JP3329321B2 - Traveling vehicle detection device and traveling vehicle detection method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a traveling vehicle detection device and a traveling vehicle detection method for detecting the position of a vehicle traveling on a road surface, and more particularly to detecting the position of a vehicle traveling on a sloped road surface. The present invention relates to a traveling vehicle detection device and a traveling vehicle detection method.

[0002]

2. Description of the Related Art Conventionally, in order to obtain effective information for efficient use of roads and for ensuring safety, the vehicle is driven on a road using a camera which is a sensor capable of extracting three-dimensional information of space and time. Road conditions such as the positions and behaviors of all vehicles are being grasped one by one.

[0003]

However, in the prior art, since the road surface is a plane and only simple parameters such as the height and angle of the camera are handled, only the position information measured by the projection onto a simple plane can be obtained. However, there is a problem that the position of a vehicle running on a road cannot be accurately measured on a sloped road surface such as a mountain area.

Further, when handling detailed parameters regarding the positional relationship between the camera and the road surface and the three-dimensional structure of the road surface,
Although it is possible to accurately measure the position of a vehicle traveling on a sloping road surface such as a mountain area, the logic is completely different from the conventional method of calculating position information using simple parameters such as camera height and angle. However, there is a problem that the cost of the apparatus increases because it is necessary to perform advanced arithmetic at a high speed.

The present invention has been made in view of such a problem, and an object of the present invention is to perform a gradient correction on position information measured by a simple projection on a plane, thereby making it possible to perform a slope correction in a mountainous area. Even on roads with gradients like
An object of the present invention is to provide a traveling vehicle detection device and a traveling vehicle detection method capable of performing highly accurate vehicle position measurement with a simple configuration and reducing costs.

[0006]

Means for Solving the Problems In order to solve the above problems, the present invention has the following constitution. A gist of the invention according to claim 1 is a traveling vehicle detection device that detects position information of a vehicle from a video signal of a vehicle traveling on a road surface photographed from above by a camera, and extracts the vehicle from the video signal. Vehicle extracting means, a screen position specifying means for specifying a screen position at which the vehicle is extracted by the vehicle extracting means in the video signal, and the screen position specifying means based on the screen position specified by the screen position specifying means. Plane distance calculation means for calculating a plane projection distance between a virtual position on the road surface of the vehicle assumed when the road surface is assumed to be a plane and a camera position on the road surface where the camera is installed, The camera position corresponding to the gradient of the road surface on which the vehicle is traveling by correcting the plane projection distance calculated by the plane distance calculating means, and the vehicle extracting means It consists in the traveling vehicle detection apparatus characterized by comprising a gradient distance calculating means for calculating an actual distance between the vehicle is more extracted. The gist of the invention described in claim 2 is that the slope distance calculation means calculates the actual distance by defining the slope of the road surface on which the vehicle is traveling as an arc. The present invention resides in the traveling vehicle detection device. The gist of the invention according to claim 3 is that the plane distance calculation means calculates a predetermined constant determined from a size of an image sensor of the camera and a focal length of a lens, a mounting angle of the camera, and a height of the camera. 3. The traveling vehicle detection device according to claim 1, wherein the plane projection distance is calculated using the screen position of the vehicle as a parameter. The gist of the invention described in claim 4 is that the gradient distance calculating means calculates the actual distance from the predetermined radius of the gradient of the road surface and the height of the camera using the plane projection distance as a parameter. The traveling vehicle detection device according to any one of claims 1 to 3, wherein: The gist of the invention according to claim 5 is that the screen position specified by the screen position specifying means is the center of the vehicle width extracted by the vehicle extracting means and the vehicle extracted by the vehicle extracting means. The traveling vehicle detection device according to any one of claims 1 to 4, wherein the traveling vehicle detection device is specified by a lower side of a vertical width of the traveling vehicle. Claim 6
The gist of the described invention is a traveling vehicle detection method for detecting position information of the vehicle from a video signal of a vehicle traveling on a road surface photographed by a camera, wherein the vehicle is extracted from the video signal, and the video signal is extracted. A virtual position on the road surface of the vehicle and the camera assumed when the road surface is assumed to be a flat surface based on the specified screen position. Calculating the plane projection distance with the camera position on the road surface where the camera is installed, correcting the calculated plane projection distance and the camera position according to the gradient of the road surface on which the vehicle is traveling, and A running vehicle detection method is characterized in that an actual distance to the extracted vehicle is calculated. The gist of the invention according to claim 7 is that the actual distance is calculated by defining the gradient of the road surface on which the vehicle is traveling as an arc. Exist. The gist of the invention according to claim 8 is that the screen position of the vehicle is determined from a predetermined constant determined from a size of an image sensor of the camera and a focal length of a lens, a mounting angle of the camera, and a height of the camera. 7. The plane projection distance is calculated as a parameter.
Or a running vehicle detection method according to 7. Claim 9
The gist of the present invention is that the actual distance is calculated from a predetermined radius of the slope of the road surface and a height of the camera using the plane projection distance as a parameter. The present invention relates to a traveling vehicle detection method described in The gist of the invention according to claim 10 is that the specified screen position is specified by a center of a width of the extracted vehicle and a lower side of a height of the extracted vehicle. Item 10 is a running vehicle detection method according to any one of Items 6 to 9.

[0007]

Embodiments of the present invention will be described below in detail with reference to the drawings.

FIG. 1 is a block diagram showing a configuration of an embodiment of a traveling vehicle detection device according to the present invention.

In this embodiment, an input device 1 for inputting a video signal as a binary image from a camera installed on a pole or the like and photographing a road surface from above, and a video signal from the input device 1 A data processing device 2 for performing data processing, a storage device 3 for storing intermediate information in a data processing process of a video signal in the data processing device 2, and an output device 4 for outputting a data processing result of the video signal by the data processing device 2. The data processing device 2 includes a vehicle extraction unit 21, a plane projection calculation unit 22, and a gradient correction calculation unit 23. The storage device 3 includes a screen position storage unit 31 and a plane position storage unit 32. Become.

The vehicle extracting means 21 extracts vehicle information from the video signal from the input device 1 to calculate position information on the vehicle screen, and stores the calculated vehicle position information on the screen in the screen position storage unit 31. Remember.

The plane projection calculation means 22 uses the parameters determined by the height, angle of view, installation angle and the like of the camera from the position information on the screen of the vehicle stored in the screen position storage unit 31 to project the plane from the camera to the vehicle. The distance is calculated, and the calculated distance from the camera to the vehicle is stored in the plane position storage unit 32.

The slope correction calculating means 23 corrects a calculation error due to the slope based on the plane projection distance from the camera to the vehicle stored in the plane position storage unit 32 and calculates an accurate distance from the camera to the vehicle. The calculated accurate distance from the camera to the vehicle is output to the output device 4.

Next, the operation of this embodiment will be described with reference to FIGS.
This will be described in detail with reference to FIG. FIG. 2 is a flowchart for explaining the operation of the embodiment of the traveling vehicle detection device according to the present invention, and FIG. 3 is an explanatory diagram for explaining the operation of the vehicle extracting means shown in FIG. FIG. 4 is an explanatory diagram for explaining the operation of the plane projection calculating means shown in FIG. 1, and FIG. 5 is an explanatory diagram for explaining the operation of the gradient correction calculating means shown in FIG.

When a binarized image as shown in FIG. 3 is input from the input device 1 to the data processing device 2 as a video signal from a camera, first, the input binarized image is converted into a two-dimensional image by the vehicle extracting means 21. Each vehicle is extracted from the binarized image,
An ID is assigned to each extracted vehicle, and a screen position on the binarized image for each extracted vehicle is measured and stored in the screen position storage unit 31 in association with the assigned ID (A1). The screen position on the extracted binarized image for each vehicle is measured by the horizontal position and the vertical position, the horizontal position is the center of the horizontal width of the extracted vehicle, and the vertical position is the vertical position of the extracted vehicle. Measure the lower side of the width, that is, the place where the road surface is installed. In FIG. 3, two vehicles are extracted from the input binary image,
The screen position data representing the screen position of each of the two extracted vehicles on the binarized image (horizontal position: 20, vertical position: -100, ID: A) and (horizontal position: -90, vertical position: 60, ID: B) are stored in the screen position storage unit 31.

Next, the plane projection calculating means 22 calculates the projection onto the plane for all the screen position data to which the IDs stored in the screen position storage section 31 are assigned, and calculates the plane projection distance between the vehicle and the camera. Then, the calculated distance between the vehicle and the camera is stored in the plane position storage unit 32 (A
2). The plane projection distance WY between the vehicle and the camera is, as shown in FIG. 4, the mounting angle of the camera α, the height of the camera H, the screen position of the vehicle N, the size of the image sensor of the camera and the focal length of the lens. Assuming that a constant determined from is K, it is calculated by the following equation 1. Here, the plane projection distance WY between the vehicle and the camera is, as shown in FIG. 4, the virtual position on the road surface assumed from the screen position of the vehicle when the road surface is assumed to be flat, and the camera is installed. It is the distance to a place on the road.

[0016]

(Equation 1)

Next, a calculation error due to the gradient is corrected by the gradient correction calculation means 23 based on the plane projection distance WY between the vehicle and the camera stored in the plane position storage unit 32 (A
3). That is, the actual distance y between the vehicle and the camera is calculated based on the plane projection distance WY between the vehicle and the camera calculated by the plane projection calculation means 22. Here, the actual distance between the vehicle and the camera is the actual distance along the road surface between the position of the vehicle and the position on the road surface where the camera is installed as shown in FIG.

A method of calculating the actual distance y between the vehicle and the camera by the gradient correction calculating means 23 will be described below. First, FIG.
, The distance between the vehicle and the camera in the plane is Y,
The vertical distance of the vehicle due to the gradient is Z, and the radius of the gradient is R
Then, the vertical distance Z of the vehicle due to the gradient can be expressed by the following equation 2. Note that the gradient radius R
Is set in advance by regarding the gradient of the road surface of the road as an arc. In the example shown in FIG. 5, the gradient is considerably emphasized in order to easily explain the calculation operation in the gradient correction calculating means 23.

[0019]

(Equation 2)

Next, when Expression 2 is modified, the distance Y between the vehicle and the camera on the plane can be expressed by Expression 3 shown below.

[0021]

(Equation 3)

The camera installation location a and the vehicle location b
Is a point on an arc, and the distance Y between the vehicle and the camera on the plane, the vertical distance Z of the vehicle based on the gradient, and the radius R of the gradient can be expressed by the following Equation 4.
The radius R of the gradient is the radius of the circular arc when the road surface of the road is along the circular arc.

[0023]

(Equation 4)

Next, by substituting equation (3) into equation (4), the following equation (5) is obtained.

[0025]

(Equation 5)

Here, if s is defined as in the following equation 6, equation 5 can be expressed as in the following equation 7.
From the formula of the solution of the quadratic equation, the vertical distance Z of the vehicle due to the gradient can be obtained by Expression 8 shown below.

[0027]

(Equation 6)

Equation 7

(Equation 8)

Accordingly, the distance Y between the vehicle and the camera on the plane can be obtained by substituting the equation 8 into the equation 3, and the actual distance y between the vehicle and the camera can be obtained by calculating the distance between the vehicle and the camera along the arc of the road surface. Is calculated by the following equation 9 as the actual distance from

[0029]

[Equation 9]

In practice, since the radius R of the gradient is generally a considerably large value compared to the height H of the camera, the vertical distance Z of the vehicle due to the gradient is small, and the vertical distance Z of the vehicle due to the gradient is small. Is sufficiently small, the actual distance y between the vehicle and the camera and the distance Y between the vehicle and the camera on the plane may be set to y ≒ Y, and when the vertical distance Z of the vehicle due to the gradient cannot be ignored. However, the actual distance y between the vehicle and the camera may be represented by the following Expression 10 as an approximate value.

[0031]

(Equation 10)

In this manner, the accurate vehicle position in consideration of the gradient is calculated by the gradient correction calculating means 23, and the accurate vehicle position in consideration of the calculated gradient is output to the output device 4.
(A4).

The above-described steps A1 to A4 are performed for all the vehicles extracted by the vehicle extracting means 21, and the exact positions of all the vehicles extracted by the vehicle extracting means 21 are grasped.

As described above, according to the present embodiment, the gradient of the road surface is regarded as a circular arc.
The gradient correction can be applied to the position information measured by projecting it on a simple plane, and the effect of being able to measure the position of the vehicle with high accuracy using a simple configuration can be obtained even on a sloped road such as a mountain area. To play.

It should be noted that the present invention is not limited to the above embodiments, and that the embodiments can be appropriately modified within the scope of the technical idea of the present invention. Further, the number, position, shape, and the like of the constituent members are not limited to the above-described embodiment, but can be set to numbers, positions, shapes, and the like suitable for carrying out the present invention. In the drawings, the same components are denoted by the same reference numerals.

[0036]

The traveling vehicle detection device and the traveling vehicle detection method of the present invention perform gradient correction on position information measured by projection onto a simple plane by capturing the gradient of the road surface as a circular arc. This makes it possible to measure the position of the vehicle with a simple structure and with high accuracy even on a sloped road surface such as a mountain area.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of an embodiment of a traveling vehicle detection device according to the present invention.

FIG. 2 is a flowchart for explaining the operation of the embodiment of the traveling vehicle detection device according to the present invention.

FIG. 3 is an explanatory diagram for explaining an operation of the vehicle extracting means shown in FIG. 1;

FIG. 4 is an explanatory diagram for explaining an operation in the plane projection calculation means shown in FIG. 1;

FIG. 5 is an explanatory diagram for explaining an operation of the gradient correction calculating means shown in FIG. 1;

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 input device 2 data processing device 3 storage device 4 output device 21 vehicle extraction means 22 plane projection calculation means 23 gradient correction calculation means 31 screen position storage unit 32 plane position storage unit

Claims (10)

(57) [Claims] 1. A traveling vehicle detection device for detecting position information of a vehicle from a video signal of a vehicle traveling on a road surface photographed from above by a camera, comprising: vehicle extraction means for extracting the vehicle from the video signal. A screen position specifying means for specifying a screen position where the vehicle is extracted by the vehicle extracting means in the video signal; and the road surface is flat based on the screen position specified by the screen position specifying means. Plane distance calculation means for calculating a plane projection distance between a virtual position on the road surface of the vehicle assumed when the vehicle is assumed and a camera position on the road surface where the camera is installed, and the plane distance calculation means The camera position corresponding to the gradient of the road surface on which the vehicle travels is corrected by correcting the plane projection distance calculated by Traveling vehicle detecting apparatus characterized by comprising a gradient distance calculating means for calculating an actual distance between the vehicle was. 2. The traveling vehicle detection device according to claim 1, wherein the gradient distance calculation means calculates the actual distance by defining the gradient of the road surface on which the vehicle is traveling as an arc. . 3. The screen position of the vehicle based on a constant determined from a predetermined size of an image sensor of the camera and a focal length of a lens, an attachment angle of the camera, and a height of the camera. The traveling vehicle detection device according to claim 1, wherein the plane projection distance is calculated using the parameter as a parameter. 4. The method according to claim 1, wherein the gradient distance calculation means calculates the actual distance from the predetermined radius of the gradient of the road surface and a height of the camera using the plane projection distance as a parameter. The traveling vehicle detection device according to any one of claims 1 to 3. 5. The screen position specified by the screen position specifying means is located below a center of a lateral width of the vehicle extracted by the vehicle extracting means and a vertical width of the vehicle extracted by the vehicle extracting means. The traveling vehicle detection device according to any one of claims 1 to 4, characterized in that: 6. A traveling vehicle detection method for detecting position information of a vehicle from a video signal of a vehicle traveling on a road surface photographed by a camera, comprising: extracting the vehicle from the video signal; A screen position where the extracted vehicle is located is specified, and a virtual position of the vehicle on the road surface and an installation of the camera are assumed when the road surface is assumed to be flat based on the specified screen position. The plane projection distance with the camera position on the road surface that has been calculated is calculated, and the calculated plane projection distance is corrected to extract the camera position and the camera position according to the gradient of the road surface on which the vehicle is traveling. A traveling vehicle detection method, wherein an actual distance to the vehicle is calculated. 7. The running vehicle detection method according to claim 6, wherein the actual distance is calculated by defining the slope of the road surface on which the vehicle is running as an arc. 8. The plane projection distance using the screen position of the vehicle as a parameter from a predetermined constant determined from the size of the image sensor of the camera and the focal length of the lens, the mounting angle of the camera, and the height of the camera. The traveling vehicle detection method according to claim 6 or 7, wherein is calculated. 9. The actual distance is calculated from a predetermined radius of the gradient of the road surface and a height of the camera, using the plane projection distance as a parameter. The traveling vehicle detection method according to any one of the preceding claims. 10. The screen position to be specified is specified by a center of a horizontal width of the extracted vehicle and a lower side of a vertical length of the extracted vehicle. The method for detecting a traveling vehicle according to any one of claims 1 to 3.