image to the first vehicle height measurement unit 27.

The first vehicle height measurement unit 27 measures the vehicle height of the vehicle based on the side surface distance image generated by the side surface coordinate conversion unit 26. For example, the first vehicle height measurement unit 27 determines that an area that is lighter than the side distance image when no vehicle is present (an area where the distance is close) is the vehicle area. That is, the first vehicle height measurement unit 27 determines that the area surrounded by the road surface or landscape area and the area without an image is a vehicle area. When determining the area where the vehicle is, the first vehicle height measurement unit 27 measures the number of pixels of the height of the vehicle area (that is, the distance from the road in front to the area (or landscape area) where there is no image), The vehicle height is calculated from the measured number of pixels. The method by which the vehicle height measuring unit 27 measures the vehicle height is not limited to a specific method.
When the vehicle height is calculated, the vehicle height measurement unit 27 transmits first vehicle height information indicating the measured vehicle height to the vehicle height measurement integration unit 44.

Next, the laser scanner 4 and the control unit 5 will be described.
FIG. 15 is a diagram for explaining an installation example of the laser scanner 4.
As shown in FIG. 15, the laser scanner 4 is installed beside the road. The laser light source of the laser scanner 4 irradiates the side surface of the vehicle C with laser. The laser light source of the laser scanner 4 rotates in the vertical direction. As a result, the laser light source of the laser scanner 4 scans the laser on the broken line 101.

The laser irradiated by the laser light source of the laser scanner 4 is reflected by the object.
The light receiving unit of the laser scanner 4 receives the reflected light reflected by the object.

  The laser scanner 4 measures the distance between the laser scanner 4 and the object based on the time difference in which the light receiving unit receives the reflected light after the laser light source irradiates the laser. When the distance is measured, the laser scanner 4 transmits information indicating the measured distance to the distance conversion unit 31 of the control unit 5. Further, the laser scanner 4 transmits information indicating the angle of the laser light source when the distance is measured to the distance conversion unit 31. The laser scanner 4 transmits information indicating the distance and angle to the distance conversion unit 31 at predetermined intervals while the laser light source rotates the movable range.

The distance conversion unit 31 calculates the distance between the distance measurement reference plane and the object based on the distance between the laser scanner 4 and the object.
FIG. 16 is a diagram illustrating a method in which the distance conversion unit 31 calculates the distance between the distance measurement reference plane and the object.

  As shown in FIG. 16, the distance measurement reference plane 112 is a plane that passes through the laser scanner 4 and is perpendicular to the road. The distance conversion unit 31 calculates a distance 114 between the object and the distance measurement reference plane 112. Here, it is assumed that the laser scanner 4 has measured the distance 113 when the angle of the laser light source of the laser scanner 4 is θ.

The distance conversion unit 31 calculates the distance between the distance measurement reference plane 112 and the object from the angle of the laser light source and the distance measured by the laser scanner. That is,
Distance 114 = cos θ × distance 113
Using this, the distance conversion unit 31 calculates the distance 114. When the distance 114 is calculated, the distance conversion unit 31 transmits information indicating the calculated distance 114 to the distance measurement unit 32. Each time the distance conversion unit 31 receives the information indicating the distance and the angle from the laser scanner 4, the distance conversion unit 31 calculates the distance between the distance measurement reference plane 112 and the object and the information indicating the calculated distance to the second distance measurement unit 32. Send.

The second distance measurement unit 32 generates a second distance image from the distance between the distance measurement reference surface 112 and the object. Since the laser scanner 4 measures a one-dimensional distance in the vertical direction, the entire side surface of the vehicle C can be scanned when the object vehicle C passes in front of the laser scanner 4. When the laser scanner 4 scans the entire side surface of the vehicle C, the second distance measurement unit 32 can obtain information indicating the distance between each part on the side surface of the vehicle C and the distance measurement reference surface 112.
The second distance measurement unit 32 generates a second distance image indicating the distance between the distance measurement reference surface 112 and the object based on information indicating the distance between each part of the image and the distance measurement reference surface 112.

FIG. 17 shows an example of the second distance image generated by the second distance measuring unit 32.
The second distance image shown in FIG. 17 indicates the farther the darker the color, and the closer the lighter the color.
Since there is a road on the near side, in the second distance image, the color is light in the lower part and becomes darker as it goes to the upper part.

  The vehicle 121 and the vehicle 122 are examples of a second distance image through which a black sedan passes. The black paint used in the vehicle is less likely to reflect the laser emitted by the laser light source of the laser scanner 4. Therefore, the light receiving unit of the laser scanner 4 cannot receive the reflected light from the black paint, and the laser scanner 4 cannot measure the distances between the vehicle 121 and each part of the vehicle 122. Therefore, in the second distance image, the vehicle 121 and the vehicle 122 are not clearly shown as a whole.

  Vehicles 123 to 125 are examples of second distance images through which a non-black track has passed. Since the vehicles 123 to 125 do not use black paint, the lasers emitted from the laser light source of the laser scanner 4 are reflected. Therefore, the light receiving unit of the laser scanner 4 can receive the reflected light, and the laser scanner 4 can measure the distance from each unit of the vehicles 123 to 125. Therefore, in the second distance image, the vehicles 123 to 125 are clearly visible as a whole as compared with the vehicles 121 and 122.

The reason why the vehicle 125 is shorter than the vehicle 123 and the vehicle 124 is that the vehicle 125 passes in front of the laser scanner 4 at a higher speed than the vehicle 123 and the vehicle 124.
The second distance measurement unit 32 transmits the generated second distance image to the second vehicle detection unit 33, the circular figure detection unit 34, and the second vehicle height measurement unit 37. The second distance measurement unit 32 transmits distance information indicating the distance between the vehicle and the distance measurement reference plane 112 to the vehicle width measurement integration unit 43.

  The second vehicle detection unit 33 distinguishes one vehicle based on the second distance image generated by the second distance measurement unit 32. For example, when distinguishing a vehicle in which distance measurement is impossible (for example, a vehicle in which black paint is used), the second vehicle detection unit 33 determines that the region in which the distance cannot be measured is a region where the vehicle is present. judge. Moreover, when distinguishing the vehicle in which distance measurement is possible, the 2nd vehicle detection part 33 determines with the area | region which shows the distance nearer than the distance in the state without a vehicle being an area | region with a vehicle. The method by which the second vehicle detection unit 33 distinguishes vehicles is not limited to a specific method.

  When the second vehicle detection unit 33 detects the front end of the vehicle, the second vehicle detection unit 33 transmits a signal indicating that the vehicle has entered (an entry detection signal) to the vehicle detection integration unit 42. Further, when the second vehicle detection unit 33 detects the rear end of the vehicle, the second vehicle detection unit 33 transmits a signal (exit detection signal) indicating that the vehicle has exited to the vehicle detection integration unit 42.

The circular graphic detector 34 detects a circular graphic as a tire candidate based on the distance image generated by the second distance measuring unit 32.
FIG. 18 is a diagram for explaining a method in which the circular graphic detector 34 detects a circular graphic as a tire candidate.
As shown in FIG. 18, the laser scanner 4 measures the distance in the order of point 131, point 132, point 133, and point 134. The circular figure detection unit 34 extracts, as a feature point, a point where the distance changes to a predetermined value or more from the second distance image based on the distance measured by the laser scanner 4. In the example shown in FIG. 18, the circular graphic detector 34 determines that the distance between the point 132 and the point 133 changes by a predetermined value or more. The circular figure detection unit 34 extracts a point 133 (black spot) as a feature point.

  There is a certain gap between the tire and the vehicle body. Therefore, the distance from the laser scanner 4 changes stepwise between the tire heel and the tire periphery. As a result, tire wrinkles become a feature point.

  Further, since the tire is circular, the feature points along the tire ridge are circular or elliptical. Therefore, the circular figure detection unit 34 determines that a feature point group in which the feature points are arranged in a circle or an ellipse is a feature point group indicating a tire candidate.

  For example, the circular figure detection unit 34 generates an image in which n feature points are visualized. A coordinate value group of feature points is represented by {(x1, y1), (x2, y2),... (Xn, yn)}. The circular figure detection unit 34 adapts the coordinate value group of feature points while changing the parameters (a, b) with respect to (x, y) of the following elliptic equation (Equation 1), and the error is equal to or less than a predetermined value. It is determined that the coordinate group is a feature point group indicating a circular figure (that is, a tire candidate).

(Y−y0) / a ^ 2 + (x−x0) / b ^ 2 = 1 (1)
Note that the method by which the circular figure detection unit 34 extracts tire candidates is not limited to a specific method.
The circular figure detection unit 34 transmits a coordinate group of feature points extracted as a feature point group indicating a tire candidate to the tire detection unit 35.

The tire detection unit 35 detects a tire constituting the axle from the tire candidates extracted by the circular figure detection unit 34.
19 and 20 are diagrams illustrating an example of a coordinate group indicating tire candidates extracted by the circular figure detection unit 34. FIG.
Tire candidates shown in FIG. 19 are tires in contact with the ground (tires constituting the axle). Since the tire is in contact with the ground, there is no point where the distance changes stepwise at the lower part of the tire, and there is no feature point. Therefore, when the tire candidate indicates a tire (axle tire) constituting the axle, the feature point group indicating the tire candidate exists on the upper side of the circular figure and does not exist on the lower side.

  The tire candidate shown in FIG. 20 is a tire that is not in contact with the ground (a tire that does not constitute an axle (for example, a spare tire)) or a circular figure other than a tire. In this case, since the lower side of the circular figure is not in contact with the ground, there are characteristic points on the lower side of the circular figure. Therefore, when the tire candidate indicates other than the axle tire, the feature point group indicating the tire candidate is also present on the lower side of the circular figure.

  Therefore, the tire detection unit 35 determines that a tire candidate whose feature point is not on the lower side of the circular figure is an axle tire. Note that the method by which the tire detection unit 35 determines whether the tire candidate is an axle tire is not limited to a specific method. The tire detection unit 35 transmits a determination result of whether or not the tire candidate is an axle tire to the axle detection unit 36.

  The axle detection unit 36 detects the axle of the vehicle based on the determination result of the tire detection unit 35. That is, the axle detection unit 36 extracts the axle tire detected by the tire detection unit 35 from the tire candidates detected by the circular figure detection unit 34.

  When the axle tire is extracted, the axle detection unit 36 transmits a signal (axle detection signal) indicating that the axle tire has been extracted to the axle counting integration unit 41.

The second vehicle height measurement unit 37 measures the vehicle height based on the second distance image generated by the second distance measurement unit 32. For example, the second vehicle height measurement unit 37 determines that an area that is lighter in color than the distance image when there is no vehicle (an area where the distance is close) is the vehicle area. When the area where the vehicle is located is determined, the vehicle height measuring unit 37 measures the number of pixels at the height of the vehicle area (that is, the distance from the road in front to the area (or landscape area) where there is no image). The vehicle height is calculated from the number of pixels obtained. The method by which the vehicle height measuring unit 37 measures the vehicle height is not limited to a specific method.
The second vehicle height measurement unit 37 transmits second vehicle height information indicating the measured vehicle height to the vehicle height measurement integration unit 44.

Next, the vehicle type determination unit 6 will be described.
As shown in FIG. 2, the vehicle type determination unit 6 includes an axle count integration unit 41, a vehicle detection integration unit 42, a vehicle width measurement integration unit 43, a vehicle height measurement integration unit 44, and the like.

The axle count integration unit 41 determines the number of axles of the vehicle based on the approach detection signal and the exit detection signal transmitted by the second vehicle detection unit 33 and the axle detection signal transmitted by the axle detection unit 36. In other words, the axle count integration unit 41 counts the number of times the axle detection signal is received from the axle detection unit 36 after receiving the entry detection signal and before receiving the exit detection signal. The axle count integration unit 41 determines the counted number as the number of axles of the vehicle.
The axle count integration unit 41 transmits the determined number of axles to the determination unit 45.

The vehicle detection integration unit 42 detects the vehicle by integrating the detection information transmitted by the first vehicle detection unit 23 and the entry detection signal and the exit detection signal transmitted by the second vehicle detection unit 33. For example, the vehicle detection integration unit 42 determines that the vehicle has entered the vehicle type determination device 1 when receiving an entry detection signal from the second vehicle detection unit 33. In this case, the vehicle detection integration unit 42 identifies forward or backward movement of the vehicle from the detection information transmitted by the first vehicle detection unit 23. This is because the detection information generated by the first vehicle detection unit 23 is likely to cause an error due to noise generated in the stereo camera 2, and the timing at which the vehicle enters cannot be measured accurately. Moreover, it is because the 2nd vehicle detection part 33 cannot detect the advance or retreat of a vehicle.
The vehicle detection integration unit 42 transmits a signal indicating the entry of the vehicle to the determination unit 45. Further, the vehicle detection integration unit 42 transmits a signal indicating forward or backward movement of the vehicle to the determination unit 45.

The vehicle width measurement integration unit 43 determines the vehicle width based on the vehicle width information transmitted by the vehicle width measurement unit 25 and the distance information transmitted by the second distance measurement unit 32. For example, the vehicle width measurement integration unit 43 acquires the distance between the laser scanner 4 and the vehicle from the distance information. The vehicle width measurement integration unit 43 corrects the vehicle width indicated by the vehicle width information based on the acquired distance (vehicle distance). In other words, the vehicle width measurement integration unit 43 determines the vehicle width by replacing the position on the near side of the vehicle area detected by the vehicle width measurement unit 25 with a position away from the laser scanner 4 by the vehicle distance. This is because the vehicle width measured by the vehicle width measuring unit 25 is likely to cause an error due to noise generated in the stereo camera 2, and it is necessary to correct the position of the vehicle region to increase the accuracy.
The vehicle width measurement integration unit 43 transmits information indicating the determined vehicle width to the determination unit 45.

The vehicle height measurement integration unit 44 determines the vehicle height based on the first vehicle height information transmitted by the first vehicle height measurement unit 27 and the second vehicle height information transmitted by the second vehicle height measurement unit 37. For example, the vehicle height measurement integration unit 44 determines the maximum value of the vehicle height indicated by the first vehicle height information and the vehicle height indicated by the second vehicle height information as the vehicle height. This is because when the vehicle is coated with black paint, the laser scanner 4 cannot accurately measure the distance, and the second vehicle height measuring unit 37 may not be able to accurately measure the vehicle height. is there. This is also because the first vehicle height measurement unit 27 using the stereo camera 2 may cause a measurement error.
The vehicle height measurement integration unit 44 transmits information indicating the determined vehicle height to the determination unit 45.

  The determination unit 45 determines the vehicle type from the number of axles, vehicle height, and vehicle width of the vehicle determined by each unit. For example, the determination unit 45 determines that the vehicle is a large vehicle if the number of axles is four or more. The determination unit 45 determines that the vehicle is a large vehicle if the number of axles is three or less and the vehicle height or vehicle width is equal to or greater than a predetermined value. Otherwise, the determination unit 45 determines that the vehicle is a normal vehicle. The vehicle type and vehicle type requirements determined by the determination unit 45 are not limited to a specific configuration.

Next, an operation example of the control unit 3 of the stereo camera 2 will be described.
FIG. 21 is a flowchart for explaining an operation example of the control unit 3 of the stereo camera 2.
First, the control unit 3 acquires an image from the stereo camera 2 (step S11). When the image is acquired, the control unit 3 converts the image using the image conversion unit 21 (step S12).

  When the image is converted, the control unit 3 generates a first distance image using the first distance measurement unit 22 (step S13). When the first distance image is generated, the control unit 3 determines whether the vehicle has moved forward using the first vehicle detection unit 23 (step S14).

When it is determined that the vehicle is not moving forward (NO in step S14), the control unit 3 returns to step S11.
If it determines with having advanced the vehicle (step S14, YES), the control part 3 will transmit the signal which shows that the vehicle advanced to the vehicle type determination part 6 (step S15).

  When a signal indicating that the vehicle has moved forward is transmitted to the vehicle type determination unit 6, the control unit 3 generates an upper surface distance image using the upper surface coordinate conversion unit 24 (step S16). When the top surface distance image is generated, the control unit 3 measures the vehicle width using the vehicle width measurement unit 25 and generates vehicle width information (step S17).

When the vehicle width is measured, the control unit 3 generates a side surface distance image using the side surface coordinate conversion unit 26 (step S18). When the side distance image is generated, the control unit 3 measures the vehicle height using the first vehicle height measurement unit 27 and generates first vehicle height information (step S19).
When the vehicle height is measured, the control unit 3 transmits the vehicle width information and the first vehicle height information to the vehicle type determination unit 6 (step S20). If vehicle width information and 1st vehicle height information are transmitted to the vehicle type determination part 6, the control part 5 will complete | finish operation | movement.

Next, an operation example of the control unit 5 of the laser scanner 4 will be described.
FIG. 22 is a flowchart for explaining an operation example of the control unit 5 of the laser scanner 4.
First, the control unit 5 acquires scan data from the laser scanner 4 (step S31). When the scan data is acquired, the control unit 5 calculates the distance from the distance measurement reference plane 112 using the distance conversion unit 31 (step S32).

  When the distance to the distance measurement reference plane 112 is calculated, the control unit 5 generates a second distance image using the second distance measurement unit 32 (step S33). That is, the control unit 5 additionally writes an image to the second distance image already generated by the control unit 5 based on the distance sequentially calculated in step S32.

  When the second distance image is generated, the control unit 5 determines whether a vehicle approach detection signal has already been transmitted (step S34). If it determines with not having transmitted the approach detection signal (step S34, NO), the control part 5 will determine whether the front-end | tip of the vehicle was detected using the 2nd vehicle detection part 33 (step S35).

If it determines with not detecting the front-end | tip of a vehicle (step S35, NO), the control part 5 will return to step S31.
If it determines with having detected the front-end | tip of a vehicle (step S35, YES), the control part 5 will transmit an approach detection signal to the vehicle detection integration part 42 (step S36).

  When it determines with having already transmitted the approach detection signal (step S34, YES), or when the approach detection signal is transmitted to the vehicle detection integration part 42 (step S36), the control part 5 uses the circular figure detection part 34. Then, tire candidates are extracted from the second distance image (step S37). When the tire candidate is extracted, the control unit 5 extracts the axle tire using the tire detection unit 35 (step S38).

  When the axle tire is extracted, the control unit 5 determines whether there is an axle tire using the axle detection unit 36 (step S39). When it is determined that there is an axle tire (step S39, YES), the control unit 5 transmits an axle detection signal to the vehicle type determination unit 6 (step S40).

  When it is determined that there is no axle tire (step S39, NO), or when an axle detection signal is transmitted to the vehicle type determination unit 6 (step S40), the control unit 5 uses the second vehicle detection unit 33 to It is determined whether the rear end has been detected (step S41). If it is determined that the rear end of the vehicle is not detected (step S41, NO), the control unit 5 returns to step S31.

  If it determines with having detected the rear end of the vehicle (step S41, YES), the control part 5 will transmit an exit detection signal to the vehicle type determination part 6 (step S42). When the exit detection signal is transmitted to the vehicle type determination unit 6, the control unit 5 measures the vehicle height using the second vehicle height measurement unit 37 and generates second vehicle height information (step S43). When the second vehicle height information is generated, the control unit 5 transmits the generated second vehicle height information to the vehicle type determination unit 6 (step S44). When the second vehicle height information is transmitted to the vehicle type determination unit 6, the control unit 5 uses the second distance measurement unit 32 to transmit distance information indicating the distance between the vehicle and the distance measurement reference plane 112 to the vehicle type determination unit 6. (Step S45). When the distance information is transmitted to the vehicle type determination unit 6, the control unit 5 ends the operation.

Next, an operation example of the vehicle type determination unit 6 will be described.
FIG. 23 is a flowchart for explaining an operation example of the vehicle type determination unit 6.
First, the vehicle type determination unit 6 determines whether an entry detection signal has been received from the second vehicle detection unit 33 of the control unit 5 using the vehicle detection integration unit 42 (step S51). If it determines with not having received the approach detection signal (step S51, NO), the vehicle type determination part 6 will return to step S51.

  If it is determined that an approach detection signal has been received (step S51, YES), the vehicle type determination unit 6 determines whether an axle detection signal has been received from the axle detection unit 36 of the control unit 5 using the axle count integration unit 41 (step S51). S52).

  When it is determined that the axle detection signal has been received (step S52, YES), the vehicle type determination unit 6 counts up the number of axles (step S53). Here, the number of axles is a counter that counts the number of axles of the target vehicle. The number of axles stores “0” in the initial state.

  When it is determined that the axle detection signal is not received (step S52, NO), or when the number of axles is counted up (step S53), the vehicle type determination unit 6 uses the vehicle detection integration unit 42 to control the control unit 5 It is determined whether an exit detection signal is received from the second vehicle detection unit 33 (step S54).

If it is determined that the exit detection signal has not been received (step S54, NO), the vehicle type determination unit 6 returns to step S52.
When it is determined that the exit detection signal has been received (step S54, YES), the vehicle type determination unit 6 receives first vehicle height information, second vehicle height information, vehicle width information, distance information, and the like from each unit (step S55). .

  When each information is received, the vehicle type determination unit 6 determines the vehicle width based on the vehicle width information and the distance information using the vehicle width measurement integration unit 43 (step S56). When the vehicle width is determined, the vehicle type determination unit 6 determines the vehicle height based on the first vehicle height information and the second vehicle height information using the vehicle height measurement integration unit 44 (step S57).

  When the vehicle height is determined, the vehicle type determination unit 6 uses the determination unit 45 to determine the vehicle type based on the vehicle width, the vehicle height, the number of axles, and the like (step S58). When the vehicle type is determined, the vehicle type determination unit 6 ends the operation.

The vehicle type determination unit 6 may transmit information indicating the determined vehicle type to another device that determines a toll fee.
Further, when the vehicle type determination unit 6 determines that the vehicle is moving backward based on the detection information transmitted by the first vehicle detection unit 23 of the control unit 3, the vehicle type determination unit 6 transmits information indicating an error to another device. May be.

  Further, the control unit 5 may detect tire tires by extracting tire candidates after the vehicle leaves.

Next, another method in which the first vehicle detection unit 23 detects the vehicle will be described.
FIG. 24 is a diagram for explaining another method in which the first vehicle detection unit 23 detects a vehicle.

  In this method, the first distance measuring unit 22 measures a distance from a point (a point 141 in FIG. 24) where the laser light source of the laser scanner 4 irradiates a laser. The first distance measuring unit 22 detects the vehicle based on the distance between the point 141 and the stereo camera 2. For example, the first distance measurement unit 22 determines whether a change has occurred in the distance between the point 141 and the stereo camera 2. When it is determined that the distance between the point 141 and the stereo camera 2 has changed, the first distance measurement unit 22 may determine that the vehicle has entered.

  The vehicle type determination apparatus configured as described above can measure the vehicle width and vehicle height using a stereo camera. In addition, the vehicle type determination device can measure the number of axles using a laser scanner, and can correct the vehicle width, the vehicle height, and the like. As a result, the vehicle type determination device can effectively determine the vehicle type.

In addition, the basic concept of the vehicle type determination device 1 according to the embodiment is as follows.
FIG. 25 is a block diagram illustrating a configuration example of the vehicle type determination device 1 according to the embodiment.
As shown in FIG. 25, the configuration example of the vehicle type determination device 1 includes a stereo camera 2, a first distance measurement unit 22, a coordinate conversion unit 201, a vehicle width measurement unit 25, a first vehicle height measurement unit 27, and a vehicle type determination unit 6. Etc. The first distance measuring unit 22 calculates the distance between the stereo camera 2 and the subject from the video captured by the stereo camera 2. The coordinate conversion unit 201 generates a distance image indicating the distance between each part of the image captured by the stereo camera 2 and a predetermined position based on the distance calculated by the first distance measurement unit. The coordinate conversion unit 201 includes an upper surface coordinate conversion unit 24, a side surface coordinate conversion unit 26, and the like. The vehicle width measuring unit 25 measures the vehicle width of the vehicle captured by the stereo camera 2 based on the first distance image. The first vehicle height measurement unit 27 measures the vehicle height of the vehicle captured by the stereo camera 2 based on the first distance image. The vehicle type determination unit 6 determines the vehicle type of the vehicle based on the vehicle width and the vehicle height of the vehicle.

  Moreover, since the vehicle type determination apparatus 1 operates when the CPU executes a program, the operation of the vehicle type determination apparatus 1 is not restricted by the connection order of the blocks as shown in FIG. The program calculates the distance between each part of the image captured by the stereo camera 2 and a predetermined position based on the code for calculating the distance between the stereo camera 2 and the subject from the video captured by the stereo camera 2 and the calculated distance. A code for generating a distance image to be shown, a code for measuring the vehicle height of the vehicle captured by the stereo camera 2 based on the generated distance image, and a vehicle of the vehicle captured by the stereo camera 2 based on the generated distance image A code for measuring the width, and a code for determining a vehicle type based on the vehicle width and the vehicle height of the vehicle.

  The vehicle type determination method of the vehicle type determination apparatus 1 calculates the distance between the stereo camera 2 and the subject from the video captured by the stereo camera 2, and based on the calculated distance, each part of the image captured by the stereo camera 2. A distance image indicating a distance to a predetermined position is generated, the vehicle height of the vehicle captured by the stereo camera 2 is measured based on the generated distance image, and the stereo camera 2 captures the image based on the generated distance image. The vehicle width of the vehicle is measured, and the vehicle type of the vehicle is determined based on the vehicle width of the vehicle and the vehicle height.

  Although several embodiments of the present 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, replacements, and changes can be made without departing from the scope of the invention. These embodiments and modifications thereof are included in the scope and gist of the invention, and are included in the invention described in the claims and the equivalents thereof.

  DESCRIPTION OF SYMBOLS 1 ... Vehicle type determination apparatus, 2 ... Stereo camera, 3 ... Control part, 4 ... Laser scanner, 5 ... Control part, 6 ... Vehicle type determination part, 22 ... 1st distance measurement part, 23 ... 1st vehicle detection part, 24 ... Upper surface coordinate conversion unit, 25 ... vehicle width measurement unit, 26 ... side coordinate conversion unit, 27 ... first vehicle height measurement unit, 32 ... second distance measurement unit, 33 ... second vehicle detection unit, 34 ... circular figure detection unit 35 ... Tire detection unit 36 ... Axle detection unit 37 ... Second vehicle height measurement unit 41 ... Axle counting integration unit 42 ... Vehicle detection integration unit 43 ... Vehicle width measurement integration unit 44 ... Vehicle height measurement integration Part, 45 ... determination part, 101 ... coordinate conversion part, C ... vehicle.

Claims (8)

In the vehicle type determination device for determining the vehicle type of the vehicle,
A stereo camera,
A first distance measuring unit that calculates a distance between the stereo camera and a subject from an image captured by the stereo camera;
A coordinate conversion unit that generates a distance image indicating a distance between each part of the image captured by the stereo camera and a predetermined position based on the distance calculated by the first distance measurement unit;
A first vehicle height measuring unit that measures the vehicle height of the vehicle captured by the stereo camera based on the distance image;
A vehicle width measuring unit that measures the vehicle width of the vehicle captured by the stereo camera based on the distance image;
A determination unit that determines a vehicle type of the vehicle based on a vehicle width of the vehicle and a vehicle height of the vehicle;
A vehicle type determination device. The stereo camera simultaneously captures the upper and side surfaces of the vehicle,
The coordinate converter is
An upper surface coordinate conversion unit that generates an upper surface distance image indicating a distance between a predetermined position directly above the vehicle and each part of the image;
A side coordinate conversion unit that generates a side distance image indicating a distance between a predetermined position directly beside the vehicle and each part of the image;
With
The vehicle width measurement unit measures the vehicle width of the vehicle based on the upper surface distance image generated by the upper surface coordinate conversion unit,
The vehicle height measurement unit measures the vehicle height of the vehicle based on the side distance image generated by the side coordinate conversion unit.
The vehicle type determination apparatus according to claim 1. further,
A laser scanner for measuring the distance to the side of the vehicle;
An axle detection unit for detecting an axle of the vehicle based on a distance measured by the laser scanner;
An axle counting integration unit that counts the axles detected by the axle detection unit;
With
The determination unit determines the vehicle type of the vehicle based on the number of axles of the vehicle.
The vehicle type determination apparatus according to claim 1 or 2. further,
A second distance measuring unit that generates a second distance image indicating a distance between the laser scanner and a measurement object of the laser scanner based on a distance measured by the laser scanner;
A feature point whose distance difference changes stepwise is extracted from the second distance image generated by the second distance measurement unit, and a tire candidate in which the extracted feature point is arranged in a circle or an ellipse is detected. A circular figure detector;
From the tire candidates detected by the circular figure detection unit, a tire detection unit that determines that a tire candidate having no feature point below the tire candidate is an axle tire that constitutes an axle, and
With
The axle detection unit detects the tire candidate determined as the axle tire by the tire detection unit as an axle.
The vehicle type determination apparatus according to claim 3. further,
A second vehicle height measurement unit that measures the vehicle height of the vehicle based on the second distance image generated by the second distance measurement unit;
A vehicle height measurement integration unit that determines the vehicle height of the vehicle based on the vehicle height measured by the first vehicle height measurement unit and the vehicle height measured by the second vehicle height measurement unit;
A vehicle width measurement integration unit that corrects the vehicle width measured by the vehicle width measurement unit based on the distance measured by the laser scanner and determines the vehicle width of the vehicle;
The determination unit sets the vehicle height determined by the vehicle height measurement integration unit as the vehicle height of the vehicle, and sets the vehicle width determined by the vehicle width measurement integration unit as the vehicle width of the vehicle.
The vehicle type determination apparatus according to claim 3 or 4. The vehicle height measurement integration unit sets a vehicle height having a large value between the vehicle height measured by the first vehicle height measurement unit and the vehicle height measured by the second vehicle height measurement unit as the vehicle height of the vehicle. ,
The vehicle width measurement integration unit determines a distance between the laser scanner and the vehicle based on the distance measured by the laser scanner, and corrects the vehicle width measured by the vehicle width measurement unit based on the distance.
The vehicle type determination device according to claim 5. further,
A first vehicle detection unit for detecting that the vehicle has moved forward or backward based on the first distance image;
The first distance measurement unit generates a first distance image when the first vehicle detection unit moves forward.
The vehicle type determination apparatus according to any one of claims 1 to 6. further,
A second vehicle detection unit that detects that the vehicle has entered and the vehicle has exited based on the distance measured by the laser scanner;
The axle counting and counting unit counts the axle between the time when the second vehicle detection unit detects that the vehicle has entered and the time when the second vehicle detection unit detects the vehicle leaving,
The vehicle type determination apparatus according to any one of claims 3 to 7.

Images