JP2020160850A - Processing device - Google Patents

Abstract

To provide technology to improve the accuracy of detecting tailgating.SOLUTION: A first detection unit 300 detects whether or not there is a following vehicle whose distance from an own vehicle is equal to or less than a second threshold value based on a first image of a first range captured by a rear imaging device 10 when the traveling speed of the own vehicle is equal to or higher than a first threshold value. A second detection unit 310 detects whether or not the following vehicle is tailgating based on the first image when the first detection unit 300 detects the presence of the following vehicle. A notification unit 366 gives warning information when the second detection unit 310 detects the presence of the following vehicle that is tailgating. A control unit 360 records a second image of a second range captured by an omnidirectional imaging device 20 when the first detection unit 300 detects the presence of the following vehicle, or when the notification unit 366 gives the warning information.SELECTED DRAWING: Figure 2

Description

The present disclosure relates to a detection technique, and more particularly to a processing device for detecting a following vehicle during road rage.

It is required to increase the possibility of avoiding a collision between the own vehicle and the following vehicle. Therefore, by comparing the image of the behavior of the following vehicle with the unnatural behavior pattern, the following vehicle that behaves unnaturally is detected. Unnatural behavior patterns include behavior patterns that may cause dozing driving, behavior patterns that may cause drunk driving, behavior patterns that correspond to dangerous driving such as meandering driving, shoulder driving, and sudden acceleration / deceleration (behavior patterns that correspond to dangerous driving such as meandering driving, road shoulder driving, and sudden acceleration / deceleration). For example, see Patent Document 1).

Japanese Patent No. 5748030

Road rage is a dangerous act that causes a serious accident or trouble. In the case of road rage, there are peculiar behaviors such as meandering while maintaining the distance between vehicles or excessive passing, and the conventional technology is not suitable for detecting road rage.

The present disclosure has been made in view of such circumstances, and an object thereof is to provide a technique for improving the detection accuracy of road rage.

In order to solve the above problems, the processing device of a certain aspect of the present invention is a first image of a first range captured by a first imaging device when the traveling speed of the own vehicle is equal to or higher than the first threshold value. When the first detection unit that detects whether the distance between the own vehicle and the following vehicle is equal to or less than the second threshold value and the first detection unit detects the presence of the following vehicle, the first Based on the video, the second detection unit that detects whether the following vehicle is driving or not, and when the second detection unit detects the presence of the following vehicle that is driving, it notifies the warning information. When the notification unit and the first detection unit detect the presence of the following vehicle, or when the notification unit notifies the warning information, the second image of the second range captured by the second imaging device is recorded. It includes a control unit.

It should be noted that any combination of the above components and the conversion of the expression of the present disclosure between methods, devices, systems, recording media, computer programs and the like are also effective as aspects of the present disclosure.

According to the present disclosure, it is possible to improve the detection accuracy of the road rage.

It is a figure which shows the own vehicle and the following vehicle which concerns on Example. It is a figure which shows the structure of the processing apparatus mounted on the own vehicle of FIG. 3A and 3B are diagrams showing an outline of processing of the second detection unit of FIG. It is a figure which shows the processing outline of the 2nd detection part of FIG. It is a figure which shows the other processing outline of the 2nd detection part of FIG. 6 (a)-(b) are views showing a screen displayed in the notification unit of FIG. It is a figure which shows another screen which is displayed in the notification part of FIG. It is a figure which shows the processing outline of the control part of FIG. It is a flowchart which shows the detection procedure by the processing apparatus of FIG.

Before concretely explaining this embodiment, the outline of the embodiment will be described. This embodiment relates to a processing device that detects a tilting operation by a following vehicle. Until now, technology that detects tilting driving using an inter-vehicle distance sensor and guides the driver to a place where shelter can be evacuated, and warns when the behavior of surrounding vehicles is judged to be unnatural behavior by comparing with stored data. Technology is being considered. However, as described above, these techniques do not detect behavior peculiar to road rage such as meandering while maintaining the distance between vehicles or excessive passing. In addition, if the road rage develops into an accident or trouble, it is desirable to leave a video recording from the start to the end of the road rage as evidence.

The processing device according to the present embodiment detects the behavior peculiar to the road rage by capturing the scenery behind the vehicle using the image pickup device and executing the image recognition process on the captured image. When it detects a following vehicle that performs road rage, the processing device alerts the driver and proposes appropriate countermeasures. Further, when the following vehicle that executes the tilting operation is detected, the processing device records the image captured by the omnidirectional image pickup device. This recording continues until there are no vehicles behind. Hereinafter, examples of the present disclosure will be described in detail with reference to the drawings. It should be noted that each of the examples described below is an example, and the present disclosure is not limited to these examples.

FIG. 1 shows the own vehicle 100 and the following vehicle 200. The own vehicle 100 is a vehicle on which the processing device (not shown) according to the present embodiment is mounted, and is a vehicle that may be driven in a tilted manner. At the rear of the own vehicle 100, a rear imaging device 10 for imaging a rear view is installed. Further, in the own vehicle 100, a first omnidirectional image pickup device 20a, a second omnidirectional image pickup device 20b, a third omnidirectional image pickup device 20c, and a fourth omnidirectional image pickup device 20d, which are collectively called the omnidirectional image pickup device 20, are installed. Will be done. The first omnidirectional image pickup device 20a is installed in the front part of the own vehicle 100 to image the scenery in front, and the second omnidirectional image pickup device 20b is installed in the left part of the own vehicle 100 to capture the left side view. Take an image. The third omnidirectional image pickup device 20c is installed at the rear part of the own vehicle 100 and images the rear view, and the fourth omnidirectional image pickup device 20d is installed at the right part of the own vehicle 100 and images the right side view. To do. The external microphone 30 is installed at the rear of the own vehicle 100, for example, and acquires external voice. The following vehicle 200 is a vehicle that travels behind the own vehicle 100 and has a possibility of performing a tilting operation.

FIG. 2 shows the configuration of the processing device 1000 mounted on the own vehicle 100. The processing device 1000 includes a first detection unit 300, a second detection unit 310, a sensor 320, and an in-vehicle device 350. The sensor 320 includes a speed sensor 322 and a steering angle sensor 324, and the in-vehicle device 350 includes a control unit 360, an operation unit 362, a notification unit 366, a storage device 368, and a communication device 370. A rear image pickup device 10, an omnidirectional image pickup device 20, an external microphone 30, and a timer 40 are connected to the processing device 1000. Hereinafter, the configuration of the processing device 1000 will be described in the order of (1) first detection processing, (2) second detection processing, and (3) post-detection processing.

(1) First Detection Process The first detection process is a process for detecting the following vehicle 200 in the following state. The following state is a state in which the own vehicle 100 is being followed, and is not determined to be a tilting operation, but is a state in which there is a possibility of a tilting operation. The speed sensor 322 detects the traveling speed of the own vehicle 100 and outputs the traveling speed information to the first detection unit 300. The first detection unit 300 receives information on the traveling speed from the speed sensor 322. The first detection unit 300 receives a rear image (hereinafter, referred to as "first image") from the rear image pickup device 10 when the traveling speed of the own vehicle 100 is a first threshold value, for example, 30 km / h or more. The imaging range in the first image is referred to as, for example, the "first range".

The first detection unit 300 detects whether or not the following vehicle 200 is included in the first image by executing the image recognition process on the first image. When the following vehicle 200 is included in the first image, the first detection unit 300 specifies the inter-vehicle distance between the own vehicle 100 and the following vehicle 200 from the first image, and the inter-vehicle distance is a second threshold value, for example. Identify whether it is 5 m or less. This corresponds to detecting whether or not there is a following vehicle 200 whose inter-vehicle distance from the own vehicle 100 is equal to or less than the second threshold value. When the inter-vehicle distance is equal to or less than the second threshold value, the first detection unit 300 determines that the following vehicle 200 is in the following state. On the other hand, if the first detection unit 300 does not satisfy any of the conditions, the first detection unit 300 determines that the following vehicle 200 is not in the following state.

When the first detection unit 300 determines that the following vehicle 200 is in the following state, the first detection unit 300 instructs the second detection unit 310 to execute the second detection process, and when it determines that the following vehicle 200 is not in the following state, the first detection unit 300 performs the process. finish. Even after instructing the execution of the second detection process, the first detection unit 300 continues the execution of the first detection process, and when any of the conditions is not satisfied, the first detection unit 300 causes the second detection unit 310. Is instructed to end the second detection process.

(2) Second detection process The second detection process is a process for detecting the following vehicle 200 in which the road rage is being performed, and is performed on the following vehicle 200 in the following state. In the second detection process, (i) detection of continuation of the following state, (ii) detection of passing, (iii) detection of horn, and (iv) detection of meandering operation are performed.

(I) Detection of Continuation of Follow-up State When the second detection unit 310 receives an instruction to execute the second detection process from the first detection unit 300, the timer 40 causes the timer 40 to start measurement. The timer 40 measures the period after the follow-up state is detected. When the second detection unit 310 receives the instruction to end the second detection process from the first detection unit 300, the timer 40 terminates the measurement. When the period measured by the timer 40 is a predetermined period, for example, 5 seconds or more, the second detection unit 310 determines the detection of the continuation of the tracking state.

(Ii) Passing Detection When the second detection unit 310 receives an instruction to execute the second detection process from the first detection unit 300, the second detection unit 310 receives the first image from the rear imaging device 10. The second detection unit 310 derives the brightness in each of the plurality of images forming the first image. For example, the brightness for an image is calculated by integrating the brightness of a plurality of pixels forming one image. The second detection unit 310 derives the brightness of the background by calculating the average value of the brightness of the plurality of images. In addition, the second detection unit 310 measures the number of transitions from a value smaller than the brightness of the background to a value larger than the brightness of the background in a predetermined period. For example, the predetermined period is set between 1 second and 10 seconds, and a value larger than the brightness of the background is set to a value twice the brightness of the background. When the number of times measured is a predetermined number of times, for example, three times or more, the second detection unit 310 detects that the following vehicle 200 being passed is present. On the other hand, if such a condition is not satisfied, the second detection unit 310 detects that there is no following vehicle 200 being passed.

FIGS. 3A and 3B show an outline of processing of the second detection unit 310. This shows the change in the brightness of the image over time. In FIG. 3A, the number of transitions from the vicinity of the brightness of the background to a value larger than the brightness of the background is three times. Therefore, the second detection unit 310 detects the presence of the following vehicle 200 that is passing in the case shown in FIG. 3A. In FIG. 3B, it is detected that the following vehicle 200 is passing when the time during which a value larger than the brightness of the background is continued from the vicinity of the brightness of the background is equal to or longer than the threshold value. Return to FIG.

(Iii) Horn detection When the second detection unit 310 receives an instruction to execute the second detection process from the first detection unit 300, the second detection unit 310 receives an external voice from the external microphone 30. The second detection unit 310 measures the volume of external voice. When the measured volume is a predetermined value, for example, 95 dB or more, the second detection unit 310 detects that the following vehicle 200 that sounds the horn is present. On the other hand, if such a condition is not satisfied, the second detection unit 310 detects that the following vehicle 200 that sounds the horn does not exist. FIG. 4 shows an outline of processing of the voice recognition unit 312. This indicates a change in volume over time. Here, the volume is 95 dB or more. Therefore, the second detection unit 310 detects the existence of the following vehicle 200 that sounds the horn in the case as shown in FIG. Return to FIG.

(Iv) Detection of meandering operation When the second detection unit 310 receives an instruction to execute the second detection process from the first detection unit 300, the second detection unit 310 receives the first image from the rear imaging device 10. The second detection unit 310 detects whether or not there is a following vehicle 200 that is meandering based on the first image. Here, this process will be described with reference to FIG.

FIG. 5 shows another processing outline of the second detection unit 310. This indicates one of a plurality of images (hereinafter, referred to as “image”) forming the first image. The image includes the following vehicle 200, and the license plate 210 is arranged in the central portion of the front portion of the following vehicle 200. The axis extending in the front-rear direction through the license plate 210 is the following vehicle central axis 400. Further, the distance between the following vehicle 200 and the own vehicle 100 along the following vehicle central axis 400 is the inter-vehicle distance 410. On the other hand, the axis extending in the front-rear direction through the center of the own vehicle 100 in the left-right direction is the own vehicle central axis 420.

Here, the own vehicle 100 may swing left and right while traveling. As a result, the following vehicle 200 may appear to be relatively meandering when viewed from the own vehicle 100, even if the following vehicle 200 is not meandering. In order to prevent this, the second detection unit 310 may adjust the position of the own vehicle central axis 420 based on the steering angle received from the steering angle sensor 324. Further, the deviation between the following vehicle central axis 400 and the own vehicle central axis 420 is the central deviation 430. When the following vehicle 200 is performing meandering driving, the following vehicle 200, in particular the license plate 210, moves on the travel path 440 over a plurality of images.

The second detection unit 310 sets the own vehicle center region 450 centered on the own vehicle central axis 420 in order to detect the presence of the following vehicle 200 during meandering operation. The own vehicle center region 450 has a width of, for example, 50 cm to 1 m. The second detection unit 310 measures the number of times the travel path 440 enters the own vehicle center region 450 and the travel path 440 exits the own vehicle center region 450 in a predetermined period. Further, the second detection unit 310 may measure the number of times the traveling path 440 crosses the own vehicle center region 450. The second detection unit 310 detects that there is a following vehicle 200 in meandering operation when the number of times measured is equal to or greater than the threshold value, and when the number of times measured is smaller than the threshold value, the second detection unit 310 is in meandering operation. Detects that the following vehicle 200 does not exist.

Alternatively, the second detection unit 310 may calculate the variance value of the center deviation 430. At that time, the second detection unit 310 detects that there is a following vehicle 200 in meandering operation when the dispersion value is equal to or more than the threshold value, and when the dispersion value is smaller than the threshold value, meandering operation. It detects that the following vehicle 200 inside does not exist. That is, the second detection unit 310 detects whether or not the following vehicle 200 in meandering operation exists based on the change in the relative position of the following vehicle 200 with respect to the own vehicle 100. Return to FIG.

If at least one of (i) to (iv) is satisfied, the second detection unit 310 detects the presence of the following vehicle 200 during the tilting operation. Here, if all of (i) to (iv) are satisfied, the second detection unit 310 may detect the presence of the following vehicle 200 during the tilting operation. The processes (i) to (iv) may be executed in parallel or serially. When it is executed serially, for example, when the process of (iii) is executed and it is detected that the following vehicle 200 being passed does not exist, the process of (iii) is executed.

(3) Post-Detection Processing When the second detection unit 310 detects the presence of the following vehicle 200 during the tilting operation, the control unit 360 notifies the following vehicle 200 of information for alerting the following vehicle 200 from the notification unit 366. The notification unit 366 is, for example, a display device, and displays a screen on which information for calling attention is shown. 6 (a)-(b) show a screen displayed by the notification unit 366. In FIG. 6A, a warning message is displayed stating, "There is a possibility that the vehicle is being hit by a following vehicle." This sounds a horn when it detects the presence of the following vehicle 200 that continues to follow, detects the presence of the following vehicle 200 during meandering driving, or detects the presence of the following vehicle 200 during passing. If at least one of the cases where the presence of the following vehicle 200 is detected is satisfied, it corresponds to the notification of the alert information. FIG. 6B will be described later, and the process returns to FIG.

The control unit 360 may perform guidance to an evacuable place by executing the route guidance function after notifying the attention alert information from the notification unit 366. An example of an evacuation site is a police station. In FIG. 6B, a warning message "There is a possibility of being hit by a following vehicle" is displayed, and a "Yes" button and a "No" button are also displayed. When the driver selects the "Yes" button, the control unit 360 executes route guidance to the nearest police station. FIG. 7 shows another screen displayed by the notification unit 366. As shown, the route to the police station is shown. Return to FIG.

When the first detection unit 300 determines that the following vehicle 200 is in the following state, or when the notification unit 366 notifies the warning information, the control unit 360 receives the image from each omnidirectional imaging device 20. The control unit 360 generates a bird's-eye view image by executing a viewpoint conversion process on the image from each omnidirectional image pickup device 20. When an image formed by a plurality of bird's-eye views images is called a "second image", the imaging range in the second image is called a "second range", and the second range is wider than the first range. The control unit 360 records the second video in the storage device 368.

The control unit 360 monitors the second video while the second video is being recorded in the storage device 368, and ends the recording of the second video when the vehicle no longer exists in the second video. FIG. 8 shows an outline of processing of the control unit 360. A second range 500 is formed around the own vehicle 100. When the following vehicle 200 existing in the second range 500 travels along the traveling path 440 and exits the second range 500, the control unit 360 ends the recording of the second video. Further, when the control unit 360 detects the presence of the following vehicle 200 during the tilting operation in the second detection unit 310, the control unit 360 may transmit the information from the communication device 370.

This configuration can be realized by a CPU (Central Processing Unit), memory, or other LSI (Large Scale Integration) of any computer in terms of hardware, and by a program loaded in memory in terms of software. However, here we are drawing a functional block realized by their cooperation. Therefore, it will be understood by those skilled in the art that these functional blocks can be realized in various forms only by hardware and by a combination of hardware and software.

The operation of the processing device 1000 with the above configuration will be described. FIG. 9 is a flowchart showing a detection procedure by the processing device 1000. The speed sensor 322 measures the vehicle speed (S10). If the speed of the own vehicle 100 is not 30 km / h or more (N in S12), the process returns to step 10. When the speed of the own vehicle 100 is 30 km / h or more (Y in S12), the first detection unit 300 measures the inter-vehicle distance 410 with the following vehicle 200 (S14). If the inter-vehicle distance 410 is not within 5 m (N in S16), the process returns to step 10.

When the inter-vehicle distance 410 is within 5 m (Y in S16), the control unit 360 starts recording by the omnidirectional image pickup device 20, and the second detection unit 310 starts the timer 40 and starts the measurement of meandering driving. (S18). Control if passing is not detected (N in S20), horn is not detected (N in S22), meandering operation is not detected (N in S24), and the following state does not continue for 5 seconds (N in S26). The unit 360 ends the recording by the omnidirectional image pickup device 20, and the second detection unit 310 stops the timer 40 and ends the measurement of the meandering operation (S28). Return to step 10.

When passing is detected (Y in S20), horn is detected (Y in S22), or meandering operation is detected (Y in S24), and the following state continues for 5 seconds (S26). Y), the second detection unit 310 detects the tilting operation (S30). The notification unit 366 executes a warning to the driver of the own vehicle 100 (S32). The control unit 360 selects control (S34). If the position of the following vehicle 200 is not outside the second range 500 (N in S36), the processing so far is continued. If the position of the following vehicle 200 is outside the second range 500 (Y in S36), the control unit 360 protects the recorded data of the storage device 368, ends the recording, stops the timer 40, and measures the meandering operation. (S38).

According to the embodiment of the present disclosure, since the presence of the following vehicle 200 during the meandering operation is detected, the detection accuracy of the tilting operation can be improved. Further, when the presence of the following vehicle 200 during meandering driving is detected, the second image is recorded, so that evidence can be left. Further, by measuring the number of times the traveling path 440 has entered and exited the center region 450 of the own vehicle, it is detected whether or not there is a following vehicle 200 that is meandering, so that the detection accuracy can be improved. Further, since it is detected whether or not there is a following vehicle 200 in meandering operation based on the dispersion value of the center deviation 430, the detection accuracy can be improved.

Further, since the presence of the following vehicle 200 during passing is detected, the detection accuracy of the tilting operation can be improved. Further, since the presence of the following vehicle 200 that sounds the horn is detected, the detection accuracy of the tilting operation can be improved. In addition, since the information for alerting is notified when the tilting driving is detected, the driver can be alerted. In addition, since the guidance to the evacuation area is notified, the occurrence of trouble can be avoided. Further, when the vehicle no longer exists, the recording of the second video is terminated, so that the recording can be automatically terminated.

The outline of one aspect of the present disclosure is as follows. The processing device 1000 of a certain aspect of the present disclosure is based on the first image of the first range captured by the rear imaging device 10 when the traveling speed of the own vehicle 100 is equal to or higher than the first threshold value. When the first detection unit 300 that detects whether or not there is a following vehicle 200 whose distance 410 from the vehicle 100 is equal to or less than the second threshold value and the first detection unit 300 detects the existence of the following vehicle 200. When the second detection unit 310 for detecting whether or not the following vehicle 200 is driving in a tilted manner and the second detection unit 310 detect the presence of the following vehicle 200 in a tilted driving based on the first image, When the notification unit 366 for notifying the attention alert information and the first detection unit 300 detect the presence of the following vehicle 200, or when the notification unit 366 notifies the attention alert information, the image is captured by the omnidirectional imaging device 20. It also includes a control unit 360 that records a second image in the second range.

According to this aspect, since the presence of the following vehicle 200 during the meandering operation is detected, the detection accuracy of the tilting operation can be improved.

The second detection unit 310 may detect whether or not there is a following vehicle 200 that is meandering based on a change in the position of the following vehicle 200 relative to the own vehicle 100. In this case, it is possible to improve the detection accuracy because it is detected whether or not there is a following vehicle 200 that is meandering based on the change in the relative position of the following vehicle 200 with respect to the own vehicle 100.

When the first detection unit 300 detects the presence of the following vehicle 200, the second detection unit 310 detects whether or not the following vehicle 200 is passing, and the notification unit 366 causes the second detection unit 310 to detect the presence of the following vehicle 200. When the presence of the following vehicle 200 during passing is detected, the warning information may be notified. In this case, since the presence of the following vehicle 200 during passing is detected, the detection accuracy of the tilting operation can be improved.

When the first detection unit 300 detects the presence of the following vehicle 200, the second detection unit 310 detects whether or not the following vehicle 200 that sounds the horn exists based on an external voice, and the notification unit 366. When the second detection unit 310 detects the presence of the following vehicle 200 that sounds the horn, may notify the warning information. In this case, since the presence of the following vehicle 200 that sounds the horn is detected, the detection accuracy of the tilting operation can be improved.

The notification unit 366 may notify the guidance to the evacuation place after notifying the information of the alert. In this case, since the guidance to the evacuation area is notified, the occurrence of trouble can be avoided.

The control unit 360 may end the recording of the second video when the vehicle no longer exists in the second video. In this case, when the vehicle no longer exists, the recording of the second video is terminated, so that the recording can be automatically terminated.

The present disclosure has been described above based on the examples. This embodiment is an example, and it will be understood by those skilled in the art that various modifications are possible for each of these components or combinations of each processing process, and that such modifications are also within the scope of the present disclosure. ..

10 Rear imager, 20 Omnidirectional imager, 30 External microphone, 40 Timer, 100 Own vehicle, 200 Subsequent vehicle, 210 Number plate, 300 1st detector, 310 2nd detector, 320 sensor, 322 speed sensor, 324 Steering angle sensor, 350 in-vehicle device, 360 control unit, 362 operation unit, 366 notification unit, 368 storage device, 370 communication device, 1000 processing device.

Claims (6)

When the traveling speed of the own vehicle is equal to or higher than the first threshold value, the distance between the own vehicle and the following vehicle is second based on the first image of the first range captured by the first imaging device. The first detector that detects whether or not it is below the threshold value,
When the first detection unit detects the presence of the following vehicle, the second detection unit that detects whether or not the following vehicle is driving based on the first image, and
When the second detection unit detects the presence of the following vehicle while driving, a notification unit that notifies information of alerting and
When the first detection unit detects the presence of the following vehicle, or when the notification unit notifies the information of the warning, the second image of the second range captured by the second imaging device is recorded. Control unit and
A processing device characterized by comprising. The second detection unit according to claim 1, wherein the second detection unit detects whether or not there is a following vehicle during meandering operation based on a change in the relative position of the following vehicle with respect to the own vehicle. Processing equipment. When the first detection unit detects the presence of the following vehicle, the second detection unit detects whether or not there is a following vehicle being passed.
The processing device according to claim 1 or 2, wherein the notification unit notifies information of a warning when the second detection unit detects the presence of the following vehicle during passing. When the first detection unit detects the presence of the following vehicle, the second detection unit detects whether or not the following vehicle that sounds the horn exists based on an external voice.
The processing device according to any one of claims 1 to 3, wherein the notification unit notifies information of alerting when the second detection unit detects the presence of the following vehicle that sounds a horn. .. The processing device according to any one of claims 1 to 4, wherein the notification unit notifies guidance to a place where evacuation is possible after notifying the information of the alert. The processing device according to any one of claims 1 to 5, wherein the control unit ends recording of the second video when the vehicle no longer exists in the second video.

Images