JP2014209686A - Vehicle detection device and vehicle detection system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a vehicle detection device and a vehicle detection system capable of accurately correcting photographic images photographed by cameras.SOLUTION: The vehicle detection device comprises image acquisition means, light emission control means, deviation detection means, setting means, and image processing means. The image acquisition means receives images photographed by plural cameras having predetermined photographing areas. The light emission control means controls a light source for emission of light for irradiating therewith a fixed position in the photographing area of each camera. The deviation detection means detects the positional deviation of each camera on the basis of the image obtained by photographing by each camera the photographing area including the fixed position irradiated with the light by the light source. The setting means sets to each camera a correction amount corresponding to the positional deviation detected by the deviation detection means. The image processing means performs image processing on an image obtained by correcting the image photographed by each camera, by the correction amount set by the setting means.

Description

  Embodiments described herein relate generally to a vehicle detection device and a vehicle detection system.

  Conventionally, some vehicle detection devices detect a vehicle from images captured by a plurality of mechanically fixed cameras. The vehicle detection device detects the vehicle by comparing the image in the reference state with the image captured by each camera. However, if there is a shift in the shooting position of each camera due to the installation environment or aging, the vehicle recognition rate may be reduced.

JP-A-10-307352

  An object of this invention is to provide the vehicle detection apparatus and vehicle detection system which can correct | amend the picked-up image which a camera image | photographs with high precision.

  According to the embodiment, the vehicle detection device includes an image acquisition unit, a light emission control unit, a deviation detection unit, a setting unit, and an image processing unit. The image acquisition means inputs images taken by a plurality of cameras having a predetermined shooting area. The light emission control means and a light source that emits light for irradiating a fixed position in the imaging region of each camera are controlled. The deviation detecting means detects a positional deviation of each camera based on an image obtained by each camera photographing an imaging region including a fixed position where the light source emits light. The setting means sets a correction amount corresponding to the positional deviation detected by the deviation detecting means for each camera. The image processing means performs image processing on an image obtained by correcting an image captured by each camera with a correction amount set by the setting means.

FIG. 1 is a diagram schematically illustrating a first configuration example of a vehicle detection system including a vehicle detection device according to the present embodiment. FIG. 2 is a diagram schematically illustrating a second configuration example of the vehicle detection system including the vehicle detection device according to the present embodiment. FIG. 3 is a block diagram illustrating a configuration example of the vehicle detection device according to the present embodiment. FIG. 4A shows the timing at which the LED emits light. FIG. 4B is a diagram illustrating an example of a standard time for the camera to receive the LED light emitted at the timing illustrated in FIG. FIG. 4C shows an example of timing at which the camera actually receives the light from the LED emitted at the timing shown in FIG. FIG. 5 is a flowchart for explaining a flow of a correction amount setting process for a shift in the shooting direction of the camera in the vehicle detection device according to the present embodiment. FIG. 6A is a diagram illustrating an example of a standard position of LED light in a captured image of the camera. FIG. 6B is a diagram illustrating an example of LED light shift in a captured image of the camera. FIG. 6C is a diagram illustrating a correction amount with respect to the deviation of the LED light in the captured image of the camera. FIG. 7 is a flowchart for explaining a flow of a correction amount setting process for a shift in the plane direction of the camera in the vehicle detection device according to the present embodiment.

Hereinafter, embodiments will be described with reference to the drawings.
FIG. 1 is a diagram illustrating a first configuration example of a vehicle detection system including a vehicle detection device according to the present embodiment. Moreover, FIG. 2 is a figure which shows the 2nd structural example of the vehicle detection system containing the vehicle detection apparatus which concerns on this embodiment.

  The vehicle detection system according to the present embodiment is a system that detects a vehicle as a detection target. The vehicle detection system detects a vehicle passing through a predetermined position on the road. Further, in the vehicle detection system, a plurality of cameras called stereo cameras capture one image area. The vehicle detection system has a function of calculating position information of a subject (vehicle) using parallax information between an image captured by one camera and an image captured by the other camera. In the present embodiment, the vehicle detection system is a system that detects a vehicle traveling on a road such as an expressway or a toll road.

1 and 2, the vehicle detection system includes a vehicle detection device 10, a first camera 11, a second camera 12, an LED 13 (13 '), a support column 14, a reflector 15, and the like. .
For example, the first camera 11 and the second camera 12 are installed at an entrance or exit of an expressway or a toll road. The vehicle detection device detects a vehicle that enters or exits from images captured by the first camera 11 and the second camera 12. The detection result of the vehicle by the vehicle detection device 10 is used for a fee settlement process for the vehicle user.

  The first camera 11 and the second camera 12 constitute a stereo camera as a plurality of cameras that shoot the same shooting region. For example, the first camera 11 and the second camera 12 are CCD cameras or the like. The 1st camera 11 and the 2nd camera 12 should just be installed in the position where parallax arises. In the configuration example shown in FIGS. 1 and 2, the first camera 11 and the second camera 12 are installed up and down at a predetermined distance.

  Moreover, the 1st camera 11 and the 2nd camera 12 are fixed and installed by the side of the road used as the detection object area | region of the vehicle to pass. The first camera 11 and the second camera 12 are installed, for example, so as to look down on the road obliquely so as to capture an image including an upper surface and a side surface of a vehicle passing through the road (vehicle to be detected). . The first camera 11 and the second camera 12 transmit the captured image to the vehicle detection device 10. In the configuration example shown in FIGS. 1 and 2, the first camera 11 and the second camera 12 are fixed to a column 14 installed on the side of the road. However, in practice, it is difficult to fix the first camera 11 and the second camera 12 so that the photographing position is completely stationary. For this reason, it is assumed that the first camera 11 and the second camera 12 may change their shooting positions.

  The LED 13 illuminates (marks) one fixed point in the area captured by the first camera 11 and the second camera 12. The first camera 11 and the second camera 12 capture the light emitted from the LED 13 as a point in the captured image. The LED 13 is fixedly installed so that the light irradiation position becomes one fixed point.

  For example, in the configuration shown in FIG. 1, the LED 13 is installed between the first camera 11 and the second camera 12. Further, in the configuration example shown in FIG. 1, the reflecting plate 15 is provided at a position where the light emitted from the LED 13 is irradiated. The light emitted from the LED 13 is applied to the fixed plate 15 at one fixed point. The light emitted by the LED 13 is reflected by the reflecting plate 15 and is photographed as a point in an image photographed by the first camera 11 and the second camera 12.

  Therefore, in the configuration example shown in FIG. 1, the light of the LED 13 photographed (received) by the first camera 11 has an optical path of a distance from the first camera 11 to the reflector and a distance from the reflector to the first camera 11. It will be the added distance. Similarly, the light of the LED 13 that is imaged (received) by the second camera 12 is a distance obtained by adding the distance from the second camera 12 to the reflector 15 and the distance from the reflector 15 to the second camera 12 in the optical path. Become.

In the configuration example shown in FIG. 2, the LED 13 ′ is fixed and installed so as to directly irradiate light to the first camera 11 and the second camera 12. In the configuration example shown in FIG. 2, the reflecting plate 15 shown in FIG. 1 is omitted.
Therefore, in the configuration example shown in FIG. 2, the light of the LED 13 ′ taken (received) by the first camera 11 has a light path from the LED 13 ′ to the first camera 11. Similarly, the light of the LED 13 ′ photographed (received) by the second camera 12 has a light path from the LED 13 ′ to the second camera 12.

Note that the LED 13 (or 13 ′) may be any one that irradiates light to a fixed point in the area captured by the first camera 11 and the second camera 12, and the installation positions of the LEDs 13 and 13 ′ are as shown in FIG. Or it is not limited to the structural example shown in FIG. Moreover, you may provide LED13, 13 'with respect to the 1st camera 11 and the 2nd camera 12, respectively.
However, since the LEDs 13 and 13 'emit light for detecting a shift in the photographing position of the first camera 11 and the second camera 12, the first camera 11 and the second camera 12 are subject to secular change and environment. Even if it is shifted due to the above, it is assumed that the light irradiation position is fixed at one point.

Next, the configuration of the vehicle detection device 10 will be described.
The vehicle detection device 10 includes a computer having a processor such as a CPU, various memories, various interfaces, and the like. The vehicle detection device 10 is connected to the first camera 11, the second camera 12, the LED 13 (13 ′), and the like. The vehicle detection device 10 is a device that performs processing for detecting a vehicle from images taken by the first camera 11 and the second camera 12. The vehicle detection device 10 has an image processing function for processing image data captured by the first camera 11 and the second camera 12. For example, the vehicle detection device 10 can determine the distance to the vehicle included in the images captured by the first camera 11 and the second camera 12 based on the image captured by the first camera 11 and the image captured by the second camera 12. Is calculated.

  Each function of the vehicle detection device 10 may be realized by the CPU executing a program stored in the memory, or may be realized by hardware. For example, the image processing function for processing images taken by the first camera 11 and the second camera 12 may be realized by executing a program for image processing stored in the memory by the CPU. It may be realized by hardware such as a general-purpose LSI.

FIG. 3 is a block diagram illustrating a configuration example of the vehicle detection device 10.
In the configuration example illustrated in FIG. 3, the vehicle detection device 10 includes a control unit 21, a storage unit 22, a first image interface (I / F) 23, a second image interface (I / F) 24, and an LED control unit 25. An image processing unit 26 and a data communication unit 27.

  The control unit 21 performs control of the entire vehicle detection device 10 and data processing. The control unit 21 includes, for example, a CPU, RAM, ROM, and various internal interfaces. The control unit 21 implements control of each unit and various processing functions by the CPU executing a program stored in the internal memory or the storage unit 22. The data communication unit 27 is an interface for communicating with an external device. For example, the data communication unit 27 is connected to a host device.

  The storage unit 22 stores various data. The storage unit 22 is, for example, a rewritable nonvolatile memory. The storage unit 22 stores setting information, application programs, and the like. Further, the storage unit 22 stores correction information such as information indicating the correction amount for the image captured by the first camera 11 and information indicating the correction amount for the image captured by the second camera 12. The storage unit 22 also has a standard time (standard time of the first camera) from when the LED 13 emits light until the first camera 11 receives the light of the LED 13 (photographing), and after the LED 13 emits light, 2 The standard time until the camera 12 receives the light of the LED 13 (photographing) (standard time of the second camera), the standard position of the light of the LED 13 in the image photographed by the first camera 11 (first standard position), the second Reference information such as the standard position (second standard position) of the light of the LED 13 in the image captured by the camera 12 is stored.

  The first and second image I / Fs 23 and 24 are configured by an image capture board. The first image I / F 23 is connected to the first camera 11. The first image I / F 23 acquires an image taken by the first camera 11. The second image I / F 24 is connected to the second camera 12. The second image I / F 24 acquires an image captured by the second camera 12. In addition, the first and second image I / Fs 23 and 24 notify the control unit 21 of information indicating the timing at which the first and second cameras 11 and 12 capture images.

  The LED control unit 25 is connected to the LED 13 (13 ′). The LED control unit 25 controls the driving of the LED 13 (13 ′). For example, the LED control unit 25 causes the LED 13 (13 ′) to emit light. The LED control unit 25 causes the LED to emit light according to a control signal instructing light emission from the control unit 21. Further, the LED control unit 25 supplies a signal indicating the timing at which the LED 13 (13 ′) emits light to the control unit 21.

  The image processing unit 26 processes images captured by the first camera 11 and the second camera 12. For example, the image processing unit 26 has a function of detecting a vehicle from an image taken by the first camera 11 and an image taken by the second camera 12, and an image taken by the first camera 11 and an image taken by the second camera 12. A function of calculating vehicle position information (distance) based on the parallax. The image processing unit 26 has a function of correcting the image captured by the first camera 11 and the image captured by the second camera 12 based on the correction amount stored in the storage unit 22. The image processing unit 26 may be a function realized by the CPU of the control unit 21 executing a program, or may be realized by an LSI for image processing.

Next, deviation correction with respect to the shooting direction (viewing direction) of the first camera 11 and the second camera 12 will be described.
The deviation with respect to the photographing direction is determined by the time until the first camera 11 and the second camera 12 receive the light emitted by the LED 13 (13 ′). In order to calculate the deviation with respect to the shooting direction, the vehicle detection device 10 stores in the storage unit 22 a standard time until the first camera 11 and the second camera 12 receive the light emitted by the LED 13 (13 ′). Shall.

  The vehicle detection device 10 uses the difference between the time until the first camera 11 receives the light actually emitted by the LED 13 (actual measurement time) and the standard time of the first camera 11 stored in the storage unit 22. And a correction amount for the first camera 11 is set according to the calculated deviation. Further, the vehicle detection device 10 determines the first time by the difference between the time until the second camera 12 receives the light actually emitted by the LED 13 (actual measurement time) and the standard time of the second camera 12 stored in the storage unit 22. The deviation of the two cameras 12 with respect to the shooting direction is calculated, and the correction amount for the second camera 12 is set according to the calculated deviation.

FIG. 4 is a timing chart for explaining a method of calculating a deviation amount in the shooting direction of the first camera 11.
FIG. 4A shows the timing at which the LED 13 emits light. FIG. 4B is a diagram showing the standard time of the first camera 11 with respect to the light emission timing of the LED 13 shown in FIG. In other words, if there is no deviation in the shooting direction, as shown in FIG. 4B, the first camera 11 receives the light (reflecting plate) from the LED 13 at the standard time ta of the first camera 11 from the timing when the LED 13 emits light. The reflected light reflected at 15 is received.

  FIG. 4C shows an example of timing when the first camera 11 receives light actually emitted from the LED 13 (reflected light reflected by the reflecting plate 15). In the example shown in FIG. 4C, the first camera 11 receives the light emitted from the LED 13 for a time (measurement time) t1 that is longer than the standard time ta by a time (difference time) Δt.

  For example, in the configuration example shown in FIG. 1, the light from the LED 13 is received by the first camera 11 via the reflector 15. Therefore, in the configuration example shown in FIG. 1, the first camera 11 that receives the light emitted by the LED 13 during the actual measurement time t1 until the first camera 11 is shifted in the photographing direction by the difference time Δt1 / 2. For this reason, in the configuration example shown in FIG. 1, if the speed of light is c, the first camera 11 is shifted by a distance (c × Δt1) / 2 in the photographing direction. This distance (c × Δt) / 2 is set as a correction amount with respect to the shooting direction of the first camera 11.

  In the configuration example shown in FIG. 2, the first camera 11 directly receives light from the LED 13. Therefore, in the configuration example shown in FIG. 2, the first camera 11 that receives the light emitted by the LED 13 during the actual measurement time t1 until the first camera 11 is shifted in the photographing direction by the difference time Δt1. Therefore, in the configuration example shown in FIG. 2, if the speed of light is c, the first camera 11 is shifted by a distance (c × Δt1) in the photographing direction. This distance (c × Δt) is set as a correction amount for the shooting direction of the first camera 11.

  Further, the correction amount with respect to the shooting direction of the first camera 11 is set by, for example, storing it in the storage unit 22. Further, the displacement of the second camera 12 with respect to the shooting direction is also calculated by the difference time Δt2 between the actual measurement time t2 until the second camera 12 receives the light emitted by the LED 13 and the standard time tb of the second camera 12. The correction amount with respect to the shooting direction of the second camera 12 is also calculated from the difference time Δt2 and stored in the storage unit 22, for example.

Next, the shift correction amount setting process with respect to the shooting direction of the first camera 11 and the second camera 12 will be described.
FIG. 5 is a flowchart for explaining the flow of the setting process of the shift correction amount with respect to the shooting direction of the first camera 11 and the second camera 12.
First, the control unit 21 of the vehicle detection apparatus 10 causes the LED control unit 25 to cause the LED 13 (13 ′) to emit light (step 10). The control unit 21 actually measures the time from when the LED 13 emits light until the first camera 11 and the second camera 12 receive the light of the LED 13 (steps S11 and S21).

  When the actual measurement time until the first camera 11 receives the light emitted from the LED 13 is the time t1, the control unit 21 determines the actual measurement time t1 and the standard time ta of the first camera 11 stored in the storage unit 22. Is calculated (step S12). The control unit 21 determines whether or not the difference time Δt1 is not 0, that is, whether or not the actual measurement time t1 and the standard time ta are not equal (step S13). When Δt = 0 (step S13, NO), the control unit 21 ends the setting of the correction amount for the shift in the shooting direction of the first camera 11 without setting the correction amount.

  When Δt is not 0, that is, when the actual measurement time t1 and the standard time ta are not the same value (step S13, YES), the control unit 21 determines the first difference time Δt1 between the actual measurement time t1 and the standard time ta. A deviation amount with respect to the shooting direction of the camera 11 is calculated (step S14). For example, in the configuration example illustrated in FIG. 1, the first camera 11 determines that the distance (c (speed of light) × Δt1) / 2 is shifted from the shooting direction. In the configuration example shown in FIG. 2, the first camera 11 is determined to be shifted by a distance (c (light speed) × Δt1) with respect to the shooting direction.

  When the shift amount with respect to the shooting direction of the first camera 11 is calculated, the control unit 21 stores the calculated shift amount with respect to the shooting direction of the first camera 11 in the storage unit 22, thereby correcting the shooting direction of the first camera 11. The amount is set (step S15).

  When the actual measurement time until the second camera 12 receives the light emitted from the LED 13 is time t2, the control unit 21 determines the actual time t2 and the standard time of the second camera 12 stored in the storage unit 22. A difference time Δt2 from tb is calculated (step S22). The control unit 21 determines whether or not the difference time Δt2 is not 0, that is, whether or not the actual measurement time t2 and the standard time tb are not equal (step S23). When Δt = 0 (step S23, NO), the control unit 21 ends the setting of the correction amount for the shift in the shooting direction of the second camera 12 without setting the correction amount.

  When Δt is not 0, that is, when the actual measurement time t2 and the standard time tb are not the same value (step S23, YES), the control unit 21 determines the second difference Δt2 between the actual measurement time t2 and the standard time tb. A deviation amount with respect to the shooting direction of the camera 12 is calculated (step S24). For example, in the configuration example illustrated in FIG. 1, the second camera 12 determines that the distance (c (speed of light) × Δt2) / 2 is shifted from the shooting direction. In the configuration example illustrated in FIG. 2, the second camera 12 determines that the distance (c (speed of light) × Δt2) is shifted from the shooting direction.

  When the deviation amount with respect to the shooting direction of the second camera 12 is calculated, the control unit 21 stores the calculated deviation amount with respect to the shooting direction of the second camera 12 in the storage unit 22, thereby correcting the shooting direction of the second camera 12. The amount is set (step S25).

  According to the correction amount setting process for the deviation in the shooting direction described above, the time from when the LED emits light until each camera constituting the stereo camera receives the light of the LED is measured. Based on the difference from the standard time until the LED light is received, a correction amount for the shift in the shooting direction of each camera is set. Thereby, according to the vehicle detection apparatus of this embodiment, the shift | offset | difference in the imaging | photography direction of each camera can be correct | amended easily. As a result, the vehicle detection device can easily identify the exact position of the vehicle in the image captured by each camera, and can realize highly accurate vehicle detection.

Next, displacement correction in the upper and lower left and right directions (plane direction) of the first camera 11 and the second camera 12 will be described.
The shift in the vertical and horizontal directions (planar direction) of the first camera 11 and the second camera 12 is determined by the position of the light of the LED 13 (13 ′) in the images taken by the first camera 11 and the second camera 12. In order to calculate the deviation in the plane direction, the vehicle detection device 10 stores information indicating the standard position of the light of the LED 13 (13 ′) in the image captured by the first camera 11 and the second camera 12 in the storage unit 22. I shall keep it.

  The vehicle detection device 10 detects the light position (photographing position) of the LED 13 detected from the image actually captured by the first camera 11 and the standard position (first position of the light of the LED 13 with respect to the first camera 11 stored in the storage unit 22. A deviation in the plane direction of the first camera 11 is detected based on a difference from the standard position), and a correction amount with respect to the plane direction of the first camera 11 is set according to the detected deviation amount.

  The vehicle detection device 10 also detects the position of the light of the LED 13 (photographing position) detected from the image actually captured by the second camera 12 and the standard position of the light of the LED 13 with respect to the second camera 12 stored in the storage unit 22 (first position). The deviation in the plane direction of the second camera 12 is detected from the difference from the standard position (1), and a correction amount for the plane direction of the second camera 12 corresponding to the detected deviation is set.

FIG. 6A is a diagram illustrating an example of the standard position of the light of the LED 13 (13 ′) in the captured image of the first camera 11 (or the second camera 12). FIG. 6B is a diagram illustrating an example of the light shift of the LED 13 (13 ′) in the captured image of the first camera 11 (or the second camera 12). FIG. 6C is a diagram illustrating a correction amount with respect to the light shift of the LED 13 (13 ′) in the captured image of the first camera 11 (or the second camera 12).
In other words, if there is no deviation in the plane direction of the first camera 11 (or the second camera 12), the vehicle detection device 10 in the image photographed by the first camera 11 (or the second camera 12) is shown in FIG. As shown in FIG. 4, the light of the LED 13 is detected at the first standard position (or the second standard position). If the light of the LED 13 is detected at the first standard position (or the second standard position), the vehicle detection device 10 does not need to correct the plane direction of the first camera 11 (or the second camera 12). to decide.

  When there is a deviation in the plane direction, the vehicle detection device 10 uses the first standard position (or as shown in FIG. 6B) in the image captured by the first camera 11 (or the second camera 12). The light of the LED 13 is detected at a position deviated from the second standard position. When the light of the LED 13 is detected at a position deviated from the first standard position (or the second standard position), the vehicle detection device 10 calculates a deviation amount (for example, a coordinate value indicating the deviation amount) from the standard position.

  When a deviation amount from the standard position (for example, a coordinate value indicating the deviation amount) is calculated, the vehicle detection apparatus 10 sets a correction amount for correcting the calculated deviation amount for the first camera (or the second camera 12). Set as a value. By correcting with this correction amount, the image taken by the first camera (or the second camera 12) is corrected for deviation in the plane direction as shown in FIG. 6C.

Next, a description will be given of a process for setting a deviation correction amount with respect to the planar direction of the first camera 11 and the second camera 12.
FIG. 7 is a flowchart for explaining the flow of the setting process of the correction amount of the displacement of the first camera 11 and the second camera 12 with respect to the planar direction.
First, the control unit 21 of the vehicle detection device 10 causes the LED control unit 25 to cause the LED 13 (13 ′) to emit light (step 30). After causing the LED 13 to emit light, the control unit 21 acquires images including light spots (light spots) of the LEDs 13 taken by the first camera 11 and the second camera 12 (steps S31 and S41).

  When the image including the light spot of the LED 13 captured by the first camera 11 is acquired, the control unit 21 extracts the light spot of the LED 13 in the actually captured image, and obtains the coordinate values (p1, q1) of the extracted light spot. It detects (step S32). When the coordinate value (p1, q1) of the light spot of the LED 13 in the photographed image of the first camera 11 is detected, the control unit 21 detects the detected coordinate value (p1, q1) and the first position as the standard position with respect to the first camera 11. The coordinate values (pa, qa) of the standard position and the coordinate values (Δp1, Δq1) of the difference are calculated (step S33).

  When the coordinate values (Δp1, Δq1) of the difference between the light spot position of the LED 13 actually taken by the first camera 11 and the standard position are calculated, the control unit 21 sets the difference coordinate values (Δp1, Δq1) to (0). , 0), that is, whether the light spot of the photographed LED is not the standard position (step S34). When the difference coordinate values (Δp1, Δq1) are (0, 0) (step S34, NO), the control unit 21 sets the correction amount for the shift in the plane direction of the first camera 11 without setting the correction amount. Finish the setting.

  When the calculated difference coordinate values (Δp1, Δq1) are not (0, 0), that is, when the light spot of the LED 13 taken by the first camera 11 is not the standard position (step S34, YES), the control unit 21 By storing the difference coordinate values (Δp1, Δq1) in the storage unit 22 as a shift amount in the plane direction of the first camera 11, a correction amount with respect to the plane direction of the first camera 11 is set (step S35).

  When the image including the light spot of the LED 13 captured by the second camera 12 is acquired, the control unit 21 extracts the light spot of the LED 13 in the actually captured image, and the coordinate value (p2, q2 of the extracted light spot). ) Is detected (step S42). When the coordinate value (p2, q2) of the light spot of the LED 13 in the captured image of the second camera 12 is detected, the control unit 21 detects the detected coordinate value (p2, q2) and the second position as the standard position with respect to the second camera 12. The coordinate values (pb, qb) of the standard position and the difference coordinate values (Δp2, Δq2) are calculated (step S43).

  When the coordinate values (Δp2, Δq2) of the difference between the light spot position of the LED 13 actually taken by the second camera 12 and the standard position are calculated, the control unit 21 sets the difference coordinate values (Δp2, Δq2) to (0). , 0), that is, whether the light spot of the captured LED is not in the standard position (step S44). When the difference coordinate values (Δp2, Δq2) are (0, 0) (step S44, NO), the control unit 21 sets the correction amount for the shift in the planar direction of the second camera 12 without setting the correction amount. Finish the setting.

  When the calculated coordinate values (Δp2, Δq2) of the difference are not (0, 0), that is, when the light spot of the LED 13 photographed by the second camera 12 is not the standard position (step S44, YES), the control unit 21 By storing the difference coordinate values (Δp2, Δq2) in the storage unit 22 as a shift amount in the plane direction, a correction amount for the plane direction of the second camera 12 is set (step S45).

  According to the processing described above, the position of the light spot of the LED that irradiates (marks) a fixed point is detected from each image captured by each camera constituting the stereo camera, and the light spot of the LED in the captured image is detected. Based on the difference between the position and the standard position of the light spot for each camera, a correction amount for the shift in the planar direction of each camera is set. Thereby, according to the vehicle detection apparatus of this embodiment, the shift | offset | difference in the plane direction of each camera can be correct | amended easily. As a result, the vehicle detection device can easily identify the exact position of the vehicle in the image captured by each camera, and can realize highly accurate vehicle detection.

  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 10 ... Vehicle detection apparatus, 11 ... 1st camera, 12 ... 2nd camera, 13 ... LED (light source), 14 ... Support | pillar, 15 ... Reflecting plate, 21 ... Control part (deviation detection means, setting means), 22 ... Memory Unit (storage unit), 23 ... first image interface (image acquisition unit), 24 ... second image interface (image acquisition unit), 25 ... LED control unit (light emission control unit), 26 ... image processing unit (image) Processing means).

Claims (10)

Image acquisition means for inputting images taken by a plurality of cameras having a predetermined shooting area;
A light emission control means for controlling a light source that emits light for irradiating a fixed position in the imaging region of each camera;
A shift detection means for detecting a shift in the position of each camera based on an image captured by each camera in a shooting area including the fixed position where the light source emits light;
Setting means for setting a correction amount according to the positional deviation detected by the deviation detection means for each camera;
Image processing means for performing image processing on an image obtained by correcting the image captured by each camera with a correction amount set by the setting means;
A vehicle detection device. Furthermore, it has a memory | storage means to memorize | store the standard information with respect to the imaging state of the light from the said light source by each said camera,
The deviation detecting means detects a position deviation based on a difference between a state in which the camera captures light from the light source and standard information stored in the storage means;
The vehicle detection device according to claim 1. The storage means stores, as the standard information, a standard time from when the light source emits light until each camera captures light from the light source,
The deviation detection unit is configured to detect a deviation in a shooting direction of the camera based on a difference between an actual measurement time from when the light source emits light until the camera actually captures light from the light source and the standard time with respect to the camera. Detect
The vehicle detection device according to claim 2. The storage means stores, as the standard information, a standard position indicating a position of light from the light source in an image captured by each camera,
The shift detection means detects a shift in the vertical and horizontal directions of the camera based on a difference between a position of light from the light source in an image actually captured by the camera and the standard position with respect to the camera.
The vehicle detection device according to claim 2. Furthermore, it has a reflecting plate that reflects light from the light source to the cameras.
The vehicle detection device according to any one of claims 1 to 4. In a vehicle detection system having a plurality of cameras and a vehicle detection device in which a predetermined shooting area is set,
Furthermore, it has a light source that emits light that illuminates a fixed position in the imaging region of each camera,
The vehicle detection device is
A shift detection means for detecting a shift in the position of each camera based on an image captured by each camera in a shooting area including the fixed position where the light source emits light;
Setting means for setting a correction amount according to the positional deviation detected by the deviation detection means for each camera;
Image processing means for performing image processing on an image obtained by correcting the image captured by each camera with a correction amount set by the setting means,
Vehicle detection system. The vehicle detection device further includes storage means for storing standard information for a shooting state of light from the light source by each camera.
The deviation detecting means detects a position deviation based on a difference between a state in which the camera captures light from the light source and standard information stored in the storage means;
The vehicle detection system according to claim 6. The storage means stores, as the standard information, a standard time from when the light source emits light until each camera captures light from the light source,
The deviation detection unit is configured to detect a deviation in a shooting direction of the camera based on a difference between an actual measurement time from when the light source emits light until the camera actually captures light from the light source and the standard time with respect to the camera. Detect
The vehicle detection system according to claim 7. The storage means stores, as the standard information, a standard position indicating a position of light from the light source in an image captured by each camera,
The shift detection means detects a shift in the vertical and horizontal directions of the camera based on a difference between a position of light from the light source in an image actually captured by the camera and the standard position with respect to the camera.
The vehicle detection system according to claim 7. Furthermore, it has a reflecting plate that reflects light from the light source to the cameras.
The vehicle detection system according to any one of claims 6 to 9.

Images