Drive recorder device, video recording method, and program
The drive recorder device adjusts the acceleration threshold based on door closure events to accurately initiate video recording, addressing variability issues and ensuring event capture.
Patent Information
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- TOYOTA JIDOSHA KK
- Filing Date
- 2023-10-27
- Publication Date
- 2026-06-23
AI Technical Summary
Existing drive recorder systems face challenges in accurately determining when to start video recording due to the variability of acceleration threshold values based on vehicle type and sensor mounting position, leading to potential missed recordings during door closure or accidental events.
A drive recorder device that uses an acceleration sensor to detect impacts exceeding a predetermined threshold when ignition is off, combined with a learning mechanism to adjust the threshold based on door closure events, ensuring appropriate video recording initiation.
The system sets an appropriate acceleration threshold for video recording, reducing false starts and ensuring accurate capture of events, including accidents, by learning from door closure impacts.
Smart Images

Figure 0007878256000001 
Figure 0007878256000002
Abstract
Description
Technical Field
[0001] The present disclosure relates to a drive recorder device, a video recording method, and a program.
Background Art
[0002] Patent Document 1 describes a technique for starting video shooting by a camera and recording the captured video when vibration is detected by an acceleration sensor, and stopping video recording when an opening / closing signal of a vehicle door is detected within a predetermined time after vibration is detected by the acceleration sensor.
Prior Art Documents
Patent Documents
[0003]
Patent Document 1
Summary of the Invention
Problems to be Solved by the Invention
[0004] As in the technique described in Patent Document 1, if one tries to determine whether the vehicle door has changed from the open state to the closed state or an accident has occurred only from the timing of the impact detected by the acceleration sensor, there is a risk that video cannot be recorded when the timing of the vehicle door changing from the open state to the closed state overlaps with the timing of an accident occurring. Even if one tries to determine whether the vehicle door has changed from the open state to the closed state or an accident has occurred by setting an acceleration threshold value, it is not easy to set an appropriate acceleration threshold value because it varies depending on the vehicle and also depends on the mounting position of the acceleration sensor.
[0005] In view of the above, the present disclosure aims to provide a drive recorder device, a video recording method, and a program that can set an appropriate value for the acceleration threshold used to determine whether or not to start video recording when the vehicle's ignition power is turned off. [Means for solving the problem]
[0006] One aspect of the present disclosure is a drive recorder device comprising: a control unit that starts recording video when an impact exceeding a predetermined acceleration threshold is detected by an acceleration sensor while the vehicle's ignition power is turned off; a determination unit that determines whether or not the vehicle's door has changed from an open state to a closed state; and a learning unit that, when the door has changed from an open state to a closed state, learns the acceleration threshold using the impact detected by the acceleration sensor during a learning period between a first time point in time when the door changed from an open state to a closed state and a second time point set back a predetermined time from the first time point. One aspect of the present disclosure is a video recording method comprising: a control step in which a drive recorder device starts recording video when an impact exceeding a predetermined acceleration threshold is detected by an acceleration sensor while the vehicle's ignition power is turned off; a determination step in which the drive recorder device determines whether or not the vehicle's door has changed from an open state to a closed state; and a learning step in which, when the door has changed from an open state to a closed state, the drive recorder device learns the acceleration threshold using the impact detected by the acceleration sensor during a learning period between a first time point in time when the door changed from an open state to a closed state and a second time point set back a predetermined time from the first time point. One aspect of the present disclosure is a program that causes a processor to perform the following steps: a control step of starting video recording when an impact exceeding a predetermined acceleration threshold is detected by an acceleration sensor while the vehicle's ignition power is turned off; a determination step of determining whether or not the vehicle's door has changed from an open state to a closed state; and a learning step of learning the acceleration threshold using the impact detected by the acceleration sensor during a learning period between a first time point in time when the door changed from an open state to a closed state and a second time point set back a predetermined time from the first time point. [Effects of the Invention]
[0007] According to this disclosure, the acceleration threshold used to determine whether or not to start video recording when the vehicle's ignition power is turned off can be set to an appropriate value. [Brief explanation of the drawing]
[0008] [Figure 1] This figure shows an example of a vehicle 1 to which the drive recorder device 16 of the first embodiment is applied. [Figure 2] This is a flowchart illustrating an example of processing performed by the processor 163 of the drive recorder device 16 of the first embodiment. [Modes for carrying out the invention]
[0009] The embodiments of the drive recorder device, video recording method, and program of this disclosure will be described below with reference to the drawings.
[0010] <First Embodiment> Figure 1 shows an example of a vehicle 1 to which the drive recorder device 16 of the first embodiment is applied. In the example shown in Figure 1, vehicle 1 is equipped with a camera 11, a door opening / closing sensor 12, an acceleration sensor 13, an ignition power supply 14, a constant power supply 15, and a drive recorder device 16. In the example shown in Figure 1, vehicle 1 is equipped with a camera 11 as a surrounding environment sensor and a door open / close sensor 12 and an acceleration sensor 13 as vehicle state sensors. However, in other examples, vehicle 1 may be equipped with surrounding environment sensors other than the camera 11 (e.g., LiDAR (Laser Imaging Detection and Ranging), radar, sonar, etc.) or vehicle state sensors other than the door open / close sensor 12 and acceleration sensor 13 (e.g., a yaw rate sensor, etc.).
[0011] In the example shown in Figure 1, the camera 11 is positioned on the vehicle 1, for example, on the windshield or rear window, and takes pictures of the front, rear, sides, and interior of the vehicle 1, and transmits the images of the front, rear, sides, and interior of the vehicle 1 to the drive recorder device 16. More specifically, the camera 11 not only takes pictures and transmits the images to the drive recorder device 16 when the ignition power supply 14 is on, but also has the function of taking pictures and transmitting the images to the drive recorder device 16 when the ignition power supply 14 is off (when the vehicle 1 is parked). In other examples, the position (placement) of camera 11 may differ from the example shown in Figure 1, or the object that camera 11 photographs may differ from the example shown in Figure 1.
[0012] In the example shown in Figure 1, the door open / close sensor 12 detects the open / closed state of the vehicle's door (not shown) and transmits a signal indicating the door's open / closed state (door open / closed signal) to the drive recorder device 16. More specifically, the door open / close sensor 12 not only detects the door's open / closed state and transmits a door open / closed signal to the drive recorder device 16 when the ignition power supply 14 is on, but also has the function of detecting the door's open / closed state and transmitting a door open / closed signal to the drive recorder device 16 when the ignition power supply 14 is off (when the vehicle 1 is parked). The acceleration sensor 13 detects acceleration, such as an impact acting on the vehicle 1, and transmits the detection result (acceleration signal) to the drive recorder device 16. More specifically, the acceleration sensor 13 not only detects acceleration and transmits the acceleration signal to the drive recorder device 16 when the ignition power supply 14 is on, but also has the function of detecting acceleration and transmitting the acceleration signal to the drive recorder device 16 when the ignition power supply 14 is off (when the vehicle 1 is parked). The ignition power supply 14 supplies power to devices (not shown) connected to it when the ignition power supply 14 is turned on. When the ignition power supply 14 is turned off, the ignition power supply 14 transmits a signal to the drive recorder device 16 indicating that the ignition power supply 14 is off. The constant power supply 15 supplies power to the drive recorder device 16, etc.
[0013] In the example shown in Figure 1, the drive recorder device 16 is comprised of a computer (for example, a drive recorder ECU (Electronic Control Unit)) having a communication interface (I / F) 161, memory 162, and a processor 163. The communication interface 161 is connected to, for example, the camera 11, the door opening / closing sensor 12, the acceleration sensor 13, the ignition power supply 14, etc. The memory 162 stores (records) predetermined acceleration thresholds, predetermined time thresholds, video and signals acquired by the acquisition unit 163A (described later), and programs that realize the functions of the processor 163. The processor 163 has the functions of an acquisition unit 163A, a determination unit 163B, a control unit 163C, and a learning unit 163D. The acquisition unit 163A acquires video transmitted from the camera 11. The acquisition unit 163A also acquires the door open / close signal transmitted from the door open / close sensor 12. Furthermore, the acquisition unit 163A acquires the acceleration signal transmitted from the acceleration sensor 13. In addition, the acquisition unit 163A acquires a signal transmitted from the ignition power supply 14 indicating that the ignition power supply 14 is turned off.
[0014] The determination unit 163B determines whether an impact exceeding an acceleration threshold stored in the memory 162 has been detected by the acceleration sensor 13. The determination unit 163B also determines whether the door of the vehicle 1 has changed from an open state to a closed state based on the door opening / closing signal acquired by the acquisition unit 163A. Furthermore, the determination unit 163B determines whether the video transmitted from the camera 11 is being recorded in the memory 162. The determination unit 163B also determines whether the elapsed time since the start of video recording to the memory 162 has exceeded a time threshold (e.g., 60 seconds) stored in the memory 162. Furthermore, the determination unit 163B determines whether the ignition power supply 14 is turned off based on a signal transmitted from the ignition power supply 14.
[0015] The control unit 163C executes the process of starting video recording to the memory 162. Specifically, when the ignition power supply 14 of the vehicle 1 is turned off, the control unit 163C starts recording video to the memory 162 when the acceleration sensor 13 detects an impact exceeding an acceleration threshold stored in the memory 162. Furthermore, the control unit 163C executes a process to terminate the recording of video to the memory 162. Specifically, the control unit 163C terminates the recording of video to the memory 162 when the elapsed time since the start of video recording to the memory 162 exceeds a time threshold stored in the memory 162.
[0016] The learning unit 163D learns the acceleration threshold value stored in the memory 162 by using the impact (acceleration) detected by the acceleration sensor 13. Specifically, when the door of the vehicle 1 changes from the open state to the closed state, the learning unit 163D uses the impact detected by the acceleration sensor 13 during the learning period between the first time point when the door of the vehicle 1 changes from the open state to the closed state and the second time point that is a predetermined time (for example, 900 ms, etc.) back from the first time point, to learn the acceleration threshold value stored in the memory 162. Specifically, the learning unit 163D learns the acceleration threshold value stored in the memory 162 by using the maximum acceleration (impact) among the accelerations (impacts) detected multiple times by the acceleration sensor 13 during the learning period from the second time point to the first time point. That is, the learning unit 163D stores the maximum acceleration (impact) among the accelerations (impacts) detected multiple times by the acceleration sensor 13 during the learning period from the second time point to the first time point in the memory 162 as the acceleration threshold value after learning. In the example shown in FIG. 1, every time the ignition power supply 14 of the vehicle 1 is turned off, the acceleration (impact) is detected by the acceleration sensor 13 N times (N is an integer of 2 or more) during the above-described learning period. Further, every time the ignition power supply 14 of the vehicle 1 is turned off M times (M is an integer of 2 or more), the maximum acceleration (impact) among the (N×M) times of accelerations (impacts) detected by the acceleration sensor 13 is used to learn the acceleration threshold value. In another example, every time the ignition power supply 14 of the vehicle 1 is turned off, the maximum acceleration (impact) among the N times of accelerations (impacts) detected by the acceleration sensor 13 may be used to learn the acceleration threshold value.
[0017] FIG. 2 is a flowchart for explaining an example of the process executed by the processor 163 of the drive recorder device 16 according to the first embodiment. The process shown in FIG. 2 starts when the acquisition unit 163A acquires a signal indicating that the ignition power supply 14 transmitted from the ignition power supply 14 is turned off. In the example shown in FIG. 2, in step S10, acquisition unit 163A acquires the acceleration signal transmitted from acceleration sensor 13. Further, determination unit 163B determines whether an impact greater than or equal to a predetermined acceleration threshold value stored in memory 162 in a step not shown is detected by acceleration sensor 13 based on the acceleration signal acquired by acquisition unit 163A. If YES, the process proceeds to step S11; if NO, the process proceeds to step S12.
[0018] In step S11, acquisition unit 163A acquires the video transmitted from camera 11. Further, control unit 163C starts the process of recording the video acquired by acquisition unit 163A in memory 162. In step S12, acquisition unit 163A acquires the door opening / closing signal transmitted from door opening / closing sensor 12. Further, determination unit 163B determines whether the door of vehicle 1 has changed from the open state to the closed state based on the door opening / closing signal acquired by acquisition unit 163A. If YES, the process proceeds to step S13; if NO, the process proceeds to step S14.
[0019] In step S13, learning unit 163D performs learning of the acceleration threshold value stored in memory 162 by using the acceleration signal acquired by acquisition unit 163A. Specifically, learning unit 163D uses the impact (specifically, the acceleration signal acquired by acquisition unit 163A) detected by acceleration sensor 13 during the learning period between the first time point when it is determined that the door of vehicle 1 has changed from the open state to the closed state and the second time point that is a predetermined time (for example, 900 ms, etc.) back from the first time point to perform learning of the acceleration threshold value stored in memory 162.
[0020] In step S14, the determination unit 163B determines whether the video transmitted from the camera 11 (more specifically, the video acquired by the acquisition unit 163A) is being recorded in the memory 162. If the result is YES, the process proceeds to step S15. On the other hand, if the result is NO, the process returns to step S10. If step S10 is executed after step S13 has been executed, the acceleration threshold obtained by the execution of step S13 (the learned acceleration threshold) is used in step S10.
[0021] In step S15, the determination unit 163B determines whether the elapsed time since the start of recording video to the memory 162 has exceeded a predetermined time threshold (e.g., 60 seconds) stored in the memory 162 in a step not shown. If YES, the process proceeds to step S16; otherwise, it returns to step S12.
[0022] In step S16, the control unit 163C finishes recording the video acquired by the acquisition unit 163A (video transmitted from the camera 11) into the memory 162.
[0023] As described above, in the examples shown in Figures 1 and 2, the learning unit 163D learns the acceleration threshold. Therefore, even if a predetermined acceleration threshold is not suitable for vehicle 1, the learning unit 163D changes the acceleration threshold to an appropriate value through learning. As a result, in the examples shown in Figures 1 and 2, even if a predetermined acceleration threshold is not suitable for vehicle 1, video recording can be started appropriately by using the learned acceleration threshold. In other words, the risk of video recording being unnecessarily started when the door of vehicle 1 changes from an open state to a closed state can be suppressed.
[0024] In the examples shown in Figures 1 and 2, if an accident occurs during the learning period described above, a warning is issued prompting the user of vehicle 1 to repair vehicle 1, and the process shown in Figure 2 is stopped. As a result, the impact detection results associated with the accident are not used in learning the acceleration threshold, thus avoiding the acceleration threshold becoming excessively large after learning. Therefore, in the examples shown in Figures 1 and 2, if an accident occurs when the ignition power supply 14 is turned off after the acceleration threshold has been learned, video recording can be started appropriately. More specifically, even if an accident occurs simultaneously with the change from the open state to the closed state of the door of vehicle 1 when the ignition power supply 14 is turned off after the acceleration threshold has been learned, video recording can be started appropriately.
[0025] In the examples shown in Figures 1 and 2, when vehicle 1 is used by a user who detects a large impact by the acceleration sensor 13 when the door of vehicle 1 changes from an open state to a closed state (i.e., a user who closes the door with force), the acceleration threshold value learned by the learning unit 163D becomes larger than when vehicle 1 is used by a user who detects a small impact by the acceleration sensor 13 when the door of vehicle 1 changes from an open state to a closed state (i.e., a user who closes the door gently). Therefore, even when vehicle 1 is used by a user who detects a large impact by the acceleration sensor 13 when the door of vehicle 1 changes from an open state to a closed state, the risk of unnecessary video recording starting when the door of vehicle 1 changes from an open state to a closed state (the risk of a false determination that an accident has occurred) can be suppressed.
[0026] As described above, in the examples shown in Figures 1 and 2, the learning period is set between the first point in time when the door of vehicle 1 changes from the open state to the closed state and the second point in time a predetermined time (e.g., 900 ms) prior to the first point in time. In other words, the control design takes into account the delay of the acceleration signal. Therefore, it is not necessary to place the drive recorder device 16 and the acceleration sensor 13 in close proximity. This means that the acceleration sensor 13 can be shared between the drive recorder device 16 and other systems (not shown), thereby reducing the overall cost of vehicle 1.
[0027] <Second Embodiment> The vehicle 1 to which the drive recorder device 16 of the second embodiment is applied is configured in the same way as the vehicle 1 to which the drive recorder device 16 of the first embodiment is applied, as shown in Figure 1, except for the points described later.
[0028] As described above, in the example shown in Figure 1, the learning unit 163D learns the acceleration threshold stored in the memory 162 by using the maximum acceleration (impact) among the accelerations (impacts) detected multiple times by the acceleration sensor 13 during the learning period from the second time point to the first time point. On the other hand, in an example of the drive recorder device 16 of the second embodiment, the learning unit 163D learns the acceleration threshold stored in the memory 162 by using the standard deviation of acceleration (impact) detected multiple times by the acceleration sensor 13 during the learning period from the second time point to the first time point. In other words, the learning unit 163D stores the standard deviation of acceleration (impact) detected multiple times by the acceleration sensor 13 during the learning period from the second time point to the first time point as the learned acceleration threshold in the memory 162.
[0029] In other words, in the example of the drive recorder device 16 of the second embodiment, even if an accident occurs during the learning period from the second time point to the first time point, the standard deviation of the acceleration (impact) detected multiple times by the acceleration sensor 13 during the learning period from the second time point to the first time point is used for learning the acceleration threshold, and the impact (>standard deviation) associated with the occurrence of an accident detected by the acceleration sensor 13 during the learning period from the second time point to the first time point is not used. As a result, in an example of the drive recorder device 16 of the second embodiment, the learning unit 163D learns the acceleration threshold so that the acceleration threshold becomes greater than the impact detected by the acceleration sensor 13 when the door changes from an open state to a closed state, and less than or equal to the impact detected by the acceleration sensor 13 when an accident occurs. In other words, in an example of the drive recorder device 16 of the second embodiment, the learned acceleration threshold becomes greater than the impact detected by the acceleration sensor 13 when the door changes from an open state to a closed state, and less than or equal to the impact detected by the acceleration sensor 13 when an accident occurs.
[0030] <Third Embodiment> The vehicle 1 to which the drive recorder device 16 of the third embodiment is applied is configured in the same way as the vehicle 1 to which the drive recorder device 16 of the first embodiment is applied, as shown in Figure 1, except for the points that will be described later.
[0031] In an example of the drive recorder device 16 of the third embodiment, the learning unit 163D learns the acceleration threshold stored in the memory 162 by applying machine learning (for example, clustering such as DBSCAN (Density-Based Spatial Clustering of Applications with Noise)) to the acceleration (impact) detected multiple times by the acceleration sensor 13 during the learning period from the second time point to the first time point. In other words, the learning unit 163D stores in the memory 162 the learned acceleration threshold, which is the acceleration (impact) detected multiple times by the acceleration sensor 13 during the learning period from the second time point to the first time point, with noise such as impacts associated with accidents removed.
[0032] In other words, in the example of the drive recorder device 16 of the third embodiment, even if an accident occurs during the learning period from the second time point to the first time point, the impact associated with the accident is removed as noise and is not used in learning the acceleration threshold. As a result, in an example of the drive recorder device 16 of the third embodiment, the learning unit 163D learns the acceleration threshold so that the acceleration threshold becomes greater than the impact detected by the acceleration sensor 13 when the door changes from an open state to a closed state, and less than or equal to the impact detected by the acceleration sensor 13 when an accident occurs. In other words, in an example of the drive recorder device 16 of the third embodiment, the learned acceleration threshold becomes greater than the impact detected by the acceleration sensor 13 when the door changes from an open state to a closed state, and less than or equal to the impact detected by the acceleration sensor 13 when an accident occurs.
[0033] As described above, embodiments of the drive recorder device, video recording method, and program of this disclosure have been explained with reference to the drawings. However, the drive recorder device, video recording method, and program of this disclosure are not limited to the embodiments described above, and can be modified as appropriate without departing from the spirit of this disclosure. The configurations of each example of the embodiments described above may be combined as appropriate. In each example of the embodiments described above, the processing performed in the drive recorder device 16 was described as software processing performed by executing a program. However, the processing performed in the drive recorder device 16 may be hardware processing. Alternatively, the processing performed in the drive recorder device 16 may be a combination of both software and hardware processing. Furthermore, the program stored in the memory 162 of the drive recorder device 16 (the program that realizes the functions of the processor 163 of the drive recorder device 16) may be recorded on a computer-readable storage medium such as a semiconductor memory, magnetic recording medium, or optical recording medium and provided and distributed. [Explanation of symbols]
[0034] 1...Vehicle, 11...Camera, 12...Door opening / closing sensor, 13...Accelerometer, 14...Ignition power supply, 15...Constant power supply, 16...Drive recorder device, 161...Communication interface, 162...Memory, 163...Processor, 163A...Acquisition unit, 163B...Determination unit, 163C...Control unit, 163D...Learning unit
Claims
1. A control unit that starts recording video when an impact exceeding a predetermined acceleration threshold is detected by an acceleration sensor while the vehicle's ignition power is off, A determination unit that determines whether or not the door of the vehicle has changed from an open state to a closed state, A drive recorder device comprising: a learning unit that, when the door changes from an open state to a closed state, learns the acceleration threshold using the impact detected by the acceleration sensor during a learning period between a first time point in time when the door changes from an open state to a closed state and a second time point set back a predetermined time from the first time point.
2. The aforementioned learning unit, The drive recorder device according to claim 1, wherein the acceleration threshold is learned such that it becomes greater than the impact detected by the acceleration sensor when the door changes from an open state to a closed state, and less than or equal to the impact detected by the acceleration sensor when an accident occurs.
3. The aforementioned learning unit, The drive recorder device according to claim 1, wherein the impacts associated with accidents detected by the acceleration sensor during the learning period are not used for learning the acceleration threshold.
4. A control step in which the drive recorder device starts recording video when an impact exceeding a predetermined acceleration threshold is detected by the acceleration sensor while the vehicle's ignition power is turned off, The drive recorder device performs a determination step of determining whether the vehicle door has changed from an open state to a closed state, A video recording method comprising: a learning step in which, when the drive recorder device changes from an open state to a closed state, the drive recorder device learns the acceleration threshold using the impact detected by the acceleration sensor during a learning period between a first time point in time when the door changes from an open state to a closed state and a second time point set back a predetermined time from the first time point.
5. In the processor, A control step in which video recording is started when an impact exceeding a predetermined acceleration threshold is detected by an acceleration sensor while the vehicle's ignition power is turned off, A determination step to determine whether the door of the vehicle has changed from an open state to a closed state, A program for causing the following to be executed: when the door changes from an open state to a closed state, a learning step in which the acceleration threshold is learned using the impact detected by the acceleration sensor during a learning period between a first time point in time when the door changes from an open state to a closed state and a second time point set back a predetermined time from the first time point.