DETECTION FUNCTION CONTROL DEVICE, DETECTION FUNCTION CONTROL METHOD, AND PROGRAM
The detection function control device optimizes resource usage on small terminals by activating AI functions like facial recognition and distracted driving detection only when needed, addressing the computing power limitations and enabling effective DMS implementation.
Patent Information
- Application Number
- JP2021034404
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2021-03-04
- Publication Date
- 2024-11-13
- Estimated Expiration
- 2041-03-04
AI Technical Summary
Small terminals lack the computing power to execute multiple AI inference models simultaneously, making it difficult to implement functions like DMS (Driver Monitoring Systems) that require significant computational resources.
A detection function control device and method that determines vehicle conditions such as ignition, seating, acceleration, position, and speed to activate specific AI functions like facial recognition and distracted driving detection only when necessary, optimizing resource usage.
Enables more detection functions to be operated on small terminals with limited computing power by intelligently activating functions based on vehicle conditions, allowing for effective implementation of DMS without overwhelming the system.
Smart Images

Figure 0007581972000001 
Figure 0007581972000002 
Figure 0007581972000003
Abstract
Description
[Technical field]
[0001] The present invention relates to a detection function control device, a detection function control method, and a program. [Background technology]
[0002] There is known a technology that equips a small terminal with an AI (Artificial Intelligence) function to perform face recognition of a user, detection of drowsiness, etc. For example, a device that detects drowsiness without installing a camera is known (Patent Document 1). [Prior art documents] [Patent documents]
[0003] [Patent Document 1] JP 2010-250577 A Summary of the Invention [Problem to be solved by the invention]
[0004] AI functions need to execute different inference models depending on their purpose. Executing an inference model requires a large amount of computing resources (CPU resources or dedicated circuit resources). Therefore, there is a limit to the number of inference models that can be executed simultaneously on a small terminal. Some of the high-performance drive recorders available in recent years are equipped with a DMS (Driver Monitoring System). This DMS is equipped with functions such as facial recognition, detecting the driver's inattentiveness, drowsiness, and talking on the phone, and warns the driver of dangerous driving. However, small devices often do not have the computing power to execute all of these functions. For this reason, it has been difficult to implement functions such as a DMS on small devices. There is a demand for more detection functions to be able to be performed relative to the computing power available.
[0005] The present invention has been made in consideration of the above points, and provides a detection function control device, a detection function control method, and a program that can operate more detection functions relative to the available computing power. [Means for solving the problem]
[0006] The present invention has been made to solve the above problems, and one aspect of the present invention is a vehicle status determination unit that determines a status of the vehicle based on a result of detection of one or more of an ignition state, whether or not a seat is occupied, the acceleration of the vehicle, the position of the vehicle, the speed of the vehicle, and a picked-up voice, and a detection function control unit that determines which of one or more detection functions to activate for detecting dangerous driving based on the status determined by the vehicle status determination unit. The detection function control unit activates a face recognition function when the vehicle situation determination unit determines that all of the amounts of change in the acceleration, the position, and the speed are less than a predetermined amount and when the vehicle situation determination unit determines that a predetermined time has passed since the ignition was turned on, or when the vehicle situation determination unit determines that a change has occurred in the seating state, and the detection function control unit activates a function for determining distracted driving when the vehicle situation determination unit determines that the amount of change in any one or more of the acceleration, the position, and the speed is equal to or greater than a predetermined amount. It is a detection function control device.
[0008] In addition, one aspect of the present invention is the above-mentioned detection function control device, the detection function control unit is The vehicle status determination unit determines that the vehicle is moving, and When the vehicle state determination unit determines that the driver is speaking, an image recognition function is activated to determine a conversation action; The vehicle status determination unit determines that the vehicle is moving, and When the vehicle state determination unit determines that the driver is silent, it activates an image recognition function that determines whether the driver is drowsy at the wheel.
[0009] According to another aspect of the present invention, there is provided a vehicle status determination step of determining a status of the vehicle based on a result of detection of one or more of an ignition state, whether or not a person is seated in a seat, the acceleration of the vehicle, the position of the vehicle, the speed of the vehicle, and a picked-up voice, and a detection function control step of determining which of one or more detection functions for detecting dangerous driving is to be activated based on the status determined by the vehicle status determination step. The detection function control step activates a face recognition function when the vehicle status determination step determines that all of the amounts of change in the acceleration, the position, and the speed are less than a predetermined amount and when the vehicle status determination step determines that a predetermined time has elapsed since the ignition was turned on, or when the vehicle status determination step determines that a change has occurred in the seating state, and the detection function control step activates a function for determining distracted driving when the vehicle status determination step determines that the amount of change in any one or more of the acceleration, the position, and the speed is equal to or greater than a predetermined amount. A detection function control method.
[0010] Another aspect of the present invention is a program for causing a computer to execute a vehicle status determination step of determining a status of the vehicle based on a result of detection of one or more of an ignition state, whether or not a person is seated in a seat, the acceleration of the vehicle, the position of the vehicle, the speed of the vehicle, and a picked-up voice, and a detection function control step of determining which of one or more detection functions for detecting dangerous driving is to be activated based on the status determined by the vehicle status determination step. wherein the detection function control step activates a face recognition function when the vehicle status determination step determines that all of the amounts of change in the acceleration, the position, and the speed are less than a predetermined amount and when the vehicle status determination step determines that a predetermined time has elapsed since the ignition was turned on, or when the vehicle status determination step determines that a change has occurred in the seating state, and the detection function control step activates a function for determining distracted driving when the vehicle status determination step determines that the amount of change in any one or more of the acceleration, the position, and the speed is equal to or greater than a predetermined amount. It is. Effect of the Invention
[0011] The present invention allows more sensing functions to be run per available computing power. [Brief description of the drawings]
[0012] [Figure 1] 1 is a diagram illustrating an example of a hardware configuration of a drive recorder according to an embodiment of the present invention. [Diagram 2] 1 is a diagram illustrating an example of a functional configuration of a drive recorder according to an embodiment of the present invention. [Diagram 3] 1 is a diagram illustrating an example of a relationship between a vehicle situation and whether or not a plurality of detection functions are activated according to an embodiment of the present invention. [Figure 4] FIG. 11 is a diagram illustrating an example of a detection function control process according to the embodiment of the present invention. [Diagram 5] FIG. 11 is a diagram illustrating an example of a detection function control process according to the embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0013] (Embodiment) Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings. In this embodiment, as an example, a case will be described in which a drive recorder 1 is equipped with AI that detects multiple detection items such as face recognition, inattentiveness, drowsiness, and conversation behavior of the driver. The drive recorder 1 is mounted on a vehicle 3 (not shown) driven by a driver 2 (not shown). The drive recorder 1 is mainly installed on the dashboard or windshield of the vehicle 3 (not shown), for example. [Drive recorder configuration] 1 is a diagram showing an example of a hardware configuration of a drive recorder 1 according to the present embodiment. The drive recorder 1 includes an acceleration sensor 10, a gyro sensor 11, a GPS receiver 12, a vehicle signal cable 13, a camera 14, a speaker 15, a display 16, an operation button 17, a microphone 18, a timer 19, a CPU 110, a ROM 111, a RAM 112, and a communication module 113. The acceleration sensor 10, the gyro sensor 11, the GPS receiver 12, the vehicle signal cable 13, the camera 14, the speaker 15, the display 16, the operation button 17, the microphone 18, the timer 19, the CPU 110, the ROM 111, the RAM 112, and the communication module 113 are connected to each other by signal lines.
[0014] The acceleration sensor 10 measures the acceleration of the vehicle 3 . The gyro sensor 11 measures the angular velocity of the vehicle 3 . The GPS receiver 12 acquires information indicating the current position of the vehicle 3 by using a GPS (Global Positioning System).
[0015] The vehicle signal cable 13 acquires vehicle signals from the vehicle 3. The vehicle signals are signals that indicate the state of the vehicle 3. The vehicle signal cable 13 acquires various vehicle signals from an ECU (Electronic Control Unit) of the vehicle 3. The various vehicle signals include, for example, an ignition switch signal, a vehicle speed signal, a signal that indicates the steering angle of the vehicle 3, a signal that indicates the value of pressure applied to a seat of the vehicle 3, and the like.
[0016] The camera 14 includes a first camera and a second camera. The first camera captures an image of a part or the whole of the body of the driver 2. The second camera captures and records moving images while the vehicle 3 is traveling. The speaker 15 outputs an audio warning sound.
[0017] The display 16 displays various screens and is, for example, a liquid crystal display or an organic electroluminescence (EL) display. The operation button 17 is a button for the user to operate the drive recorder 1. The display 16 may be a touch panel, in which case the operation buttons 17 may be configured integrally with the display 16. Alternatively, a part of the operation buttons 17 may be configured integrally with the display 16.
[0018] The microphone 18 picks up voice. The voice includes the voice of the driver 2. The timer 19 measures time.
[0019] The CPU 110 reads a program from the ROM 111 and executes various controls in accordance with the read program. The CPU 110 has multiple built-in internal storage media such as registers. The CPU 110 temporarily stores data from the ROM 111 in the internal storage media and processes the data. The CPU 110 outputs the results of the calculations to the register, and further outputs the results from the register to the RAM 112 or an external storage medium.
[0020] The ROM 111 is a main storage device that stores various programs, data, parameters, etc. for the CPU 110 to perform various calculations and controls. The ROM 111 can retain the stored contents even when the power supplied to the ROM 111 is zero.
[0021] The RAM 112 is a main storage device used as a working memory by the CPU 110. Programs, data, and the like are written to and erased from the RAM 112 by the CPU 110. The RAM 112 is configured using a storage device such as a semiconductor storage device, for example.
[0022] The communication module 113 transmits and receives various types of information via the wireless network NW. The communication module 113 includes a communication interface (I / F).
[0023] 2 is a diagram showing an example of a functional configuration of the drive recorder 1 according to this embodiment. The drive recorder 1 includes a control unit 20, an acceleration acquisition unit 21, an angular velocity acquisition unit 22, a position acquisition unit 23, a vehicle signal acquisition unit 24, a sound collection unit 25, an image capture unit 26, a timer unit 27, an operation unit 28, an audio output unit 29, a display unit 210, a communication unit 211, and a storage unit 212.
[0024] The control unit 20 performs various controls on the drive recorder 1. The control unit 20 includes, for example, a CPU, and performs various calculations and information transmission and reception. Each functional unit of the control unit 20 is realized by the CPU reading a program from a ROM and executing processing. Each functional unit of the control unit 20 will be described later. The control unit 20 includes a CPU 110, a ROM 111, and a RAM 112 shown in FIG. 1.
[0025] The acceleration acquisition unit 21 acquires the acceleration of the vehicle 3. The acceleration acquisition unit 21 generates acceleration data A1 based on the acquired acceleration. The acceleration data A1 indicates the acceleration of the vehicle 3. The acceleration acquisition unit 21 includes the acceleration sensor 10 shown in FIG.
[0026] The angular velocity acquisition unit 22 acquires the angular velocity of the vehicle 3. The angular velocity acquisition unit 22 generates angular velocity data B1 based on the acquired angular velocity. The angular velocity data B1 indicates the angular velocity of the vehicle 3. The angular velocity acquisition unit 22 includes the gyro sensor 11 shown in FIG.
[0027] The position acquisition unit 23 acquires the position of the vehicle 3. The position acquisition unit 23 generates position data C1 based on the acquired position. The position data C1 indicates the position of the vehicle 3. The position acquisition unit 23 includes the GPS receiver 12 shown in FIG.
[0028] The vehicle signal acquisition unit 24 acquires a vehicle signal of the vehicle 3. The vehicle signal acquisition unit 24 generates vehicle signal data D1 based on the acquired vehicle signal. The vehicle signal data D1 includes data indicating whether the ignition is on or off, data indicating the vehicle speed of the vehicle 3, steering angle data, and seating data. The steering angle data indicates the steering angle of the vehicle 3. The seating data indicates the value of pressure applied to the seat of the vehicle 3. The vehicle signal acquisition unit 24 includes the vehicle signal cable 13 shown in FIG. 1.
[0029] The sound collection unit 25 collects the voice uttered by the driver 2. As a result of the collected voice, the sound collection unit 25 generates voice data E1. The voice data E1 indicates the voice uttered by the driver 2. The sound collection unit 25 includes the microphone 18 shown in FIG.
[0030] The photographing unit 26 photographs one or more of the face of the driver 2 and the behavior of the driver 2. The photographing unit 26 generates image data F1 as a result of photographing. The image data F1 indicates one or more of the face of the driver 2 and the behavior of the driver 2. The photographing unit 26 includes the camera 14 shown in FIG. 1.
[0031] The timing unit 27 performs various types of timing. The timing unit 27 starts and stops timing in response to a signal from the control unit 20. The timing unit 27 generates timing data G1 as a result of timing. The timing data G1 includes, for example, the time elapsed from when a timing start signal is obtained from the control unit 20 to when a timing stop signal is obtained. The timing unit 27 may perform multiple timings in parallel. When multiple timings are performed in parallel, the timing data G1 includes the results of the multiple timings. The timing data G1 may also include the current time. The timing unit 27 includes the timer 19 shown in FIG.
[0032] The operation unit 28 accepts various operations by a user of the drive recorder 1. The operation unit 28 includes the operation buttons 17 shown in Fig. 1. Note that, in the case where the operation buttons 17 and the display 16 are integrated as a touch panel, the operation unit 28 includes the touch panel.
[0033] The audio output unit 29 outputs various sounds. The sounds output by the audio output unit 29 include, for example, a sound to be output as a warning sound when an abnormality is detected as a result of detection regarding the driver 2.
[0034] The display unit 210 displays various screens. The screens displayed by the display unit 210 include, for example, a warning screen that is displayed as a warning when an abnormality is detected as a result of detection of the driver 2. The display unit 210 includes the display 16 shown in Fig. 1 and a RAM 112. The RAM 112 includes image data of the warning screen.
[0035] The communication unit 211 communicates with external devices. The external devices with which the communication unit 211 communicates include a server (not shown). The communication unit 211 includes the communication module 113 shown in FIG. The storage unit 212 stores various types of information. The storage unit 212 includes the RAM 112 shown in FIG.
[0036] Here, a description will be given of each functional unit included in the control unit 20. The control unit 20 includes an acquisition unit 200, an operation reception unit 2010, a display control unit 2011, a warning unit 2012, a detection execution unit 30, and a detection function management unit 31.
[0037] The acquisition unit 200 acquires various types of information. The acquisition unit 200 includes an acceleration data acquisition unit 201, an angular velocity data acquisition unit 202, a steering angle data acquisition unit 203, a position information acquisition unit 204, a vehicle signal data acquisition unit 205, a seating data acquisition unit 206, a voice data acquisition unit 207, a captured image acquisition unit 208, and a timing data acquisition unit 209.
[0038] The acceleration data acquisition unit 201 acquires the acceleration data A1 from the acceleration acquisition unit 21 . The angular velocity data acquisition unit 202 acquires the angular velocity data B 1 from the angular velocity acquisition unit 22 . The steering angle data acquisition unit 203 acquires the steering angle data included in the vehicle signal data D1 from the vehicle signal acquisition unit 24. The position information acquisition unit 204 acquires the position data C1 from the position acquisition unit 23. The vehicle signal data acquisition unit 205 acquires the vehicle signal data D1 from the vehicle signal acquisition unit 24.
[0039] The seating data acquisition unit 206 acquires the seating pressure data included in the vehicle signal data D1 from the vehicle signal acquisition unit 24. The audio data acquisition unit 207 acquires the audio data E 1 from the sound collection unit 25 . The photographed image acquisition unit 208 acquires image data F 1 from the photographing unit 26 . The timing data acquisition unit 209 acquires timing data G 1 from the timing unit 27 .
[0040] The operation reception unit 2010 receives various operations performed on the operation unit 28 . The display control unit 2011 controls the display unit 210 to cause the display unit 210 to display various screens. The warning unit 2012 executes a warning. For example, the warning unit 2012 controls the audio output unit 29 to output an alarm sound.
[0041] The detection execution unit 30 has an AI function and detects a plurality of detection items such as a conversation. In other words, the detection execution unit 30 executes a plurality of detection functions according to the plurality of detection items. Note that a learned model for the detection execution unit 30 to perform various judgments is stored in the storage unit 212. The detection execution unit 30 corresponds to a driver monitoring system (DMS). The detection execution unit 30 includes a face authentication unit 300 , a distracted driving determination unit 301 , a drowsy driving determination unit 302 , and a call operation determination unit 303 .
[0042] The face authentication unit 300 authenticates the face of the driver 2. The face authentication unit 300 performs authentication based on an AI function by comparing the face of the driver 2 indicated by the image data F1 acquired by the captured image acquisition unit 208 with a face image (not shown) of the owner of the vehicle 3 stored in the storage unit 212. The face authentication unit 300 collects the result of the face authentication as information on the driver. The face authentication is performed, for example, when the vehicle 3 is a commercial vehicle, to prevent unauthorized or unrelated persons from driving the vehicle. For example, when the face authentication unit 300 determines that the driver 2 is an unauthorized person as a result of the face authentication, it causes the warning unit 2012 to execute a warning. Alternatively, the face authentication may be used to verify whether the driver 2 is the owner of the vehicle 3 for the purpose of preventing theft.
[0043] The inattentive driving determination unit 301 determines whether the driver 2 is looking away from the road based on the behavior of the driver 2 (e.g., the direction of the face, the direction of the line of sight, etc.) indicated by the image data F1 acquired by the captured image acquisition unit 208. When the inattentive driving determination unit 301 determines that the driver 2 is looking away from the road, it causes the warning unit 2012 to issue a warning.
[0044] The doze-driving determination unit 302 determines whether the driver 2 is dozing while driving. The doze-driving determination unit 302 determines whether the driver 2 is dozing while driving based on the behavior of the driver 2 (e.g., head shaking, eyelids open, etc.) indicated by the image data F1 acquired by the captured image acquisition unit 208. When the doze-driving determination unit 302 determines that the driver 2 is dozing while driving, it causes the warning unit 2012 to issue a warning.
[0045] The calling action determination unit 303 determines the calling action. The calling action determination unit 303 determines the calling action based on the voice of the driver 2 indicated by the voice data E1 acquired by the voice data acquisition unit 207. The calling action determination unit 303 may determine the calling action based on the action of the driver 2 indicated by the image data F1 acquired by the captured image acquisition unit 208 (e.g., holding a mobile phone to the ear). When the calling action determination unit 303 determines that the driver 2 is talking on the phone, it causes the warning unit 2012 to execute a warning.
[0046] The detection function management unit 31 determines which of the multiple detection functions executed by the detection execution unit 30 is to be activated. In other words, the detection function management unit 31 selects a detection function to be activated from the multiple detection functions. As a result of the determination (or selection), the detection function management unit 31 may activate one detection function or multiple detection functions. In the following description, the process executed by the detection function management unit 31 is also referred to as a detection function control process. The detection function management unit 31 includes a vehicle state determination unit 310 and a detection function control unit 311 .
[0047] The vehicle status determination unit 310 determines the vehicle status based on the detection results of predetermined detection items. Here, the predetermined detection items are, for example, one or more of the ignition state, whether or not a person is seated in a seat, the acceleration of the vehicle, the position of the vehicle 3, the speed of the vehicle 3, and the picked-up voice. The vehicle status is the status of the vehicle 3. The vehicle status includes, for example, the movement status of the vehicle 3, the status when driving starts or when the driver is changed, the status when the driver 2 is speaking, and the like. The movement status of the vehicle 3 includes, for example, the vehicle 3 moving, the vehicle 3 moving straight ahead, and the like.
[0048] The predetermined detection items may be one or more of the ignition state, whether or not a seat is occupied, the acceleration of the vehicle, the position of the vehicle 3, the speed of the vehicle 3, and the picked-up voice.
[0049] The detection function control unit 311 determines which of one or more detection functions to activate for detecting dangerous driving, based on the vehicle situation determined by the vehicle situation determination unit 310. The detection function control unit 311 outputs the determination result to the detection execution unit 30, and causes the detection execution unit 30 to activate the detection function indicated by the determination result among the multiple detection functions.
[0050] Here, the relationship between the vehicle situation and whether or not to activate a plurality of detection functions will be described with reference to Fig. 3. Fig. 3 is a diagram showing an example of the relationship between the vehicle situation and whether or not to activate a plurality of detection functions according to the present embodiment. 3 shows vehicle situations in which the vehicle 3 is stopped, the vehicle 3 is moving, the vehicle 3 is turning, the driver 2 is silent, and the driver 2 is speaking. In FIG. 3, whether or not to activate the detection mechanism in each of these vehicle situations is shown for each of the detection items of face recognition, looking away, dozing, and talking.
[0051] It is necessary to collect driver information by face recognition when driving starts and when the driver is changed. Face recognition is executed only when it is detected that the vehicle 3 is not moving (stopped) from the acceleration sensor, GPS data, vehicle speed pulse data, etc., and face recognition is stopped for the remaining time when these are not detected.
[0052] A dangerous driving warning due to distracted driving detection becomes necessary when the vehicle 3 is moving straight ahead. The distracted driving detection function is executed only when it is detected from the acceleration sensor, GPS data, vehicle speed pulse data, etc. that the vehicle 3 is moving, and when it is detected from the gyro sensor, steering angle data, etc. that the vehicle 3 is not turning (moving straight), and the distracted driving detection function is stopped for the remaining time when these are not detected. Note that the determination of whether to execute the distracted driving detection function does not require detection of the driver's silence or the driver's speech.
[0053] Dangerous driving warnings based on drowsiness detection become necessary when the vehicle 3 is moving and the driver 2 is silent. The drowsiness detection function is executed only when it is detected from the acceleration sensor, GPS data, vehicle speed pulse data, etc. that the vehicle 3 is moving, from the gyro sensor, steering angle data, etc. that the vehicle 3 is turning, and from the voice data collected from the microphone that the driver is silent, and the drowsiness detection function is stopped for the remaining time when these are not detected.
[0054] Dangerous driving warnings based on conversation detection become necessary when the vehicle 3 is moving and the driver 2 is speaking. The conversation detection function is executed only when it is detected from the acceleration sensor, GPS data, vehicle speed pulse data, etc. that the vehicle 3 is moving, from the gyro sensor, steering angle data, etc. that the vehicle 3 is turning, and from the voice data collected from the microphone that the driver is speaking, and the conversation detection function is stopped for the remaining time when these are not detected.
[0055] [Detection function control processing] Next, a specific example of the detection function control process will be described with reference to Fig. 4 and Fig. 5. For example, while the power supply of the drive recorder 1 is on, the control unit 20 repeatedly executes the detection function control processes shown in Fig. 4 and Fig. 5 in parallel. Note that only one of the detection function control processes shown in Fig. 4 and Fig. 5 may be executed.
[0056] Before the detection function control process shown in Figures 4 and 5 is started, it is determined that the multiple detection functions will not be activated. In other words, the control unit 20 does not activate any of the multiple detection functions that are not determined to be activated in the detection function control process shown in Figures 4 and 5. However, since the detection function control process shown in Figures 4 and 5 is repeatedly executed, even if a detection function is determined not to be activated at a certain point in time, it may be determined to be activated in the next detection function control process. Similarly, even if a detection function is determined to be activated at a certain point in time, it may be determined not to be activated in the next detection function control process.
[0057] FIG. 4 is a diagram showing an example of the detection function control process according to the present embodiment. Step S10: The vehicle situation determination unit 310 acquires various detection results. Here, the vehicle situation determination unit 310 acquires acceleration data A1 from the acceleration data acquisition unit 201. The vehicle situation determination unit 310 acquires position data C1 from the position information acquisition unit 204. The vehicle situation determination unit 310 acquires vehicle signal data D1 from the vehicle signal data acquisition unit 205. The vehicle situation determination unit 310 acquires timing data G1 from the timing data acquisition unit 209. The vehicle situation determination unit 310 acquires seating pressure data from the seating data acquisition unit 206.
[0058] Step S20: The vehicle status determination unit 310 determines whether or not all of the changes in the acceleration of the vehicle 3, the position of the vehicle 3, and the speed of the vehicle 3 are less than a predetermined amount. For example, the vehicle status determination unit 310 calculates the amount of change in acceleration by dividing the acceleration indicated by the acceleration data A1 by a predetermined time indicated by the timing data G1. For example, the vehicle status determination unit 310 calculates the amount of change in position by dividing the distance between multiple positions indicated by the position data C1 by a predetermined time indicated by the timing data G1. For example, the vehicle status determination unit 310 calculates the amount of change in vehicle speed based on the vehicle speed signal indicated by the vehicle signal data D1 and the predetermined time indicated by the timing data G1.
[0059] The predetermined amount of change in the acceleration of the vehicle 3 is, for example, an amount of change that exceeds continuous vibration caused by an engine, etc. The predetermined amount of change in the acceleration may be determined in advance, or may be calculated as needed from a representative value such as the average or median of the amount of change in acceleration within a predetermined period of time in the past. The predetermined amount of change in the position of the vehicle 3 is, for example, a change in the GPS position after a unit time of 1 second of a predetermined distance of 1 meter. The unit time is not limited to 1 second, and may be an appropriate time of 10 milliseconds or more and 10 seconds or less. The predetermined distance is also not limited to 1 meter, and may be an appropriate distance of 0 meters or more and 10 meters or less. The predetermined change in speed of the vehicle 3 is, for example, 3 kilometers per hour. The speed is not limited to 3 kilometers per hour, and may be an appropriate speed between 0 and 5 kilometers per hour.
[0060] When the vehicle status determination unit 310 determines that all of the changes in the acceleration of the vehicle 3, the position of the vehicle 3, and the speed of the vehicle 3 are less than a predetermined amount (step S20; YES), the detection function management unit 31 executes the process of step S30. On the other hand, when the vehicle status determination unit 310 determines that the changes in one or more of the acceleration of the vehicle 3, the position of the vehicle 3, and the speed of the vehicle 3 are equal to or greater than a predetermined amount (step S20; NO), the detection function management unit 31 executes the process of step S50.
[0061] Step S30: The vehicle status determination unit 310 determines whether a predetermined time has elapsed since the ignition was turned on, or whether there has been a change in the seat occupancy sensor. The presence of a change in the seat occupancy sensor indicates a change in the position or number of passengers seated in the seats.
[0062] In order to determine whether it is within the predetermined time since the ignition was turned on, the vehicle status determination unit 310 makes the timer unit 27 start timing when the vehicle signal data D1 indicates that the ignition was turned on. The vehicle status determination unit 310 makes a determination by comparing the time that has passed since the ignition was turned on with the predetermined time based on the timing data G1. In order to determine whether or not it is within a predetermined time since the ignition was turned on, the determination may be made based on a vehicle signal other than the vehicle signal data D1.
[0063] In order to determine whether or not there has been a change in the seating sensor of the seat, the vehicle state determination unit 310 makes the determination based on the value of the pressure applied to the seat of the vehicle 3, which is indicated by the pressure data.
[0064] If the vehicle status determination unit 310 determines that it is within a predetermined time since the ignition was turned on or that there has been a change in the seat occupancy sensor (step S30; YES), the detection function management unit 31 executes the process of step S40. On the other hand, if the vehicle status determination unit 310 determines that it is not within a predetermined time since the ignition was turned on and that there has been no change in the seat occupancy sensor (step S30; NO), the detection function management unit 31 ends the detection function control process.
[0065] Step S40: The detection function control unit 311 determines to activate the face authentication function. The detection function control unit 311 causes the face authentication unit 300 to execute the face authentication function.
[0066] Step S50: The detection function control unit 311 determines to activate the function for determining inattentive driving. The detection function control unit 311 causes the inattentive driving determination unit 301 to execute the function for determining inattentive driving. That is, when the vehicle state determination unit 310 determines that the amount of change in one or more of the acceleration of the vehicle 3, the position of the vehicle 3, and the speed of the vehicle 3 is equal to or greater than a predetermined amount, the detection function control unit 311 activates the function for determining inattentive driving. With this, the detection function management unit 31 ends the detection function control process.
[0067] Facial recognition is required when the driver 2 is replaced. The detection function control process shown in Fig. 4 is based on the assumption that the driver 2 is replaced when the vehicle 3 is not moving. In the detection function control process shown in Fig. 4, a condition that a predetermined time has passed since the ignition was turned on is used as an example of a condition that the vehicle 3 is not moving.
[0068] Note that the determination of inattentive driving is required when the vehicle 3 is traveling. In the detection function control process shown in Fig. 4, as an example of the condition that the vehicle 3 is traveling, a condition that the amount of change in the acceleration, the position, and the speed of the vehicle 3 are each equal to or greater than a predetermined amount is used.
[0069] As described above, in the detection function control process shown in Fig. 4, an example of a case where it is determined whether or not to activate the face recognition function and whether or not to activate the function for determining inattentive driving is described, but this is not limited to this. In the detection function control process shown in Fig. 4, the determination of either the face recognition function or the function for determining inattentive driving may be omitted.
[0070] When the determination of whether to activate the face authentication function is omitted, the processes of steps S30 and S40 are omitted in the detection function control process shown in Fig. 4. In that case, when the vehicle status determination unit 310 determines in step S20 that all of the changes in the acceleration of the vehicle 3, the position of the vehicle 3, and the speed of the vehicle 3 are less than a predetermined amount (step S20; YES), the detection function management unit 31 ends the detection function control process. On the other hand, when the vehicle status determination unit 310 determines that the changes in one or more of the acceleration of the vehicle 3, the position of the vehicle 3, and the speed of the vehicle 3 are equal to or greater than a predetermined amount (step S20; NO), the detection function management unit 31 executes the process of step S50.
[0071] When the determination of whether or not to activate the function for determining inattentive driving is omitted, the processes of steps S20 and S50 are omitted in the detection function control process shown in Fig. 4. In that case, the process of step S30 is executed after the process of step S10. In step S30, if the vehicle status determination unit 310 determines that it is within a predetermined time since the ignition was turned on or that there has been a change in the seat occupancy sensor (step S30; YES), the detection function management unit 31 executes the process of step S40. On the other hand, if the vehicle status determination unit 310 determines that it is not within a predetermined time since the ignition was turned on and that there has been no change in the seat occupancy sensor (step S30; NO), the detection function management unit 31 ends the detection function control process.
[0072] Note that a change in the seating sensor of the seat is an example of a change in the seating state. Another example of a change in the seating state is the door of the vehicle 3 being opened or closed. The vehicle status determination unit 310 may determine whether the door of the vehicle 3 has been opened or closed in order to determine whether a change in the seating state has occurred. In this case, the vehicle signal data D1 includes a signal indicating whether the door has been opened or closed. The vehicle status determination unit 310 determines whether the door of the vehicle 3 has been opened or closed based on the signal. In addition, in order to determine whether a change has occurred in the seating state, the vehicle status determination unit 310 may determine both whether there has been a change in the seating sensor of the seat and whether the door of the vehicle 3 has been opened or closed.
[0073] As described above, in step S30, the vehicle status determination unit 310 may determine whether or not the predetermined time has elapsed since the ignition was turned on, or whether or not a change has occurred in the seating state. Therefore, the detection function control unit 311 activates the face recognition function when the vehicle status determination unit 310 determines that all of the amounts of change in the acceleration of the vehicle 3, the position of the vehicle 3, and the speed of the vehicle 3 are less than a predetermined amount, and when the vehicle status determination unit 310 determines that the predetermined time has elapsed since the ignition was turned on, or when the vehicle status determination unit 310 determines that a change has occurred in the seating state.
[0074] FIG. 5 is a diagram showing an example of the detection function control process according to the present embodiment. Step S110: The vehicle state determination unit 310 acquires various detection results. Here, the vehicle state determination unit 310 acquires the voice data E1 from the voice data acquisition unit 207.
[0075] Step S120: The vehicle state determination unit 310 determines whether the driver 2 is speaking or not. The vehicle state determination unit 310 determines whether the driver 2 is speaking or silent, based on the voice of the driver 2 indicated by the voice data E1.
[0076] When the vehicle state determination unit 310 determines that the driver 2 is speaking (step S120; YES), the detection function management unit 31 executes the process of step S130. On the other hand, when the vehicle state determination unit 310 determines that the driver 2 is not speaking (is silent) (step S120; NO), the detection function management unit 31 executes the process of step S140.
[0077] Step S130: The detection function control unit 311 determines to activate the image recognition function for determining a call operation. The detection function control unit 311 causes the call operation determination unit 303 to execute the image recognition function for determining a call operation. Therefore, when the vehicle state determination unit 310 determines that the driver 2 is speaking, the detection function control unit 311 activates the image recognition function for determining a call operation.
[0078] Step S140: The detection function control unit 311 determines to activate the image recognition function for determining drowsy driving. The detection function control unit 311 causes the drowsy driving determination unit 302 to execute the image recognition function for determining drowsy driving. Therefore, when the vehicle state determination unit 310 determines that the driver 2 is silent, the detection function control unit 311 activates the image recognition function for determining drowsy driving. With this, the detection function management unit 31 ends the detection function control process.
[0079] In the detection function control process shown in Fig. 5, an example of a case where it is determined whether or not to activate the image recognition function for determining whether the driver 2 is dozing off while driving is described, depending on whether or not the driver 2 is speaking, but the present invention is not limited to this. For example, it may be determined whether or not to activate the image recognition function for determining whether or not to activate the image recognition function for determining whether or not the driver 2 is dozing off while driving may be determined based on the speech of the driver 2 and the steering angle. In this case, when the driver 2 is silent and the steering angle is equal to or greater than a predetermined value, it is determined that the driver is dozing off while driving, and the detection function control unit 311 activates the function.
[0080] As described above, the detection function control device according to this embodiment (the drive recorder 1 in this embodiment) includes the vehicle state determination unit 310 and the detection function control unit 311. The vehicle status determination unit 310 determines the status of the vehicle 3 (in this embodiment, the vehicle status) based on the results of detection of one or more of the ignition state, whether or not a seat is occupied, the acceleration of the vehicle 3, the position of the vehicle 3, the speed of the vehicle 3, and the picked-up sound. The detection function control unit 311 determines which of one or more detection functions (in this embodiment, the face authentication unit 300, the inattentive driving determination unit 301, the drowsy driving determination unit 302, and the call operation determination unit 303) to activate for detecting dangerous driving based on the situation (in this embodiment, the vehicle situation) determined by the vehicle situation determination unit 310.
[0081] With this configuration, the detection function control device according to this embodiment (the drive recorder 1 in this embodiment) can activate only one or more detection functions required for detecting dangerous driving according to the situation of the vehicle 3, so that more detection functions can be activated relative to the computational power available. As described above, the detection functions are based on AI. With the detection function control device according to this embodiment, more dangerous driving warning functions can be implemented and executed even in a DMS provided in a small terminal with low computational power.
[0082] Furthermore, in the detection function control device according to this embodiment (drive recorder 1 in this embodiment), when the vehicle situation determination unit 310 determines that all of the changes in the acceleration of vehicle 3, the position of vehicle 3, and the speed of vehicle 3 are less than a predetermined amount and when the vehicle situation determination unit 310 determines that it is within a predetermined time since the ignition was turned on, or when the detection function control unit 311 determines that a change has occurred in the seating state, the detection function control unit 311 activates a face recognition function (face recognition unit 300 in this embodiment), and when the vehicle situation determination unit 310 determines that the changes in one or more of the acceleration of vehicle 3, the position of vehicle 3, and the speed of vehicle 3 are greater than or equal to a predetermined amount, the detection function control unit 311 activates a function for determining distracted driving (distracted driving determination unit 301 in this embodiment).
[0083] With this configuration, the detection function control device according to this embodiment (the drive recorder 1 in this embodiment) can determine whether or not a change in the driver 2 has occurred based on the fact that the vehicle 3 is not moving, and can therefore activate the face recognition function when a change in the driver 2 has occurred. Also, the detection function control device according to this embodiment can determine whether or not the vehicle 3 is traveling, and can therefore activate the function of determining distracted driving when the vehicle 3 is traveling.
[0084] The conditions for the detection function control device to activate the face authentication function or the function to determine inattentive driving are not limited to the above conditions. For example, the detection function control unit 311 may activate the face authentication unit 300 when the vehicle status determination unit 310 determines that the amount of change in one or more of the acceleration of the vehicle 3, the position of the vehicle 3, and the speed of the vehicle 3 is less than a predetermined amount, and when the vehicle status determination unit 310 determines that the amount of change is within a predetermined time since the ignition was turned on. The detection function control unit 311 may also activate the inattentive driving determination unit 301 when the vehicle status determination unit 310 determines that the amount of change in all of the acceleration of the vehicle 3, the position of the vehicle 3, and the speed of the vehicle 3 is greater than or equal to a predetermined amount. Furthermore, the detection function control unit 311 may activate the face authentication unit 300 when the vehicle state determination unit 310 determines that the amount of change in one or more of the acceleration of the vehicle 3, the position of the vehicle 3, and the speed of the vehicle 3 is less than a predetermined amount. The detection function control unit 311 may activate the face authentication unit 300 when the vehicle state determination unit 310 determines that a predetermined time has passed since the ignition was turned on.
[0085] Furthermore, in the detection function control device of this embodiment (in this embodiment, the drive recorder 1), when the vehicle situation determination unit 310 determines that the driver 2 is speaking, the detection function control unit 311 activates an image recognition function that determines a call operation (in this embodiment, the call operation determination unit 303) and when the vehicle situation determination unit 310 determines that the driver 2 is silent, the detection function control unit 311 activates an image recognition function that determines drowsy driving (in this embodiment, the drowsy driving determination unit 302).
[0086] With this configuration, the detection function control device of this embodiment (in this embodiment, the drive recorder 1) can determine whether the driver 2 is speaking or not, and can therefore activate an image recognition function that determines whether the driver 2 is drowsy while driving when the driver 2 is silent.
[0087] Any one or more of the face authentication unit 300, the inattentive driving determination unit 301, the drowsy driving determination unit 302, and the call operation determination unit 303 may be omitted from the configuration of the detection execution unit 30. Furthermore, in addition to the face authentication unit 300, the inattentive driving determination unit 301, the drowsy driving determination unit 302, and the call operation determination unit 303, or instead of any one of them, the detection execution unit 30 may be provided with a functional unit having another detection function.
[0088] The vehicle status determination unit 310 may distinguish between day and night as the vehicle status. It is considered that detection of drowsy driving is unnecessary during the day. In that case, the drive recorder 1 may be equipped with an illuminance sensor. When the vehicle status determination unit 310 determines that the brightness measured by the illuminance sensor is equal to or higher than a predetermined value, the detection function control unit 311 does not cause the drowsy driving determination unit 302 to operate the image recognition function for determining drowsy driving.
[0089] In addition to the functions described with reference to FIG. 2, the drive recorder 1 has a recording function for storing in the storage unit 212 the video images constantly recorded by the camera 14. The drive recorder 1 may also perform an impact detection process. The impact detection process includes a process of causing the audio output unit 29 to output an alarm sound and a process of causing the display unit 210 to display an alarm screen when an impact of a predetermined magnitude or greater is detected. The impact detection process also includes a process of generating event information and storing the event information in the storage unit 212.
[0090] The control unit 20 may be provided as a function of a device mounted on the vehicle 3 other than a drive recorder. When the device is a small terminal, the configuration of the control unit 20 is preferably used. Here, a small terminal refers to a terminal that does not have a high computing power and therefore has difficulty in simultaneously operating multiple AI functions.
[0091] In addition, a part of the drive recorder 1 in the above-mentioned embodiment, for example, the control unit 20, may be realized by a computer. In that case, a program for realizing this control function may be recorded in a computer-readable recording medium, and the program recorded in the recording medium may be read into a computer system and executed to realize the control function. In addition, the "computer system" here refers to a computer system built into the drive recorder 1, and includes hardware such as an OS and peripheral devices. In addition, the "computer-readable recording medium" refers to a portable medium such as a flexible disk, an optical magnetic disk, a ROM, a CD-ROM, and a storage device such as a hard disk built into a computer system. Furthermore, the "computer-readable recording medium" may include a medium that dynamically holds a program for a short time, such as a communication line when a program is transmitted via a network such as the Internet or a communication line such as a telephone line, and a medium that holds a program for a certain period of time, such as a volatile memory inside a computer system that is a server or client in that case. In addition, the above-mentioned program may be a program for realizing a part of the above-mentioned function, and may further be a program that can realize the above-mentioned function in combination with a program already recorded in the computer system. In addition, a part or all of the drive recorder 1 in the above-mentioned embodiment may be realized as an integrated circuit such as an LSI (Large Scale Integration). Each functional block of the drive recorder 1 may be individually processed, or may be integrated into a processor in part or in whole. The integrated circuit method is not limited to LSI, and may be realized by a dedicated circuit or a general-purpose processor. In addition, when an integrated circuit technology that replaces LSI appears due to the progress of semiconductor technology, an integrated circuit based on that technology may be used.
[0092] Although one embodiment of the present invention has been described in detail above with reference to the drawings, the specific configuration is not limited to the above, and various design changes, etc. are possible within the scope that does not deviate from the gist of the present invention. [Explanation of symbols]
[0093] 1... drive recorder, 2... driver, 3... vehicle, 310... vehicle condition determination unit, 311... detection function control unit, 300... face recognition unit, 301... inattentive driving determination unit, 302... drowsy driving determination unit, 303... call operation determination unit
Claims
1. a vehicle status determination unit that determines a status of the vehicle based on a result of detection of one or more of an ignition state, whether or not a seat is occupied, an acceleration of the vehicle, a position of the vehicle, a speed of the vehicle, and a picked-up sound; a detection function control unit that determines which of one or more detection functions to activate for detecting dangerous driving based on the situation determined by the vehicle situation determination unit; and Equipped with the detection function control unit activates a face authentication function when the vehicle situation determination unit determines that all of the amounts of change in the acceleration, the position, and the speed are less than a predetermined amount and when the vehicle situation determination unit determines that a predetermined time has passed since the ignition was turned on, or when the vehicle situation determination unit determines that a change has occurred in the seating state; The detection function control unit is a detection function control device that activates a function for determining distracted driving when the vehicle situation determination unit determines that the amount of change in one or more of the acceleration, the position, and the speed is greater than a predetermined value.
2. The detection function control unit activates an image recognition function for determining a conversation operation when the vehicle status determination unit determines that the vehicle is moving and that the driver is speaking, and activates an image recognition function for determining drowsy driving when the vehicle status determination unit determines that the vehicle is moving and that the driver is silent. The sensing function control device according to claim 1 .
3. a vehicle status determination step of determining a status of the vehicle based on a result of detection of one or more of an ignition state, whether or not a seat is occupied, an acceleration of the vehicle, a position of the vehicle, a speed of the vehicle, and a picked-up voice; a detection function control step of determining which of one or more detection functions for detecting dangerous driving is to be activated based on the situation determined by the vehicle situation determination step, the detection function control step activates a face authentication function when the vehicle status determination step determines that all of the amounts of change in the acceleration, the position, and the speed are less than a predetermined amount and when the vehicle status determination step determines that a predetermined time has elapsed since the ignition was turned on, or when the vehicle status determination step determines that a change has occurred in the seating state; The detection function control step is a detection function control method in which, when the vehicle situation determination step determines that the amount of change in one or more of the acceleration, the position, and the speed is greater than a predetermined amount, a function for determining distracted driving is activated.
4. On the computer, a vehicle status determination step of determining a status of the vehicle based on a result of detection of one or more of an ignition state, whether or not a seat is occupied, an acceleration of the vehicle, a position of the vehicle, a speed of the vehicle, and a picked-up voice; a detection function control step of determining which of one or more detection functions for detecting dangerous driving is to be activated based on the situation determined by the vehicle situation determination step; A program for executing the detection function control step activates a face authentication function when the vehicle status determination step determines that all of the amounts of change in the acceleration, the position, and the speed are less than a predetermined amount and when the vehicle status determination step determines that a predetermined time has elapsed since the ignition was turned on, or when the vehicle status determination step determines that a change has occurred in the seating state; The detection function control step activates a function for determining distracted driving when the vehicle state determination step determines that the amount of change in at least one of the acceleration, the position, and the speed is equal to or greater than a predetermined amount. program.
Citation Information
Patent Citations
Portable telephone set
JP1998234076A
Room monitor device, room monitor control device and vehicle robber specifying system
JP2003226227A
Information processing apparatus, program, portable radio communication apparatus, doze prevention system, and doze detection method
JP2010250577A
Methods for the safe operation of mobile phones inside vehicles
JP2010503262A
Driver state detection device and driver state notification device
JP2014016702A
