WO2019176418A1 - Vehicular lamp, vehicle detection method, and vehicle detection device - Google Patents

Abstract

Provided are: a lamp unit (1); an image capture means (camera) (2) that captures an image of a light irradiation region of the lamp unit (1); and a control means (lamp ECU) (6) for controlling the lamp unit (1) and the image capture means (2), the control means (6) being connected to a signal bus (CAN) (100) installed in a vehicle. Aiming of the lamp unit (1) is adjusted on the basis of signals obtained by performing signal-processing on image signals captured by the image capture means (2). When a prescribed command signal has been inputted, the control means (6) outputs the signal-processed signal to the signal bus (100), and adjusts aiming using an aiming adjuster connected to the signal bus (100).

Description

Vehicle lamp, vehicle detection method, and vehicle detection device

The present disclosure relates to a vehicle lamp suitable for application to a headlamp of an automobile, and more particularly to a vehicle lamp provided with imaging means such as a camera. The present disclosure also relates to a vehicle detection method and a vehicle detection device that are mounted on a vehicle such as an automobile and detect other vehicles such as an oncoming vehicle and a preceding vehicle.

As an automobile headlamp, ADB (Adaptive Driving Beam) control and AHB (Automatic High Beam) control, etc., are used to block out some areas so that oncoming cars and preceding cars are not dazzled and to illuminate other areas over a wide area. A headlamp capable of light control (hereinafter also referred to as ADB control) has been proposed. In this type of lamp, a headlamp in which a camera is integrated is proposed as an imaging means for detecting an oncoming vehicle or a preceding vehicle. For example, in Patent Document 1, a camera unit is housed together with a lamp unit in a lamp housing, and imaging is performed by the camera unit.

By incorporating the imaging means in the headlamp, the imaging optical axis of the imaging means and the light irradiation optical axis of the lamp unit can be brought closer as compared with the case where the imaging means is disposed in the front window of the automobile, such as ADB control. Can be performed with higher accuracy. In other words, it is possible to reduce the light distribution error due to the parallax generated between the two central axes by bringing the central axis of the image capturing the oncoming vehicle and the preceding vehicle close to the central axis of the light distribution such as ADB control. is there.

In addition, in order to improve the visibility of the front area of the host vehicle, an ADB (AdaptiveingDriving) that detects the other vehicle existing in the front area and controls the light distribution of the headlamp so as not to dazzle the other vehicle. Beam) control and AHB (Automatic High Beam) control are performed. In recent years, other vehicles have been detected for automatic driving control. As a technique for detecting such other vehicles, a technique has been proposed in which a surrounding area of the host vehicle is imaged by an imaging unit, and an image obtained by the imaging is analyzed to detect the other vehicle.

In Patent Document 2, a light spot in an image obtained by an imaging means is detected, and the attribute of the detected light spot, for example, the color and behavior (moving state) of the light spot, A technique has been proposed in which the headlamps and taillights of other vehicles are discriminated to detect other vehicles. Patent Document 2 also discloses a technique for shortening the detection time by setting a detection area for detecting another vehicle based on the detected attribute of the light spot as a predetermined area and determining the set area. Proposed.

Japanese Unexamined Patent Publication No. 2013-164913 Japanese Unexamined Patent Publication No. 2012-20672

By the way, aiming adjustment for adjusting the optical axis direction of the lamp unit is performed after the headlamp is assembled to the car body of the automobile. In the headlamp having the image pickup means as described above, the image is picked up by the image pickup means. It is conceivable to perform aiming adjustment while confirming the light distribution pattern.

In addition, when detecting an oncoming vehicle or a preceding vehicle using an imaging unit when performing ADB control or the like, vehicle detection is performed by analyzing an image captured by the imaging unit. Detection by this image analysis In order to improve accuracy, it is considered to use AI (Artificial Intelligence) having a learning function.

Conventionally, no satisfactory content has been proposed for aiming adjustment or AI learning as described later in detail. For this reason, aiming adjustment and learning by AI are problems in realizing suitable ADB control using an imaging unit.

An object of the present disclosure is to provide a vehicle lamp capable of realizing suitable light distribution control.

The technique of Patent Document 2 is effective in reducing the detection time by limiting the detection area. However, in the recent situation where the driving environment of automobiles is diversified, it is required to detect other vehicles over a wide area, and therefore it is difficult for the technology of Patent Document 2 to meet this requirement. At the same time, illuminants such as road signs and advertisements increase, and the number of light spots in the captured image tends to increase. Therefore, a process for discriminating other vehicles based on all the behaviors of these light spots is required, and the number of detection steps increases and the detection time increases.

An object of the present disclosure is to provide a vehicle detection method and a vehicle detection device that can narrow down light spots in a captured image and detect a vehicle accurately and quickly.

The present disclosure relates to a lamp unit that performs light irradiation, an imaging unit that captures at least a light irradiation region of the lamp unit, and a signal bus disposed in a vehicle, for controlling the lamp unit and the imaging unit. A vehicular lamp having a control unit and configured to perform aiming adjustment of the lamp unit based on a signal obtained by performing signal processing on an imaging signal captured by the imaging unit, and the control unit is configured to receive a predetermined command signal. The signal processed signal is output to the signal bus. The signal processed signal is preferably an image signal based on the imaging signal or an aiming adjustment signal.

In the present disclosure, it is preferable that the camera and the lamp unit are built in the lamp housing, and the camera is detachably supported with respect to the lamp housing. For example, when vehicle lamps are disposed on both sides of the vehicle, the camera is disposed on one selected vehicle lamp of each vehicle lamp.

Furthermore, in the present disclosure, the control unit includes a detection unit that detects at least another vehicle based on an imaging signal captured by the camera, and the detection unit is machine-learned based on the automatically generated traveling simulation data. It is preferable. In the traveling simulation data, it is preferable that traveling image data and teacher data are automatically generated.

The vehicle detection method according to the present disclosure is a detection method that is mounted on a vehicle, images the surroundings of the vehicle, and detects the vehicle from the behavior of a light spot that is captured in an image obtained by the imaging. A step of recognizing the shapes of a plurality of existing light spots, a step of identifying a light spot that is a vehicle detection target among the recognized shapes as a target light spot, and a light spot other than that as a non-target light spot, and an image The method includes detecting a behavior of a target light spot in the vehicle and detecting a light spot having a predetermined behavior as a vehicle.

A vehicle detection device according to the present disclosure includes an imaging unit that is mounted on a vehicle and images the surroundings of the vehicle, and a detection unit that detects a vehicle from a light spot captured in an image captured by the imaging unit. The vehicle is detected from the target light spot that is the vehicle detection target from the plurality of light spots existing in the light source, the light spot identification unit that identifies the other non-target light spots, and the behavior of the identified target light spot in the image A light detection unit that recognizes shapes of the plurality of light spots, and identifies the target light spot and the non-target light spot based on the recognition;

According to the present disclosure, the aiming adjuster can acquire a signal obtained by performing signal processing on an image pickup signal picked up by the image pickup unit, for example, an image signal or an aiming adjustment signal obtained by calculation, through the signal bus of the vehicle. High-precision aiming adjustment can be easily performed, and suitable light distribution control can be realized.

Further, according to the present disclosure, the shape of the light spot in the image captured by the imaging unit is identified, and the light spot that is likely to be a vehicle is identified as the target light spot from this shape, and the identified target light spot The process for detecting the vehicle only is executed. As a result, even when there are many light spots in the image, there is no need to perform detection processing for light spots that are not likely to be light spots of the vehicle, and the number of processing steps for detecting the vehicle is reduced. In addition, rapid vehicle detection is possible.

The conceptual diagram of the motor vehicle which applied the headlamp concerning this indication. The schematic horizontal sectional view of a right headlamp. The exploded perspective view of a camera. The block diagram of lamp ECU. The flow diagram of DL (deep learning). The conceptual diagram which shows the 1st aiming adjustment form. The conceptual diagram which shows the 2nd aiming adjustment form. Schematic which shows the CAN connection state of a right-and-left headlamp. The block diagram of lamp ECU of the headlamp of FIG. Schematic which shows the image imaged with the camera of the left-right headlamp, and its light distribution. Schematic which shows the image imaged with the camera of the left-right headlamp, and its light distribution. Schematic which shows the image imaged with the camera of the left-right headlamp, and its light distribution. FIG. 3 is a schematic horizontal sectional view of a headlamp to which the present disclosure is applied. The block diagram of lamp ECU containing a vehicle detection means. Schematic of an example of the front area | region of the own vehicle imaged with a camera. Schematic of the light spot obtained by imaging with a camera. The conceptual diagram for demonstrating the extraction area | region and shape recognition of a light spot. The conceptual diagram for demonstrating the extraction area | region and shape recognition of a light spot. The conceptual diagram for demonstrating the extraction area | region and shape recognition of a light spot. The conceptual diagram for demonstrating the extraction area | region and shape recognition of a light spot. Schematic of target light spot. Schematic detected as a light spot of a vehicle. The conceptual diagram explaining the vehicle detection method when a camera exists in a high position. The conceptual diagram explaining the vehicle detection method when a camera exists in a low position. The conceptual diagram explaining the vehicle detection method when an image inclines. The conceptual diagram explaining the vehicle detection method using the light spot of a white line.

(Vehicle lamp)
Next, the form of the vehicle lamp according to the present disclosure will be described with reference to the drawings. FIG. 1 is a conceptual diagram of an embodiment in which the present disclosure is applied to a headlamp HL of an automobile. The left and right headlamps L-HL and R-HL of the automobile CAR are each provided with a lamp unit 1 capable of ADB control in a lamp housing 3. The lamp unit 1 includes, for example, a light source 11 composed of a plurality of LEDs (light emitting diodes). All or selected LEDs 11 are caused to emit light, and the emitted light is projected onto a front area of the automobile by a projection lens 12. Thus, it is possible to perform light irradiation with a desired light distribution pattern P.

For example, in FIG. 1, each light emitting region of the plurality of LEDs 11 is set to illuminate a predetermined segmented region Ap in the front region of the automobile, and only the segmented region Ap corresponding to the LED 11 selected and emitted is included. Illuminated. Therefore, when all the LEDs 11 emit light, the entire area of the light distribution pattern P is illuminated. Moreover, since the area | region corresponding to the quenched LED is not illuminated, the dazzling with respect to these motor vehicles can be prevented by quenching LED of the area | region which detected the oncoming vehicle and the preceding vehicle, for example.

In addition, a camera 2 that captures a moving image is provided as an imaging means in the lamp housing 3 of each of the headlamps L-HL and R-HL. The camera 2 images at least a front area of the automobile irradiated with light by the lamp unit 1 and outputs an imaging signal.

The left and right headlamps L-HL and R-HL are respectively assembled to the left and right of the front part of the car body of the car CAR in the car assembly process. After this assembly, the optical axis adjustment, that is, aiming adjustment of the lamp unit 1 and the camera 2 is executed. By this aiming adjustment, the irradiation optical axis of the lamp unit 1 and the imaging optical axis of the camera 2 are set in a predetermined direction, so that highly accurate ADB control can be realized.

FIG. 2 is a schematic horizontal sectional view of the right headlamp R-HL among the headlamps HL. The lamp housing 3 includes a lamp body 31 and a translucent front cover 32. A base plate 5 is supported on the lamp body 31 by an aiming mechanism 4. The lamp unit 1 and the camera 2 are attached to the base plate 5. Here, the irradiation optical axis of the lamp unit 1 and the imaging optical axis of the camera 2 are attached to the base plate 5 in a state where they are directed in the same direction.

Accordingly, when the aiming screw 41 constituting a part of the aiming mechanism 4 is pivoted manually or by a motor, the base plate 5 is tilted up and down, left and right, and the irradiation optical axis of the lamp unit 1 and the imaging of the camera 2 are captured. Aiming adjustment of the optical axis is performed. Since the aiming mechanism 4 having such an aiming screw 41 is already known, detailed description thereof is omitted, but here, the aiming adjustment is performed manually.

As shown in the exploded perspective view of FIG. 3, the camera 2 includes a camera body 21 including an imaging lens 211 and an imaging element 212, and a camera holder 22 that holds the camera body 21. The camera holder 22 is fixed to the base plate 5 by appropriate means. The camera body 21 is detachably attached to the camera holder 22. In this example, the camera body 21 can be attached and detached by sliding up and down with respect to the camera holder 22 from the outside of the lamp housing 3 through an opening not shown in the drawing provided on the upper surface of the lamp body 31. It is configured. These camera body 21 and camera holder 22 are provided with electrodes 23 and 24, respectively. When the camera body 21 is mounted on the camera holder 22, both are electrically connected to each other by these electrodes 23 and 24.

On the other hand, as shown in FIG. 2, a lamp ECU (electronic control unit) 6 is housed in the lamp housing 3. The lamp unit 1 and the camera 2 are electrically connected to the lamp ECU 6. FIG. 4 is a block diagram of the lamp ECU 6. The lamp ECU 6 includes a main control unit 61. An input / output unit 62, a signal processing unit 63, an image analysis unit 64, and a lighting drive unit 65 are connected to the main control unit 61. The main control unit 61 includes the input / output unit 62 and the signal processing unit 63. The operations of the image analysis unit 64 and the lighting drive unit 65 are controlled.

The lamp unit 1 and the camera 2 described above are connected to the input / output unit 62. In the camera 2, the camera body 21 is connected to the input / output unit 62 via the camera holder 22. The input / output unit 62 is connected to a signal bus, for example, CAN (Controller 例 え ば Area Network) 100 that extends to the car CAR. The input / output unit 62 is connected to various other ECUs of the automobile not shown in the drawing via the CAN 100.

In the lamp ECU 6, the signal processing unit 63 processes the image pickup signal of the camera 2 input from the input / output unit 62 to generate a moving image or a still image signal. The image analysis unit 64 analyzes an image obtained from the generated image signal, and detects an object in the captured image, particularly other vehicles such as an oncoming vehicle and a preceding vehicle. Therefore, the image analysis unit 64 is configured as other vehicle detection means. The lighting drive unit 65 recognizes the object detected by the image analysis unit 64, sets an appropriate light distribution pattern that does not dazzle the oncoming vehicle and the preceding vehicle, and generates the light distribution pattern. An ADB control signal is generated. Based on this ADB control signal, the light source of the lamp unit 1, that is, the light emission of the plurality of LEDs 11 is controlled.

In this way, the lamp ECU 6 detects the other vehicle such as the oncoming vehicle and the preceding vehicle based on the image obtained by imaging with the camera 2, and does not dazzle the detected other vehicle, while other regions. The lighting of the lamp unit 1 is controlled so as to obtain a light distribution pattern capable of brightly illuminating and appropriate ADB control is executed.

Here, the image analysis unit 64, that is, the other vehicle detection means detects the other vehicle by performing machine learning based on the traveling simulation data generated by DL (Deep Learning) using AI. Preferably it is configured. In order to improve the detection performance of an object in this DL, it is desirable to perform a large amount of learning without a break. In order to perform a large amount of learning, it is necessary to create and learn in advance a large amount of traveling video data and teacher data that is the correct answer of the traveling data. Therefore, these data can be automatically generated.

In order to collect travel data, it is necessary to travel as many roads as possible in the destination with actual vehicles and acquire data, which requires enormous costs and time. In addition, the presence / absence and position of the vehicle must be extracted from the accumulated traveling video and the correct answer data must be created. The more accumulated data, the more time is required.

In this embodiment, when executing the DL, the image analysis unit 64 automatically generates a driving environment (S1), automatically generates and arranges road objects (S2), and automatically runs a road as shown in the flow of FIG. Generation (S3), acquisition of correct data by own vehicle traveling (S4), and learning by a learning device are repeated (S5). This increases the learning amount and improves the detection performance.

That is, an image of traveling on a road in a virtual space is created by a traveling simulator. The presence / absence of the vehicle and the position on the image viewed from the camera of the own vehicle are converted from 3D to 2D from the position, orientation, distance of the vehicle arranged at an arbitrary coordinate in the virtual space, and the direction and position of the camera of the own vehicle. Etc., and is recorded as correct answer data.

Also, by setting camera parameters (F value, angle of view, exposure time, etc.) so as to correctly acquire the camera image viewed from the camera mounting position, an image close to the actual camera image can be acquired. it can.

∙ Use a test course in which objects on the road that have already been created are placed in the virtual space.

On top of that, the vehicle will travel on all roads (both up and down). The own vehicle and vehicles other than the own vehicle go around the course at the set road speed limit of ± 20 km / h, and are set so as to observe traffic rules such as traffic lights and temporary stops. Also, make sure that the vehicles do not collide. Then, the learning device inputs the image and the correct answer obtained by traveling of the host vehicle as teacher data.

In the automatic generation of the travel route, a road is generated in a virtual space from an actual road map image (2D map, satellite photograph image). In this case, roads are generated by identifying roads and other roads by image processing from maps and satellite photographs. Further, the number of lanes may be set appropriately from the width of the road, or the lanes may be forcibly determined according to the test case.

Objects other than roads may be appropriately arranged according to areas other than roads such as buildings, parking lots, signboards, and mountainous areas. Pedestrian crossings and signals are placed where the roads intersect. Other vehicles may be arranged according to the length of the generated building or road, and may be set so as to continue moving from one building to another.

As described above, the detection performance by DL is improved by repeating the learning using the traveling simulator. Therefore, in this embodiment, the detection accuracy of other vehicles such as an oncoming vehicle and a preceding vehicle in the image analysis unit of the lamp ECU that has been machine-learned based on simulation data obtained based on such learning is improved. ADB control becomes possible.

The configuration of the left headlamp L-HL is substantially the same except that the arrangement of the lamp unit 1 and the camera 2 in the lamp housing 3 is bilaterally symmetrical to the configuration of FIG. Although illustration is omitted, also in the left headlamp L-HL, a lamp ECU is built in the lamp housing and connected to the CAN 100.

These left and right headlamps L-HL and R-HL are mounted on the body of an automobile CAR, and then the irradiation optical axis of the lamp unit 1 and the imaging optical axis of the camera 2 are directed in a predetermined direction. Aiming adjustment is performed. In this aiming adjustment, as shown in FIG. 1, the light of the lamp unit 1 is irradiated to the screen Sc disposed at a predetermined position in front of the car CAR, and the light distribution pattern P generated by the light irradiation is obtained. Is captured by the camera 2. Then, the irradiation optical axis of the lamp unit 1 is detected from the captured image of the light distribution pattern, and the irradiation optical axis is aimed at a predetermined direction, for example, the center O that is the intersection of the horizontal line H and the vertical line V. Aiming adjustment is performed by operating the screw 41. The aiming adjustment of the imaging optical axis of the camera 2 is also performed by the aiming adjustment of the lamp unit 1.

In such aiming adjustment, two adjustment modes are conceivable. FIG. 6 is a schematic diagram for explaining the first adjustment mode. In the first adjustment mode, an external connector 7 is provided on a part of the lamp housing 3 and is directly connected to the camera 2. Thereby, the image signal picked up by the camera 2 can be directly output through the external connector 7. The external aiming adjuster 8A is connected to the external connector 7. The first aiming adjuster 8A includes a signal processing unit 81 similar to the signal processing unit 63 provided in the lamp ECU 6, and a monitor 82 for displaying an image signal generated by the signal processing unit 63. A monitor driving unit 83 for displaying an image corresponding to the image signal on the monitor 82 is provided.

In the aiming adjustment, a reference is clearly specified in advance on the screen Sc shown in FIG. 1, and the posture of the automobile CAR is adjusted in accordance with this reference. The lamp unit 1 is turned on and the screen Sc is irradiated with light, and then the light distribution pattern P irradiated by the camera 2 is imaged. The first aiming adjuster 8 </ b> A captures the captured image signal through the external connector 7, and the signal processing unit 81 generates an image signal. The monitor driving unit 83 displays an image, that is, a captured light distribution pattern on the monitor 82 based on the generated image signal. The operator confirms how much the light distribution pattern displayed while visually recognizing the monitor 82 is deviated from a predetermined position, for example, the center in each of the three directions of yaw, roll, and pitch, and aiming to eliminate this deviation. Manually adjust the screw 41. Thereby, aiming adjustment is executed.

In this first adjustment mode, it is necessary to provide the lamp housing 3 of the headlamp HL with an external connector 7 for connecting to the first aiming adjuster 8A in order to capture an image signal captured by the camera 2. This is an obstacle to miniaturization of the lamp HL and increases the cost. Further, it is necessary to connect and remove the external connector 7, and it takes time for the work and the work efficiency is deteriorated. Further, the first aiming adjuster 8A needs to incorporate the signal processing unit 81, which is expensive.

On the other hand, in the second adjustment mode, as shown in FIG. 7, the lamp housing 3 of the headlamp HL is not provided with an external connector. The second aiming adjuster 8B is not provided with a signal processing unit, and includes a monitor 82 and a monitor driving unit 83. The second aiming adjuster 8B can be connected to the CAN 100 connected to the lamp ECU 6. The connection to the CAN 100 can be easily realized by connecting to an existing CAN connector provided in the automobile CAR.

In order to execute this second adjustment mode, the lamp ECU 6 incorporates a predetermined program in the main control unit 61 in advance. When a predetermined command signal output from the second aiming adjuster 8B through the CAN 100 is input, the lamp ECU 6 outputs (transmits) the image signal generated by the signal processing unit 63 from the input / output unit 62 to the CAN 100. It is configured. As this predetermined command signal, for example, a signal output when the power of the second aiming adjuster 8B is turned on or a signal output in advance by an operation in the second aiming adjuster 8B can be adopted.

In the second adjustment mode, the second aiming adjuster 8B inputs (receives) the image signal output from the lamp ECU 6 to the CAN 100 and causes the monitor driving unit 83 to display the image signal on the monitor 82. Thereby, the operator who performs aiming adjustment can perform aiming adjustment while visually recognizing the image picked up by the camera 2 on the monitor 82 as in the first adjustment mode.

Therefore, in the second adjustment mode, it is not necessary to provide an external connector on the lamp housing 3, and connection and disconnection with respect to the external connector are not required. For this reason, according to the second adjustment mode, the headlamp HL can be reduced in size and cost, and quick and easy adjustment can be realized. Furthermore, since it is not necessary to incorporate a signal processing unit in the first aiming adjuster 8A, the first aiming adjuster 8A can be configured at a low price.

As described above, the lamp ECU 6 is configured to realize the second adjustment mode, so that the aiming adjuster having a simple configuration such as the second aiming adjuster 8B can be used quickly and easily. Aiming adjustment can be realized. Thereby, the irradiation optical axis of the lamp unit 1 and the imaging optical axis of the camera 2 in the headlamp HL can be accurately adjusted, and a suitable ADB can be realized.

Here, in the second adjustment mode, when the aiming screw 41 of the aiming mechanism 4 provided in the headlamp HL can be automatically adjusted by a motor or the like, the illustration is omitted. The aiming adjuster may include a calculation unit that calculates the amount of deviation of the light distribution pattern from a predetermined position and a motor drive unit that feedback-controls the motor so that the amount of deviation is zero. Thereby, aiming adjustment can be performed automatically. In this case, since the aiming adjustment can be performed only by using the image signal, the monitor and the monitor driving unit may be omitted.

On the other hand, in the second adjustment mode, when the main control unit 61 of the lamp ECU 6 outputs an image signal, the image signal of an area necessary for aiming adjustment, for example, a predetermined including the center of the imaging area captured by the camera 2 is included. Only the image signal in the area may be output to the CAN 100. In this way, it is possible to reduce the amount of data when outputting the image signal and to increase the speed of the aiming adjustment. Alternatively, when it is predicted that the amount of deviation in aiming is not so large, the luminance of the captured light distribution pattern is measured, and it is highly probable that the high luminance region is a region including the center. May be output to the CAN 100.

Furthermore, the main control unit 61 of the lamp ECU 6 may include a calculation unit that has calculated the deviation amount in the second aiming adjuster, and the rotation amount of the aiming screw motor calculated based on the deviation amount. You may provide the calculating part which calculates | requires. Both the shift amount and the rotation amount are output as aiming adjustment signals. In the case of the former aiming adjustment signal, it is only necessary to output the calculated deviation amount from the lamp ECU 6 to the CAN 100, and the second aiming adjustment machine only needs to include a motor drive unit. In the case of the latter aiming adjustment signal, it is only necessary to output only the calculated motor rotation speed from the lamp ECU 6, and the configuration of the motor drive unit of the second aiming adjustment machine can be simplified. Thereby, further aiming adjustment can be simplified and speeded up.

In the lamp ECU configured to realize the second aiming adjustment mode, as shown in FIG. 8, the left and right headlamps L-HL and R-HL lamp ECUs 6 send and receive signals to each other through the CAN 100. It becomes possible to do. That is, an image signal output from one headlamp HL to the CAN 100 can be captured by the lamp ECU 6 of the other headlamp HL. For example, when the lamp ECU 6 of the other headlamp HL receives a command signal output from the lamp ECU 6 of one headlamp HL to the CAN100, the received lamp ECU6 outputs an image signal generated by itself to the CAN100. This output image signal can be received by the lamp ECU 6 of one headlamp HL and captured.

Therefore, the left and right headlamps L-HL and R-HL lamp ECUs 6 (6L, 6R) are each provided with an abnormality detection unit 66 for detecting an abnormality of the camera, as shown in FIG. The main control unit 61 is configured to output a predetermined abnormality command signal when the abnormality detection unit 66 detects an abnormality of the camera 2.

With this configuration, for example, when an abnormality occurs in the camera 2R of the right headlamp R-HL, the right headlamp R-HL uses a lamp based on the image signal captured by the camera 2R on the own side. Unit 1 cannot perform ADB control. At that time, an abnormal command signal is output from the main control unit 61 of the lamp ECU 6 to the CAN 100. Here, the abnormality of the camera is a state in which a normal image cannot be acquired, such as a case where imaging is impossible or a field of view is deteriorated during imaging.

The lamp ECU 6L of the left headlamp L-HL performs ADB control based on the image signal captured by the camera 2L on its own side, but the abnormal command signal output from the lamp ECU 6R of the right headlamp R-HL is CAN100. When the signal is input through the camera 100, an image signal captured by the camera 2L on the own side is output to the CAN 100. Since this image signal is input to the lamp ECU 6R of the right headlamp R-HL in an abnormal state through the CAN 100, the right headlamp R-HL is converted into the image signal captured by the camera 2L of the input left headlamp L-HL. Based on this, ADB control is executed.

Thus, even if an abnormality occurs in one of the left and right headlamps L-HL and R-HL, both heads are used based on the image captured by the other normal camera 2R or 2L. ADB control can be secured in each of the lamps L-HL and R-HL. In this case, in a headlamp in which an abnormality of the camera actually occurs, the parallax angle between the irradiation optical axis of the lamp unit and the imaging optical axis of a normal camera becomes large. It is preferable to increase the value so as to reliably prevent dazzling oncoming vehicles and preceding vehicles.

In the embodiment described above, the left and right headlamps L-HL and R-HL are each provided with a camera. However, in the case of a low-grade automobile, it is conceivable that the camera is provided only on one headlamp. It is done. However, in this case, the parallax angle between the imaging optical axis of the camera and the irradiation optical axis of the lamp unit of the other headlamp not equipped with the camera becomes a problem. For this reason, if ADB control is performed with the other headlamp based on an image obtained from the camera of one headlamp, a situation may occur in which the other vehicle is dazzled. In particular, in an image captured by a roadside headlamp camera, it is difficult to capture an oncoming vehicle that is in the vicinity of the host vehicle in the oncoming lane.

FIGS. 10A to 10C are examples of left-hand traffic, but in a driving situation where the preceding vehicle CAR1 and the oncoming vehicle CAR2 of FIG. 10A exist, images captured by the camera 2R of the right headlamp R-HL on the oncoming lane side are FIG. 10B is an image captured by the camera 2L of the left headlamp L-HL on the shoulder side in FIG. 10B. As in this example, the camera 2L of the left headlamp L-HL on the shoulder side may not be able to reliably image the oncoming vehicle CAR2 existing in the vicinity of the host vehicle, and it is difficult to detect this.

Therefore, when the ADB control is performed based on the image of FIG. 10B, a light distribution pattern P1 like a dotted dotted area is obtained. However, when the ADB control is performed based on the image of FIG. Become. Even if the light distribution pattern P1 is applied, the oncoming vehicle CAR2 is not dazzled. However, when the light distribution pattern P2 is applied, the oncoming vehicle CAR2 is dazzled as shown by a broken line region in FIG. 10A.

Therefore, a headlamp on the opposite lane that makes it easy to reliably image and detect an oncoming vehicle in the vicinity of the host vehicle in the opposite lane, for example, in a left-hand traffic area such as Japan, a camera is installed in the right headlamp. Then, ADB control is executed in each of the left and right headlamps based on the image captured by the camera.

However, when the camera is mounted only on one headlamp in this way, it is required to mount the camera on the left headlamp in the right-hand traffic area such as Europe. Therefore, it is necessary to individually design the left and right headlamps corresponding to each region, and as a result, the cost may increase. Also, when the same vehicle travels across different traffic system areas, there is a situation where ADB control is performed based on an image captured by a camera built in a headlamp on the roadside, and the accuracy of ADB control is reduced. Is done.

In this embodiment, as shown in FIG. 3, the camera 2 includes a camera body 21 and a camera holder 22. The camera body 21 is supported by the base plate 5 via the camera holder 22 and is electrically connected to the lamp ECU 6. Accordingly, if the left and right headlamps L-HL and R-HL are configured in this manner, one camera body 21 is provided in each of the left and right headlamps L-HL and R-HL. The holder 22 can be used for attachment or removal. As a result, in the left-hand traffic area, the ADB control can be executed by attaching the camera body 21 to the right headlamp R-HL. In the area of right-hand traffic, ADB control can be performed by removing the camera body 21 attached to the right headlamp R-HL and attaching it to the left headlamp L-HL.

In this way, by adopting a configuration in which one camera body is selectively exchanged and attached to either one of the left and right headlamps, the headlamps are different even when the automobile travels across regions of different transportation systems. High-precision ADB control based on an image captured by the headlamp camera on the opposite lane side can be realized without changing to a specification.

In the above embodiment, the left and right headlamps are provided with the lamp ECUs, respectively, but one lamp ECU is connected to each lamp unit of the left and right headlamps and the camera by, for example, LIN (Local Interconnect Network). May be. In this case, in aiming adjustment, an external aiming adjuster is connected to LIN. In addition, input / output of image signals between the left and right headlamps is performed via the one lamp ECU and LIN.

In the embodiment, one lamp unit is provided in the lamp housing, but one or more lamp units, for example, a lamp unit such as a clearance lamp or a turn signal lamp, may be incorporated.

In the above description of the embodiment, an example of ADB control is shown as the light distribution control of the present disclosure. However, it goes without saying that the present invention can be similarly applied to a headlamp that performs the above-described AHB control and other light distribution controls.

(Vehicle detection method and vehicle detection device)
Next, an embodiment of the present disclosure will be described with reference to the drawings. FIG. 11 is a schematic horizontal sectional view of an embodiment in which the vehicle detection device according to the present disclosure is incorporated in a headlamp HL1 provided on the front left and right of a vehicle body 108 of an automobile. FIG. 11 shows the internal structure of the right headlamp HL1, but the left headlamp has the same configuration. The lamp housing 103 of the headlamp HL1 includes a lamp body 1031 and a translucent front cover 1032. A lamp unit 101 and a camera 102 are housed inside the lamp housing 103. The lamp unit 101 and the camera 102 are supported by a base plate 105. The lamp unit 101 and the camera 102 can be adjusted in their optical axis directions by an aiming mechanism 104 including an aiming screw 1041 and the like.

The lamp unit 101 includes a light source 1011 having a plurality of LEDs arranged in a simplified manner, and a projection lens 1012 that projects light emitted from the light source 1011 toward the front of the automobile. The lamp unit 101 can illuminate the front area of the automobile with a desired light distribution by controlling the light emission of the light source 1011.

The camera 102 is an imaging unit according to the present disclosure and is not shown in the drawing, but is a digital camera including an imaging element, for example. The camera 102 images a front area of the host vehicle, at least an area including an area irradiated with light from the lamp unit 101, and outputs an imaging signal. Objects captured by the camera 102 include signs, signs, and vehicles such as preceding vehicles and oncoming vehicles that are present in the area where the host vehicle is traveling.

A lamp ECU 106 is built in the lamp housing 103 and is connected to the lamp unit 101 and the camera 102, respectively. The lamp ECU 106 can set a suitable light distribution based on an image obtained by imaging with the camera 102. The lamp ECU 106 can perform light distribution of the lamp unit 101 based on this setting, for example, ADB control. The lamp ECU 106 is connected to a vehicle ECU or the like not shown in the figure via a CAN (Controller （Area Network) 10100, and sends and receives required signals to and from them.

As shown in the block diagram of FIG. 12, the lamp ECU 106 performs signal processing on an imaging signal output from the imaging element of the camera 102 to generate an image signal as image data, and this image signal. The image analysis unit 1062 detects the vehicle existing in the image, and the lighting control unit 1063 recognizes the detected vehicle and controls the light distribution of the lamp unit 101. And.

In the lamp ECU 106, the signal processing unit 1061 and the image analysis unit 1062 perform an operation for detecting the vehicle. These constitute a vehicle detection means in a broad sense, but in particular, the vehicle is directly detected. In a sense, the image analysis unit 1062 constitutes a vehicle detection means in a narrow sense. Therefore, the lamp ECU 106 detects the vehicle by the image analysis unit 1062, that is, the vehicle detection means, based on the image signal obtained by the camera 102, and controls the lighting so as to shield the area where the detected vehicle exists. The ADB control is executed by controlling the light distribution of the lamp unit 101 in the unit 1063.

As shown in FIG. 12, the image analysis unit 1062 serving as a vehicle detection unit identifies a light spot extraction unit 1064 that extracts a light spot existing in the image, and identifies the shape of the extracted light spot. A light spot identifying unit 1065 for removing a light spot, and a vehicle detection unit 1066 for detecting the vehicle by determining the behavior of the identified light spot, here, the movement state of the light spot in the image.

A vehicle detection operation in the image analysis unit 1062 as vehicle detection means will be described. Here, a case will be described in which the image of the front region obtained by capturing with the camera 102 is in the state of FIG. 13 when the automobile in which the ADB control of the headlamp HL1 is performed at night. In this image, the preceding vehicle CAR11 and the oncoming vehicles CAR12 and CAR13, whose lamps are lit, are traveling on a road on which a white line (including a yellow line) LINE is drawn. On the side of the road, lighting facilities such as a road lamp B and a delineator D are provided, and a rectangular road sign S1 and a rhombus road sign S2 are provided. Here, various signs are also included in the sign.

An image signal (image data) is obtained by imaging the front area with the camera 102 and processing the image signal with the signal processing unit 1061. Furthermore, the image shown in FIG. 14 is obtained from this image signal. This image is obtained by arranging bright and dark pixels (light dots) corresponding to the image sensor of the camera 102 in the X direction (row direction) and the Y direction (column direction). And in this image, the imaged vehicle, lighting equipment, a sign, etc. are displayed as a light spot. Here, for the sake of convenience, the dark area of the background is represented by a white background, and the light spot is drawn with dots.

Normally, the light spots LP1 to LP3 due to the head lamps or tail lamps of the vehicles CAR11 to CAR13, and the light spots LP4 and LP5 due to the lighting facilities B and D are directly captured by the camera 102. The light spot has a shape close to. In addition, the light spots LP6 and LP7 formed by the signs S1 and S2 are picked up by the camera 102 with respect to the high-luminance surface illuminated by the light from the illuminant, so that each shape, that is, a rectangle (square, rectangle) or rhombus The light spot has the shape of The white line LINE on the road surface is a long and narrow light spot LP8.

The light spot extraction unit 1064 extracts light spots that are candidates for inspection from this image. For example, by setting a luminance value of a predetermined level as a threshold value, the light spot extraction unit 1064 detects a region having a luminance higher than the threshold value. As a result, light spots of low brightness due to light reflected simply by the object, for example, road surface areas illuminated by the headlamps of the host vehicle are excluded, and light spots by a vehicle equipped with a sign or lamp equipped with a light emitter are detected. In the case of FIG. 14, all such light spots are detected.

The light spot extraction unit 1064 further sets the light spot area with the xy coordinates in the image for the detected light spot. That is, for the rectangular light spot LP6, a rectangular light spot area indicated by diagonal coordinates (xa, ya) to (xb, yb) is set as shown by a broken line in FIG. 15A. The same applies to the diamond light spot LP7 shown in FIG. 15B, the circular light spots LP1 to LP5 shown in FIG. 15C, and the white light spot LP8 shown in FIG. 15D, and the coordinates (xm, ym) of the maximum value in the xy directions, respectively. The light spot region is set with ~ (xn, yn). In these figures, n and m are values set for each light spot.

Next, the light spot identifying unit 1065 identifies the shape of each light spot for each set light spot area. For example, as shown by an arrow in FIG. 15A, while scanning in the x direction (row direction) from the position on the left side of the base of the set light spot area, that is, from the position of (xa, ya) to the right side, in one line The luminance of the pixel is measured. In this measurement, the above-described luminance threshold value can be used. Next, the above-mentioned row is performed, and this is repeated for several lines, for example, 2 to 5 rows in the y direction (column direction). Actually, about 3 lines are sufficient.

As shown in FIG. 15A, when most of the measured pixels in each row have a luminance higher than the threshold value, the shape of the light spot is identified as a rectangle. When the number of pixels with higher luminance than the threshold decreases toward the upper row, the shape of the light spot is identified as a rhombus or a circle as shown in FIG. 15B or 15C. In this case, the number of lines to be measured is increased by several lines, and the decrease state of the high luminance pixels is measured. If the rate of decrease is large, it is identified as a diamond in FIG. 15D, and if the rate of decrease is smaller than that, it is identified as a circle in FIG. 15C.

If the number of pixels with higher luminance than the threshold is substantially the same, but the pixels with higher luminance gradually change to the left or right as they go to the upper row, FIG. Thus, the shape of the intersection is identified as a linear shape. When the amount of change in position is constant, it is identified as a linear shape, and when the amount of change in position changes, it is identified as a curved shape.

In the above-described identification, when identifying the diamond in FIG. 15D or the circle in FIG. 15C, a stepped downward line by line from the center in the y direction on the left side of the light spot region to the center in the x direction on the bottom side. Several lines may be measured. Or you may measure from the right side of a light spot area | region, and stepwise upwards. In such a case, if all the pixels have high luminance, the possibility of being a rhombus increases. In the case of a circle, the luminance of the pixel being measured is lowered, and finally the pixel having a higher luminance is measured again. This circle includes an ellipse.

Then, when the light spot is identified as a rectangle or a diamond, a tag “non-target light spot” that is not a detection target is attached because it is highly likely that the light spot is a light spot by a marker. In addition, even when the light spot is identified as a line shape, the light spot is considered to be a light spot by a white line on the road, and therefore, a tag “non-target light spot” is attached. On the other hand, since a light spot identified as a circle is likely to be a light spot by a vehicle, a tag “target light spot” to be detected is attached. Then, the light spot identification unit 1065 outputs each light spot with these tags attached to the vehicle detection unit 1066. FIG. 16 shows a light spot with a tag “target light spot”. Here, a tag “non-target light spot” is attached to the light spots LP6 and LP7. Then, the light spots LP6 and LP7 are excluded from extraction.

Of the identified light spots, the vehicle detection unit 1066 detects the vehicle by determining the behavior of the light spots in FIG. 16 with the tag “target light spot”, in this case, the moving state in the image. . In this detection method, for example, as in Patent Document 2, the relative position change of the light spot with respect to the host vehicle, that is, the movement direction and the movement speed are tracked over time, and the road is traveled based on the result. Detects a preceding vehicle or an oncoming vehicle. FIG. 17 shows the light spot of the vehicle detected in this way. Here, the light spots LP1 and LP2 are detected as the preceding vehicle CAR11 and the oncoming vehicle CAR12. Note that the light spot LP3 due to the oncoming car CAR13 is small in size, so that it can be detected that it is located far away, and is excluded because there is little possibility of dazzling the oncoming car CAR13 by the headlamp of the host vehicle at this time.

Information on the oncoming vehicle and the preceding vehicle detected by the vehicle detection unit 1066 in this way is output to the lighting control unit 1063. Based on this information, the lighting control unit 1063 sets a light distribution that does not dazzle the preceding car CAR11 and the oncoming car CAR12, and executes the light distribution control in the lamp unit 101. Thereby, suitable ADB control is performed.

As described above, in the embodiment, the light spot identifying unit 1065 identifies a light spot that is likely to be a vehicle based on the image obtained by imaging with the camera 102, and the vehicle detecting unit 1066 identifies this. A process for detecting the vehicle only for the light spot is executed. Therefore, even when there are a large number of light spots in the captured image, detection processing is performed for light spots that are clearly not from the vehicle or light spots that are likely not from the vehicle. This eliminates the need to reduce the number of processing steps when detecting a vehicle, and enables rapid vehicle detection.

Here, it is preferable to correct the moving direction of the light spot when detecting the vehicle, particularly when detecting an oncoming vehicle. In other words, in the embodiment, the camera 102 is provided integrally with the headlamp HL1, and is substantially at the same height as the headlamp of the oncoming vehicle. In some cases, it is lower than the headlamps of the oncoming vehicle. Therefore, when the camera is arranged at a position higher than the lamp of the oncoming vehicle, for example, the moving direction of the oncoming vehicle in the captured image is larger than when the camera is arranged on the front window or side mirror of the automobile. Is different.

FIG. 18A is an image when the camera is positioned higher than the headlamp HL1 of the oncoming vehicle CAR12 as in the case where the camera is disposed at the upper part of the front window of the host vehicle. In this case, the moving direction of the light spot LP2 by the headlamp HL1 of the oncoming vehicle CAR12 is directed to the lower right with respect to the horizontal line H in the image as indicated by an arrow.

FIG. 18B is an image in the case where the camera is integrated with the headlamp as in this embodiment and is at a height or a position substantially equal to the headlamp HL1 of the oncoming vehicle CAR12. In this case, the moving direction of the light spot LP2 by the headlamp HL1 of the oncoming vehicle CAR12 is directed to the right along the horizontal line H in the image, or slightly lower right than the horizontal line H.

Therefore, the vehicle detection unit 1066 of the embodiment can detect the light spot moving in the horizontal direction in the image or in the lower right direction than the horizontal direction as the light spot of the oncoming vehicle. However, when the camera 102 provided on the host vehicle is tilted in the roll direction (rotation direction about the longitudinal axis of the automobile), for example, the camera 102 is mounted tilted in the left-down direction. Alternatively, when the vehicle body is tilted in the same direction due to a deviation in the load weight of the host vehicle, the image captured by the camera 102 is also tilted to the lower left as shown in FIG. 18C.

In such an image, the light spot LP2 of the headlamp HL1 of the oncoming car CAR12 is moved to the upper right side of the horizontal line H in the image. Therefore, when the light spot is detected only in the moving direction in the image, the light spot LP2 of the oncoming vehicle may be erroneously detected as a light spot other than the vehicle, for example, a sign or a road illumination light. Therefore, when detecting the vehicle, it is necessary to detect not only the moving direction of the light spot in the image but also the moving speed.

Therefore, the vehicle detection unit 1066 refers to the light spot that has been identified as the “non-target light spot” by the light spot identification unit 1065. For example, in the example of FIG. 18C, the vehicle detection unit 1066 refers to the light spot LP6 of the sign S1 identified as having a rectangular shape, and corrects the detected horizontal side, that is, the direction of the bottom side and the top side, in the corrected horizontal direction. Set as DH. Then, the vehicle detection unit 1066 corrects the moving direction of the light spot LP2 of the oncoming vehicle CAR12 in the image indicated by the arrow as the direction with respect to the corrected horizontal direction DH.

Usually, since the horizontal side of the rectangular sign S1 is oriented in the horizontal direction, even if the camera 102 or the vehicle body is tilted in the roll direction by performing this correction, the vehicle detection unit 1066 displays an image. Can be corrected to the corrected horizontal direction DH. If the moving direction of the light spot LP6 of the oncoming vehicle CAR12 is detected with reference to the corrected horizontal direction DH, the vehicle detection unit 1066 detects that the light spot LP6 is moving in the horizontal direction or the lower right direction. It is possible to accurately detect the oncoming vehicle CAR12. That is, even if the moving speed of the light spot is not detected at the time of detection, the vehicle can be accurately detected only by detecting the moving direction, thereby simplifying the processing and speeding up the detection.

In this case, when a plurality of rectangular light spots are identified in the image, an average of the bottom and top sides of the plurality of rectangles may be taken as the reference horizontal direction. Alternatively, the bottom side or the top side of the rectangular light spot having the largest dimension, which can easily identify the shape with high accuracy, may be employed. If a rectangular light spot is not identified, the reference horizontal direction may be set using a light spot of another shape. For example, in the case of a diamond-shaped light spot, a horizontal diagonal line may be used as the reference horizontal direction. If the vehicle has a vehicle height sensor or the like and the roll angle of the host vehicle can be detected, the moving direction of the light spot in the image may be corrected based on the detected roll angle. Good.

In the present disclosure, the light spot LP8 of the white line LINE identified by the light spot identifying unit 1065 may be used to further improve the accuracy of vehicle detection, particularly the accuracy of oncoming vehicle detection. In this embodiment, the camera 102 is provided integrally with the headlamp HL1, and is provided at a lower position closer to the road surface than when provided on the front window or side mirror. The white line LINE can be imaged more clearly. In particular, since the white line LINE is a light spot obtained by imaging light reflected from the headlamps of the oncoming vehicle and the host vehicle, it is difficult to normally capture images with high brightness. By providing in, it can image as a high-intensity light spot.

Therefore, it becomes possible to clearly capture each light spot LP8 of the slightly distant white line LINE illuminated by the headlamp of the host vehicle and the adjacent white line illuminated by the headlamp HL1 of the oncoming vehicle CAR12 with high brightness. Recognition of the white line LINE becomes easy and accurate. Accordingly, it becomes possible to accurately recognize a road state as to whether the road on which the vehicle is traveling is a straight road or a curved road.

When the road state can be accurately recognized, the vehicle detection unit 1066 is in the vicinity of the recognized white line light spot LP8 among the light spots of the plurality of “target light spots” identified by the light spot identifying unit 1065, and Only the light spot that exists at the position along the extension direction of the light spot LP8 and moves in the extension direction can be detected as the light spot of the vehicle. Therefore, by performing vehicle detection based only on the detected light spot based on the moving direction and moving speed of the light spot, the accuracy of vehicle detection is improved, and the processing steps for vehicle detection are simplified, and the detection speed is increased. Can be realized.

Further, by recognizing the road state using the light spot LP8 by the white line in this way, the moving direction of the light spot in the image is a direction close to the light spot LP8 of the white line and along the extending direction thereof. It is possible to accurately detect the vehicle simply by detecting. That is, even when the host vehicle rolls and the camera and the vehicle body are tilted, accurate vehicle detection is possible without considering the corrected horizontal direction as described in FIG. 18C. The complexity of processing can be avoided.

When the white line is a curve, that is, when the road state is a curved road, the vehicle detection unit 1066 calculates the curvature from the shape of the light spot LP8 when it is identified by the light spot identification unit 1065. When the target light spot in the image is moving in the horizontal direction, the vehicle detection unit 1066 performs correction such as subtraction or addition of the movement amount based on the direction of the curved path and the calculated curvature. It may be. This makes it possible to accurately detect the vehicle from the light spot.

In the above embodiment, an example in which the vehicle detection device of the present disclosure is used for light distribution control by ADB control of a headlamp has been shown, but it may be used to perform AHB control or other light distribution control. Further, since the vehicle can be detected quickly and accurately, it can also be used for automatic driving control of an automobile.

Also, the vehicle detection device of the present disclosure can be configured to take an image of a side region and a rear region of an automobile and detect other vehicles existing in the side and rear regions of the host vehicle. In this case, the camera may be integrated into a side mirror or tail lamp.

This application includes the contents disclosed in the Japanese patent application filed on March 15, 2018 (Japanese Patent Application No. 2018-048049) and the Japanese patent application filed on March 15, 2018 (Japanese Patent Application No. No. 2018-048050) is appropriately incorporated.

Claims (16)

1. A lamp unit that irradiates light, an imaging unit that images at least a light irradiation region of the lamp unit, a control unit that is connected to a signal bus disposed in a vehicle and controls the lamp unit and the imaging unit; A vehicle lamp configured to perform an aiming adjustment of the lamp unit by performing signal processing on an image pickup signal picked up by the image pickup means, and when the control means receives a predetermined command signal, A vehicle lamp characterized by outputting a signal-processed signal to the signal bus.
2. The vehicle lamp according to claim 1, wherein the signal-processed signal is an image signal based on an imaging signal or an aiming adjustment signal.
3. An aiming adjuster can be connected to the signal bus, and the aiming adjuster outputs the command signal to the signal bus and acquires the signal processed signal output to the signal bus, The vehicle lamp according to claim 2, wherein the aiming state of the lamp unit is displayed based on the acquired signal, or the aiming adjustment is executed based on the acquired signal.
4. 4. The vehicle lamp according to claim 1, wherein the camera and the lamp unit are housed in a lamp housing, and the camera is detachably supported with respect to the lamp housing.
5. 5. The vehicle lamp according to claim 4, wherein a vehicle lamp is disposed on each side of the vehicle, and the camera is disposed on a selected one of the vehicle lamps.
6. The control means includes at least another vehicle detection means for detecting another vehicle based on an image signal captured by the camera, and the other vehicle detection means is machine-learned based on automatically generated traveling simulation data. The vehicle lamp according to claim 1, further comprising an image analysis unit.
7. 7. The vehicle lamp according to claim 6, wherein the running simulation data includes automatically generated running video data and teacher data.
8. A vehicle detection method for detecting a vehicle from the behavior of a light spot that is mounted on a vehicle and that captures the surroundings of the vehicle and that is captured in an image obtained by the imaging, wherein a plurality of light spots present in the image are detected. A step of recognizing a shape, a step of identifying a light spot that is a vehicle detection target in the recognized shape as a target light point, and a light spot other than that as a non-target light point, and a behavior of the target light spot in the image And detecting a light spot having a predetermined behavior as a vehicle.
9. The vehicle detection method according to claim 8, wherein when detecting the behavior of the target light spot, the shape of the non-target light spot is referred to.
10. A vehicle detection apparatus comprising: an imaging unit that is mounted on a vehicle and images the surroundings of the vehicle; and a detection unit that detects a vehicle from a light spot captured in an image captured by the imaging unit, and is present in the image A light spot identifying unit for identifying a target light spot to be detected from a plurality of light spots, a non-target light spot, and a vehicle detection for detecting a vehicle from the behavior of the identified target light spot in the image A vehicle detection device that recognizes shapes of the plurality of light spots and identifies the target light spot and the non-target light spot based on the recognition.
11. The vehicle detection device according to claim 10, wherein the light spot identifying unit identifies a light spot having a rectangular shape, a diamond shape, or a line shape as a non-target light spot, and identifies a light spot having a circular shape or a shape close thereto as a target light spot. .
12. The vehicle detection device according to claim 10 or 11, wherein the vehicle detection unit detects a vehicle based on a moving direction of the target light spot in the image.
13. The vehicle detection device according to claim 12, wherein the vehicle detection unit corrects a moving direction of the target light spot based on a shape recognized by the non-target light spot.
14. When the shape recognized from the non-target light spot is a rectangle, the vehicle detection unit sets the direction of the bottom or top of the rectangle as a reference horizontal direction, and moves the target light spot based on the reference horizontal direction. The vehicle detection device according to claim 13, wherein the direction is corrected.
15. The vehicle detection device according to claim 12, wherein the vehicle detection unit recognizes a white line on a road from a linear non-target light spot and detects a light spot in a moving direction along the white line as a vehicle.
16. The vehicle detection device according to any one of claims 10 to 15, wherein the imaging unit is provided integrally with a lamp of the vehicle.

Images