Vehicle-mounted camera control method, device and system, electronic equipment and medium

Abstract

The invention provides a vehicle-mounted camera control method, device and system, electronic equipment and a medium. The method comprises the steps that the power state and the vehicle door state of a target vehicle are acquired; when the target vehicle is in the power-off state, whether the vehicle door state is a locked state or not is judged, and if yes, a remnant detection function of a cabin camera of the target vehicle is started; when the target vehicle is in the power-on state, the vehicle speed of the target vehicle is obtained; whether the vehicle speed of the target vehicle reaches or exceeds a preset vehicle speed threshold value or not is judged, if yes, it is judged that the target vehicle is in a high-speed driving state, and the high-speed driving duration time of the target vehicle is obtained; and if the high-speed driving duration time of the target vehicle exceeds a preset first time threshold value, starting a driver monitoring function of the cabin camera. Through vehicle speed monitoring and duration judgment, timely starting of a driver monitoring function is realized, fatigue driving is effectively prevented, and driving safety is enhanced.

Description

Technical Field

[0001] The present disclosure relates to the field of vehicle intelligence technology, and in particular to a method, device, system, electronic equipment and medium for controlling a vehicle-mounted camera. Background Art

[0002] With the rapid development of the automobile industry, intelligence has become a significant trend in modern automobiles. In this context, vehicle visual perception systems, as one of the key technologies for achieving intelligence, are increasingly being used. These systems provide assisted driving functions and enhanced safety for vehicles by capturing and analyzing visual information about the vehicle's surroundings.

[0003] However, with the increase in the number of vehicle visual perception systems, each system needs to be equipped with corresponding perception sensors (mainly cameras), controller units with different computing power levels, and wiring harnesses connecting these components. This configuration has led to a significant increase in the hardware cost of the entire vehicle, especially in high-end models, where the cost increase is more significant. In addition, the existence of multiple systems also increases the complexity of the interior and exterior layout of the vehicle. Each system needs to be carefully laid out to ensure that they do not interfere with each other, while also taking into account the aesthetics and aerodynamics of the vehicle.

[0004] Therefore, how to effectively integrate these systems, reduce costs, simplify layout, and maintain or improve performance has become a pressing issue to be addressed in the field of automotive intelligence. Summary of the invention

[0005] The purpose of the present disclosure is to provide a vehicle-mounted camera control method, device, system, electronic device and medium, aiming to reduce the number of visual sensors and improve computing power utilization.

[0006] A first aspect of the present disclosure provides a vehicle-mounted camera control method, comprising: obtaining a power state and a door state of a target vehicle, wherein the power state includes a power-on state and a power-off state, and the door state includes an unlocked state and a locked state; when the target vehicle is in a power-off state, determining whether the door state is a locked state, and if so, activating a debris detection function of a cabin camera of the target vehicle; when the target vehicle is in a power-on state, obtaining a vehicle speed of the target vehicle; determining whether the vehicle speed of the target vehicle reaches or exceeds a preset vehicle speed threshold, and if so, determining that the target vehicle is in a high-speed driving state, and obtaining a duration of the high-speed driving of the target vehicle; if the duration of the high-speed driving of the target vehicle exceeds a preset first time threshold, activating a driver monitoring function of the cabin camera.

[0007] In this implementation, the intelligent activation of the cabin camera's object detection function is achieved by monitoring the vehicle's power status and door status. When the vehicle is powered off and the doors are locked, object detection is automatically activated, improving vehicle safety and user convenience. At the same time, by monitoring vehicle speed and judging duration, the driver monitoring function is activated in a timely manner, effectively preventing fatigue driving and enhancing driving safety.

[0008] In an embodiment of the first aspect, the method of starting the left-behind object detection function of the cabin camera of the target vehicle includes: regularly acquiring the interior image of the target vehicle through the cabin camera; processing the interior image based on an image algorithm to determine whether there is a left-behind object in the interior of the target vehicle; if so, sending the interior image and an alarm message to a user; if not, only sending the interior image to the user.

[0009] In this implementation, by regularly acquiring images inside the vehicle and processing them using image algorithms, accurate judgment of objects left inside the vehicle is achieved. When a left-behind object is detected, an alarm message and images inside the vehicle are sent to the user in a timely manner, thereby improving the safety of the vehicle and the safety of the user's belongings.

[0010] In an embodiment of the first aspect, the driver monitoring function of the cockpit camera includes one or more combinations of fatigue status monitoring, dangerous driving behavior monitoring and distracted driving monitoring.

[0011] In this implementation, by combining fatigue status monitoring, dangerous driving behavior monitoring and distracted driving monitoring, comprehensive driver status monitoring is provided, which can detect the driver's abnormal behavior in time and take measures, significantly improving driving safety.

[0012] In an embodiment of the first aspect, the fatigue status monitoring includes: acquiring the driver's facial image in real time through the cockpit camera; obtaining the driver's eyelid closure degree and mouth opening degree based on the driver's facial image; setting eyelid closure fatigue judgment conditions and yawning fatigue judgment conditions respectively; within a preset time, counting the number of times the driver's eyelid closure degree meets the eyelid closure fatigue judgment conditions and the number of times the mouth opening degree meets the yawning fatigue judgment conditions; judging whether the driver is in a fatigue state based on the statistical results, and if so, starting a fatigue driving warning mechanism.

[0013] In this implementation, accurate monitoring of the driver's fatigue state is achieved by acquiring the driver's facial image in real time and analyzing the eyelid closure and mouth opening. Through the preset eyelid closure and yawning judgment conditions, it is possible to accurately count and judge whether the driver is fatigued, and activate the early warning mechanism in time to effectively prevent accidents caused by fatigue driving.

[0014] In an embodiment of the first aspect, it also includes selecting different sensitivity modes to monitor the driver's fatigue state: obtaining road condition information of the target vehicle; based on the road condition information, dynamically adjusting the frame rate of the driver's image captured by the cockpit camera; based on the frame rate of the driver's image captured by the cockpit camera, selecting different sensitivity modes to monitor the driver's fatigue state; wherein, when the frame rate of the driver's image captured by the cockpit camera is lower than a preset frame rate threshold, a high sensitivity mode is used to monitor the driver's fatigue state; when the frame rate of the driver's image captured by the cockpit camera is higher than the preset frame rate threshold, a low sensitivity mode is used to monitor the driver's fatigue state.

[0015] In this implementation, the cockpit camera frame rate is dynamically adjusted based on real-time road condition information, optimizing the allocation of computing power of the shared controller. The frame rate is reduced in complex road conditions to support the high computing power requirements of the front-view camera, while the frame rate is increased in simple road conditions to enhance driver monitoring, thereby improving computing power utilization and monitoring efficiency, and also achieving dynamic adjustment of the sensitivity of driver fatigue status monitoring.

[0016] In an embodiment of the first aspect, judging whether the driver is in a fatigue state based on statistical results includes: in a low-sensitivity mode, when the number of times the counted eyelid closure meets the eyelid closure fatigue judgment condition reaches a first threshold, judging that the driver is in a fatigue state; in a high-sensitivity mode, when the number of times the counted eyelid closure meets the eyelid closure fatigue judgment condition reaches a second threshold, judging that the driver is in a fatigue state; wherein the first threshold is greater than the second threshold; in a low-sensitivity mode, when the number of times the counted mouth opening meets the yawn fatigue judgment condition reaches a third threshold, judging that the driver is in a fatigue state; in a high-sensitivity mode, when the number of times the counted mouth opening meets the yawn fatigue judgment condition reaches the fourth threshold, judging that the driver is in a fatigue state; wherein the third threshold is greater than the fourth threshold.

[0017] In this implementation, different thresholds are set to distinguish fatigue judgments in high-sensitivity mode and low-sensitivity mode, so that fatigue monitoring is more accurate, the possibility of misjudgment is reduced, and the reliability of the early warning mechanism is improved.

[0018] In an embodiment of the first aspect, the frame rate of the cockpit camera for collecting the driver's image is dynamically adjusted based on the road condition information, including: when the target vehicle is in a complex road condition, reducing the frame rate of the cockpit camera for collecting the driver's image; when the target vehicle is in a simple road condition, increasing the frame rate of the cockpit camera for collecting the driver's image.

[0019] In this implementation, the frame rate of the cockpit camera is dynamically adjusted based on the road condition information to optimize the use of system resources. The frame rate is lowered in complex road conditions to reduce processing requirements, and the frame rate is increased in simple road conditions to increase the monitoring frequency, thereby improving the adaptability and efficiency of the system.

[0020] In an embodiment of the first aspect, the dangerous driving behavior monitoring includes: acquiring a head image of the driver through the cockpit camera; based on the driver's head image, obtaining a driver's face target frame and a driving interference object target frame; based on the spatial intersection and intersection duration of the face target frame and the driving interference object target frame, determining whether the driver is engaging in dangerous driving behavior; if so, initiating a dangerous driving warning mechanism.

[0021] In this implementation, by analyzing the intersection of the face target frame in the driver's head image and the target frame of the object interfering with driving, accurate judgment of dangerous driving behavior is achieved. When dangerous driving behavior is detected, the early warning mechanism is activated in time to effectively prevent traffic accidents caused by distraction or dangerous behavior.

[0022] In an embodiment of the first aspect, the distracted driving monitoring includes: acquiring a head image of the driver through the cockpit camera; calculating the driver's head posture angle based on the driver's head image; judging whether the driver is distracted driving based on the head posture angle and the duration of the head posture angle; if so, initiating a distracted driving warning mechanism.

[0023] This application realizes effective monitoring of distracted driving behavior by analyzing the driver's head posture angle and its duration. When it is judged that the driver is distracted, the early warning mechanism is activated in time, improving the safety of the driving process.

[0024] The second aspect of the present disclosure discloses a vehicle-mounted camera control device, comprising: a first acquisition module, used to acquire the power state and door state of a target vehicle, wherein the power state includes a power-on state and a power-off state, and the door state includes an unlocked state and a locked state; a first starting module, used to determine whether the door state is a locked state when the target vehicle is in a power-off state, and if so, to start the debris detection function of the cabin camera of the target vehicle; a second acquisition module, used to acquire the speed of the target vehicle when the target vehicle is in a power-on state; a high-speed judgment module, used to determine whether the speed of the target vehicle reaches or exceeds a preset speed threshold, and if so, to determine that the target vehicle is in a high-speed driving state; a third acquisition module, used to acquire the duration of the high-speed driving of the target vehicle; and a second starting module, used to start the driver monitoring function of the cabin camera if the duration of the high-speed driving of the target vehicle exceeds a preset first time threshold.

[0025] The third aspect of the present disclosure discloses a vehicle-mounted visual perception system, comprising: a visual sensor unit, comprising a front-view camera and a cockpit camera; wherein the front-view camera is used to collect forward driving images; the cockpit camera is used to collect images in the vehicle cabin, wherein the cockpit camera is used to monitor the driver's state when the vehicle is powered on and driving; and is used to detect objects left in the cabin when the vehicle is powered off and locked; a controller unit, which is communicatively connected to the visual sensor unit, comprises a SOC module and an MCU module, wherein the MCU module is used to control the SOC module to power on, receive vehicle data parameters and transmit the vehicle data parameters to the SOC module, and perform safety monitoring; the SOC module comprises an image acquisition module, an image algorithm module, a control algorithm module and a data transmission module; wherein the image acquisition module is used to control the visual sensor unit camera to acquire images; the image algorithm module is used to identify and detect the acquired images; the control algorithm module is used to adjust the algorithm function according to the vehicle data parameters and the image recognition results; the data transmission module is used to transmit the vehicle cabin image taken by the cockpit camera and the result of detecting objects left in the cabin to an external device when the vehicle is powered off and locked.

[0026] In this implementation, the cockpit camera simultaneously realizes the functions of driver monitoring and object detection, and realizes the functional integration of two independent cameras, thereby reducing the number of required visual sensors and the corresponding number of wiring harnesses, which significantly reduces the hardware cost. The cockpit camera and the front-view camera share a controller unit, which reduces the number of required controller units, thereby reducing the corresponding wiring harnesses and space occupation.

[0027] The fourth aspect of the present disclosure discloses an electronic device, comprising: a processor and a memory; wherein the memory is used to store a computer program; the processor is used to execute the computer program stored in the memory, so that the electronic device executes the vehicle camera control method as described in any one of the first aspects.

[0028] A fifth aspect of the present disclosure discloses a computer-readable storage medium having a computer program stored thereon, which, when executed by an electronic device, implements the vehicle-mounted camera control method described in any one of the first aspects. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 Shown is a schematic diagram of an exemplary application scenario of the vehicle camera control method described in an embodiment of the present application.

[0030] Figure 2 Shown is an exemplary flow chart of the vehicle camera control method described in an embodiment of the present application.

[0031] Figure 3 Shown is a schematic diagram of the cockpit camera debris detection process described in an embodiment of the present application.

[0032] Figure 4 Shown is a schematic diagram of the startup logic of the cockpit camera described in an embodiment of the present application.

[0033] Figure 5 Shown is a schematic diagram of the fatigue driving monitoring process described in an embodiment of the present application.

[0034] Figure 6 Shown is a schematic diagram of facial key points described in an embodiment of the present application.

[0035] Figure 7 Shown is a schematic diagram of the frame rate adjustment process described in an embodiment of the present application.

[0036] Figure 8 Shown is a flow chart of the relationship between frame rate adjustment and sensitivity described in an embodiment of the present application.

[0037] Fig. 9 Shown is a schematic diagram of the dangerous driving monitoring process described in an embodiment of the present application.

[0038] Fig.10 Shown is a schematic diagram of the target frame intersection described in an embodiment of the present application.

[0039] Fig.11 Shown is a schematic diagram of the distracted driving monitoring process described in an embodiment of the present application.

[0040] Fig.12 Shown is a flow chart of the cockpit camera driver monitoring function described in an embodiment of the present application.

[0041] Fig.13 Shown is an exemplary structural schematic diagram of the vehicle-mounted camera control device described in an embodiment of the present application.

[0042] Fig.14 Shown is a structural schematic diagram of the visual perception system described in an embodiment of the present application.

[0043] Fig.15 , Fig.16 Shown is a hardware schematic diagram of the visual perception system described in an embodiment of the present application.

[0044] Fig.17 Shown is an exemplary structural diagram of an electronic device described in an embodiment of the present application. DETAILED DESCRIPTION

[0045] The following describes the embodiments of the present invention by specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.

[0046] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present invention, and thus the drawings only show components related to the present invention rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0047] In the development of automobile intelligence, vehicle visual perception systems play a vital role. These systems can provide detailed information about the vehicle's surrounding environment and the internal conditions of the cabin through various intelligent sensing devices installed on the vehicle. More importantly, these systems can perceive forward information, which is crucial for realizing advanced driver assistance system (ADAS) functions such as lane keeping and collision warning. At the same time, the perception of in-cabin information is also important for monitoring the driver's status and improving vehicle safety.

[0048] However, there are some limitations in the existing technology. Usually, each specific function requires a separate camera to implement, and each camera needs to be equipped with an independent controller unit. This design leads to a significant increase in the hardware cost of the whole vehicle, especially in high-end models. In addition, as the number of visual perception devices installed on the vehicle increases, the complexity of the wiring also increases, which not only increases the difficulty of vehicle manufacturing, but also increases the difficulty of maintenance and repair.

[0049] Furthermore, each camera is equipped with an independent controller unit, which may lead to redundant computing resources. Since different cameras may have different requirements for computing power at different times, this design may cause some controller units to be idle at certain times, while other times may not be able to meet high-load computing requirements. This computing power redundancy not only reduces the overall efficiency of the system, but also causes a waste of resources.

[0050] In view of the above problems, the present invention proposes an integrated solution for a vehicle-mounted visual perception system. The technical solution in the embodiment of the present application will be described in detail below in conjunction with the accompanying drawings in the embodiment of the present application.

[0051] like Figure 1As shown, it shows a schematic diagram of an application scenario of the present application. The vehicle-mounted camera control scenario includes: a front-view camera 10, a cockpit camera 20, a data processing module 30 and an external data receiving device 40.

[0052] The forward-looking camera 10 is a camera device mounted on the front windshield of the vehicle, and its main function is to capture real-time image data in front of the vehicle. The camera has high-resolution imaging capabilities and can work stably under various lighting conditions to ensure the clarity and accuracy of the image data. The forward-looking camera 10 is usually equipped with a wide-angle lens to obtain a wide field of view and cover key visual information in front of the vehicle. In addition, the forward-looking camera 10 may also be integrated with image stabilization technology and night vision functions to adapt to different driving environments. In the application scenario of the present application, the forward-looking camera 10 is used to implement ADAS functions, such as lane departure warning, forward collision warning, etc., to provide the vehicle with detailed visual information of the environment ahead.

[0053] The cockpit camera 20 is a camera device installed inside the vehicle, which is designed to monitor the situation in the vehicle cabin, including the driver's status and detection of objects left in the vehicle. The camera has facial recognition and behavior analysis functions, and can monitor the driver's facial expressions, eye closure frequency and other physiological characteristics in real time to assess the driver's fatigue or distracted behavior. When the vehicle is powered on and driving, the cockpit camera 20 is used to monitor the driver's status; when the vehicle is powered off and locked, it is used for detection of objects left in the cabin to detect whether there are mobile phones, bags, living biological objects, etc. left in the cabin. The image acquisition function of the cockpit camera 20 is crucial to improving driving safety and vehicle management efficiency.

[0054] The data processing module 30 is the core component of the vehicle camera control system, responsible for processing the image data collected by the front camera 10 and the cockpit camera 20. The module contains a high-performance processor and advanced image processing algorithms, which can analyze and recognize the collected images in real time. The data processing module 30 adjusts the algorithm function according to the vehicle data parameters and image recognition results to achieve intelligent perception of the vehicle's surrounding environment and internal conditions. In addition, the data processing module 30 is also responsible for transmitting the processed data to the external data receiving device 40 to achieve remote monitoring and application of the data.

[0055] The external data receiving device 40 is a device that is connected to the vehicle camera control system for receiving and displaying data transmitted by the data processing module 30. The device can be a driver's smart phone, tablet computer or other dedicated monitoring device. The external data receiving device 40 can not only display the image information inside and around the vehicle in real time, but also receive warning signals and status reports from the data processing module 30, so that the driver or the remote monitoring center can understand the vehicle status in time and take corresponding measures. This design enhances the vehicle's ability to interact with the outside world and improves the convenience and safety of vehicle use.

[0056] like Figure 2 As shown, an embodiment of the present application provides a vehicle camera control method, including the following steps S210 to S290.

[0057] Step S210: Acquire the power state and door state of the target vehicle, wherein the power state includes a power-on state and a power-off state, and the door state includes an unlocked state and a locked state.

[0058] Specifically, the power state refers to the "power-on state" and the "power-off state", that is, the start-up or shutdown state of the vehicle. The power-on state generally refers to the state in which the vehicle has been started, the engine is running, and the various electronic devices of the vehicle can work normally. On the contrary, the power-off state refers to the state in which the vehicle has been shut down, the engine has stopped, and the electronic devices of the vehicle are in standby or shutdown state. By monitoring the power state, the system can determine whether the vehicle is in operation and decide whether to start or shut down the camera accordingly.

[0059] The door status refers to the "locked state" and "unlocked state" of the vehicle door, that is, the closed or open state. The locked state refers to the state in which the door is closed and locked and cannot be opened from the outside. The unlocked state refers to the state in which the door is open or can be opened.

[0060] Step S220: Determine whether the target vehicle is in a power-off state.

[0061] It should be noted that when the target vehicle is in a powered-off state, the front-view camera is turned off.

[0062] If yes, then proceed to step S230: determine whether the door is in a locked state;

[0063] If so, proceed to step S240: start the leftover object detection function of the cabin camera of the target vehicle.

[0064] In some embodiments, Figure 3 As shown, the step S240 further includes steps S241 to S244.

[0065] Step 241: Periodically obtain the interior image of the target vehicle through the cockpit camera.

[0066] Specifically, the cockpit camera automatically activates at preset time intervals or under certain conditions to capture real-time images of the interior of the vehicle. These images may include every corner of the passenger compartment to ensure full coverage without leaving blind spots. In addition, the system ensures that the captured images have sufficient resolution and clarity so that subsequent image processing and analysis can be carried out accurately.

[0067] Step 242: Process the interior image based on an image algorithm to determine whether there is any object left inside the target vehicle.

[0068] Specifically, the captured image will be fed into the image processing module, which is equipped with advanced image recognition algorithms to identify and distinguish various objects in the car. The system compares the features in the image with the features of known objects in the database to identify the objects in the image and determine whether there are any leftovers.

[0069] If yes, proceed to step 243: send the in-car image and alarm information to the user.

[0070] Specifically, if there is an object left in the car, an alarm message is generated to remind the user that there is an object left in the car, and may include detailed information such as the type and location of the object. Through a wireless communication module such as Bluetooth, Wi-Fi or mobile network, the image of the car and the alarm message are sent to the user's smartphone or other remote device.

[0071] If not, proceed to step 244: only send the in-vehicle image to the user.

[0072] Specifically, if there is no object left in the car, the captured image of the car is sent to the user's remote device, so that the user can view the situation inside the car. The user confirms the situation inside the car by viewing the sent image to ensure the safety of the vehicle.

[0073] In this implementation, through intelligent image capture, processing and analysis, effective monitoring of objects left inside the vehicle is achieved, thereby improving the safety of the vehicle and the trust of users.

[0074] Step S250: When the target vehicle is in a powered-on state, the speed of the target vehicle is obtained.

[0075] Specifically, when the vehicle is powered on, that is, the vehicle is started and ready to drive or is driving, the vehicle speed data will be obtained in real time through sensors inside the vehicle or modules connected to the vehicle data bus. These data are usually provided by the vehicle's speed sensor, which directly measures the rotation speed of the wheels or indirectly measures the vehicle speed through other means. The acquired vehicle speed data will be used for further processing, including but not limited to calculation, comparison and decision support. These data processing will provide key information for the vehicle's driving assistance systems, such as adaptive cruise control, forward collision warning system, etc.

[0076] It should be noted that when the vehicle is powered on, the front camera is automatically activated. The front camera is a key component in ADAS and is used to capture real-time image data in front of the vehicle. The front camera is equipped with high-resolution imaging capabilities and can work stably under various lighting conditions to ensure the clarity and accuracy of image data. The system will process the captured images in real time to identify road signs, traffic signals, pedestrians, other vehicles, and potential obstacles.

[0077] Step S260: determining whether the speed of the target vehicle reaches or exceeds a preset speed threshold;

[0078] If so, proceed to step S270: determine that the target vehicle is in a high-speed driving state, and obtain the duration of the high-speed driving of the target vehicle.

[0079] Specifically, the vehicle speed data is obtained through the vehicle's speed sensor or on-board diagnostic system interface, and the data is compared with a pre-set threshold. If the vehicle speed reaches or exceeds the threshold, the system determines that the vehicle is in a high-speed driving state and begins to record the duration of the vehicle maintaining this state. This determination is crucial for subsequent steps because it triggers further monitoring of the driver's status.

[0080] Step S280: If the duration of the target vehicle's high-speed driving exceeds a preset first time threshold, proceed to step S290: start the driver monitoring function of the cockpit camera.

[0081] Specifically, a time threshold is preset to determine when to activate the driver monitoring function. When the duration of high-speed driving of the vehicle exceeds this preset threshold, the system considers that the driver may be in a state of fatigue or distraction, and thus activates the driver monitoring function of the cockpit camera. This function monitors the driver's behavior and physiological state in real time, promptly detects signs of fatigue or distraction, and takes corresponding early warning measures to ensure driving safety.

[0082] For example, in order to better illustrate the vehicle camera control method in this implementation, a specific application example is listed below. Figure 4shown.

[0083] When the vehicle is powered off, if the door unlocking is detected, the system will automatically turn off the cabin camera and front camera to protect the privacy of passengers. On the contrary, when the door is locked, the system will start the cabin camera and activate the left-behind detection function. At this time, the cabin camera is responsible for collecting image information inside the vehicle to detect whether there are left-behind objects, such as mobile phones, bags or living creatures.

[0084] After the vehicle is powered on, the system will determine whether to start the monitoring functions of the cockpit camera and the front camera based on the real-time vehicle speed and duration. This judgment mechanism avoids invalid calculations in scenarios such as waiting at traffic lights, thereby saving computing resources.

[0085] When the vehicle speed V reaches or exceeds 0km / h, the system will turn on the front camera and activate the driving ADAS function in the controller unit. At this time, the front camera is responsible for collecting road information ahead to support the vehicle's ADAS functions, such as lane keeping and collision warning.

[0086] When the vehicle speed V reaches or exceeds the preset speed threshold V1km / h, and the duration reaches or exceeds the preset time threshold T1, the system will turn on the cockpit camera and start the driver monitoring function. The cockpit camera is responsible for monitoring the driver's fatigue, dangerous driving behavior, and distraction to ensure driving safety. If the vehicle speed V drops below the speed threshold V1km / h, and the duration reaches or exceeds the time threshold T1, the system will turn off the cockpit camera and stop the driver monitoring function to reduce unnecessary computing power consumption.

[0087] In some embodiments, the driver monitoring function of the cockpit camera includes one or more combinations of fatigue status monitoring, dangerous driving behavior monitoring, and distracted driving monitoring.

[0088] Further, such as Figure 5 As shown, the fatigue state monitoring includes steps S310 to S360.

[0089] Step S310: Acquire the driver's facial image in real time through the cockpit camera.

[0090] Specifically, the cockpit camera is used to capture the driver's facial image in real time. This step is the basis of the fatigue monitoring function. The cockpit camera can accurately capture the real-time image of the driver's face through a specific facial recognition algorithm, providing visual data for subsequent fatigue feature analysis.

[0091] Step S320: Based on the facial image of the driver, the eyelid closure degree and mouth opening degree of the driver are obtained.

[0092] Specifically, the degree of fatigue is determined by judging whether the driver closes his eyes or yawns. First, a deep learning model is used to detect the face frame in the image. The face frame is pre-processed and input into the face key point detection module to detect 68 face key points and output the image coordinates. The positions of the 68 face key points are as follows: Figure 6 As shown. Calculate the eyelid closure using the image coordinates of the key points of the face ear Determine whether the eyelids are closed and the mouth ear The degree of opening and closing determines whether the mouth is open.

[0093] The formula for calculating eyelid closure is:

[0094]

[0095] Among them, Eye left_ear Indicates the degree of eyelid closure of the left eye; Eye right_ear Indicates the right eyelid closure.

[0096] Furthermore, the calculation formula for the left eyelid closure is:

[0097]

[0098] The calculation formula for the right eyelid closure is:

[0099]

[0100] The calculation formula for mouth opening and closing is:

[0101]

[0102] Among them, P x ,P y Indicates the key points of the face, x=1~68, y=1~68; |P x -P y | represents the image pixel distance between two face key points.

[0103] For example, Eye ear ≤0.15 means eyelid closure, Eye ear ≤0.10 indicates complete eyelid closure.

[0104] For example, Mouth ear ≥M1 means mouth is open. ear ≥M2 means the mouth is fully open. The value range of M1 is 0.30-0.38; the value range of M2 is 0.50-0.60.

[0105] In some embodiments, Figure 7As shown, the cabin camera obtains the eyelid closure degree of the driver's face at different frame rates, and the steps are S321 to S323.

[0106] Step S321: Obtain the road condition information of the target vehicle.

[0107] Specifically, collect the road condition information of the current driving section of the target vehicle. This information is crucial for the system to decide how to allocate processing capabilities. The road condition information can be obtained in various ways, including but not limited to the vehicle's own sensors (such as front-view cameras, radars, lidars, etc.), in-vehicle GPS systems, map data services, and communication with the traffic management center. The system analyzes these data to identify whether the vehicle is in an urban congested section, highway, rural road, or other specific road conditions. This information will be used for subsequent frame rate adjustment decisions.

[0108] Step S322: Dynamically adjust the frame rate of the cabin camera to capture the driver's image based on the road condition information.

[0109] Specifically, according to the obtained road condition information, dynamically adjust the frame rate of the cabin camera to capture the driver's image. The purpose of this step is to optimize the use of the vehicle's computing resources while ensuring the effectiveness of the driver monitoring function. The system will adjust the frame rate according to the complexity of the road conditions to meet different monitoring requirements.

[0110] In some embodiments, step S322 further includes: when the target vehicle is in a complex road condition, reduce the frame rate of the cabin camera to capture the driver's image; when the target vehicle is in a simple road condition, increase the frame rate of the cabin camera to capture the driver's image.

[0111] Specifically, in this implementation, the front-view camera and the cabin camera share a controller unit, and the total computing power of the controller unit is S, in TOPS, where Sd represents the computing power required for the driving function of the front-view camera, and Sc represents the computing power required for the cabin camera function.

[0112] In the normal driving state, the total computing power requirements of the front-view camera and the cabin camera do not exceed K1 of the total computing power of the controller, where the value range of K1 is between 80% and 95%, that is, Sc + Sd < K1 * S. This design provides sufficient computing power margin for the system to handle sudden computing requirements, while ensuring the normal operation of the ADAS function and the driver monitoring function.

[0113] When encountering complex road conditions, such as urban traffic congestion or winding mountain roads, the computing power requirement of the front-view camera increases, and the controller unit will dynamically adjust the frame rate of the cabin camera to ensure that the total computing power requirement does not exceed the computing power limit of the controller. Specifically, when Sc + Sd ≥ K1 * S, the system will execute the frame rate reduction strategy to reduce the frame rate of the cabin camera and release computing power for the front-view camera. This strategy helps to maintain the responsiveness and accuracy of the ADAS function under complex road conditions.

[0114] Conversely, when the vehicle re-enters simple road conditions, such as a straight highway, and Sc + Sd < K1 * S for a duration T that reaches or exceeds the preset threshold Tc, the controller unit will execute the frame rate increase strategy to increase the frame rate of the cabin camera. This dynamic adjustment mechanism ensures that system resources are optimally allocated under different road conditions while maintaining the continuity and effectiveness of the driver monitoring function.

[0115] Exemplarily, the frame rate adjustment of the cabin camera is carried out in steps of 5 frames, which helps to smoothly adjust the monitoring frequency and reduce the impact on performance.

[0116] Step S323: Based on the frame rate at which the cabin camera captures the driver's image, select different sensitivity modes to monitor the driver's fatigue state; among them, when the frame rate at which the cabin camera captures the driver's image is lower than the preset frame rate threshold, a high-sensitivity mode is used to monitor the driver's fatigue state; when the frame rate at which the cabin camera captures the driver's image is higher than the preset frame rate threshold, a low-sensitivity mode is used to monitor the driver's fatigue state.

[0117] Specifically, the appropriate fatigue monitoring sensitivity mode is selected according to the real-time frame rate F of the cabin camera, the maximum frame rate Q of the preset cabin camera, and two key percentage thresholds K2 and K3, where the value range of K2 is 30% to 45%, and the value range of K3 is 10% to 20%. These thresholds are obtained based on experimental data and actual driving scenario analysis to determine the sensitivity requirements for driver fatigue monitoring under different road conditions.

[0118] When Q * K3 ≤ F ≤ Q * K2 of the cabin camera, a high-sensitivity mode is used for fatigue monitoring. In this mode, the system will adopt more stringent judgment criteria and a more intensive monitoring frequency to ensure that fatigue signs of the driver can still be accurately captured when the frame rate is low.

[0119] When Q * K2 ≤ F ≤ Q of the cabin camera, a low-sensitivity mode is adopted. In this mode, the system can appropriately reduce the monitoring frequency and complexity to reduce the consumption of computing resources while maintaining effective monitoring of the fatigue state.

[0120] For example, Figure 8 As shown, the relationship between computing power, dynamic frame rate adjustment, and dynamic sensitivity adjustment is demonstrated.

[0121] In this implementation, the high sensitivity mode and the low sensitivity mode are switched dynamically according to the relationship between the real-time frame rate F and the preset thresholds K2 and K3. This dynamic adjustment mechanism enables the system to flexibly adjust the monitoring strategy according to the current frame rate to adapt to different driving environments and needs.

[0122] Step S330: setting the eyelid closure fatigue judgment condition and the yawn fatigue judgment condition respectively.

[0123] In some embodiments, the driver's fatigue behavior is judged by the driver's eyelid closure degree and is divided into single eyelid closure fatigue and multiple eyelid closure fatigue.

[0124] The single eyelid closure fatigue judgment condition is to satisfy both formula 1-1 and formula 1-2; the specific formula is as follows:

[0125] T b ≥T tbs Formula 1-1

[0126] T wb ≥K4*T b Formula 1-2

[0127] Among them, T b represents the duration of a single eyelid closure; T tbs represents the single eyelid closure fatigue time threshold; T wb Indicates the duration of a single eyelid closure; the value of K4 ranges from 17% to 25%.

[0128] The multiple eyelid closure fatigue judgment condition is to satisfy formula 1-3 and formula 1-4 at the same time; the specific formula is as follows:

[0129] T tbd ≤T b <T tbs Formula 1-3

[0130] T wb ≥K5*T b Formula 1-4

[0131] Among them, T b represents the duration of a single eyelid closure; T tbs represents the single eyelid closure fatigue time threshold; T wb Indicates the duration of a single eyelid closure; T tbd : Indicates the single eye closure threshold time for multiple eyelid closure fatigue warning; the value range of K5 is 10% to 20%.

[0132] The yawn fatigue judgment condition is to satisfy both formula 1-6 and formula 1-7; the specific formula is as follows:

[0133] T m ≥T mt Formula 1-6

[0134] T wm ≥K6*T m Formula 1-7

[0135] Among them, T m Indicates the duration of a single mouth opening; T mt Indicates the fatigue time threshold of a single mouth opening; T wm Indicates the duration of a single full mouth opening; the value range of K6 is 20% to 30%.

[0136] Step S340: Within a preset time, the number of times the driver's eyelid closure degree meets the eyelid closure fatigue judgment condition and the number of times the mouth opening degree meets the yawn fatigue judgment condition are counted.

[0137] Step S350: Determine whether the driver is in a fatigue state based on the statistical results, and if so, proceed to step S360.

[0138] Specifically, since it is difficult to detect the driver's fatigue state due to instantaneous movements such as blinking, it is necessary to set fatigue determination conditions according to different sensitivities, that is, step S350 further includes:

[0139] In low sensitivity mode, when the number of times the counted eyelid closure degree meets the eyelid closure fatigue judgment condition reaches a first threshold value N1, it is determined that the driver is in a fatigue state; in high sensitivity mode, when the number of times the counted eyelid closure degree meets the eyelid closure fatigue judgment condition reaches a second threshold value N2, it is determined that the driver is in a fatigue state; wherein the first threshold value N1 is greater than the second threshold value N2.

[0140] Exemplarily, the first threshold N1≥4, and the second threshold N2≤N2<4.

[0141] In the low-sensitivity mode, when the number of times the counted mouth opening and closing degrees meet the yawn fatigue judgment condition reaches the third threshold value N3, it is determined that the driver is in a fatigue state; in the high-sensitivity mode, when the number of times the counted mouth opening and closing degrees meet the yawn fatigue judgment condition reaches the fourth threshold value N4, it is determined that the driver is in a fatigue state; wherein the third threshold value N3 is greater than the fourth threshold value N4.

[0142] Exemplarily, the third threshold N3≥5, and the fourth threshold N3≤N4<5.

[0143] Step S360: Activate the fatigue driving warning mechanism.

[0144] Specifically, the warning mechanism may include sound alarms, visual prompts or physical feedback, aiming to remind the driver to take a rest to prevent traffic accidents caused by fatigue driving.

[0145] In some embodiments, Fig. 9 As shown, the dangerous driving behavior monitoring includes steps S410 to S440.

[0146] Step S410: Acquire the driver's head image through the cockpit camera.

[0147] Specifically, dangerous driving behavior mainly detects whether the driver engages in dangerous behaviors such as talking on the phone or smoking for a long time while driving, so it is necessary to obtain images of the driver's head area through the cockpit camera.

[0148] Step S420: Based on the driver's head image, a driver face target frame and a driving interference object target frame are obtained.

[0149] Specifically, based on the driver's head image, the head image is fed into a deep learning module to identify and locate the driver's face target frame and the object target frame that may interfere with driving. This step uses computer vision technology, including but not limited to image recognition, target detection and feature tracking algorithms, to accurately identify and locate the driver's face and objects in the car that may distract the driver, such as mobile phones, food or other items.

[0150] Step S430: Based on the spatial intersection of the face target frame and the driving-interfering object target frame and the intersection duration, determine whether the driver is engaging in dangerous driving behavior.

[0151] Specifically, when the driver's face target frame intersects with the interference object target frame and this intersection lasts for a certain period of time, it will be judged that the driver may be engaging in dangerous driving behavior, such as using a mobile phone or eating while driving.

[0152] For example, the target box intersection is determined as follows: Fig.10 As shown: the face detection frame is A(X1,Y1), and the pixel coordinates of the upper left corner and lower right corner of the face detection frame in the image are (X a1 ,Y a1 ),(X a2 ,Y a2 ), the driving interference object detection box is B(X2,Y2), and the pixel coordinates of the upper left corner and lower right corner of the smoke / mobile phone detection box in the image are (X b1 ,Y b1 ),(X b2 ,Y b2). Among them, X1=max(X a1 ,X b1 ); Y1=max(Y a1 ,Y b1 ); X2 = min(X a2 ,X b2 ); Y2 = min(Y a2 ,Y b2 ).

[0153] The judgment conditions for dangerous driving behavior are Where T represents the duration of the intersection, T j Indicates the warning time threshold for dangerous driving behavior.

[0154] If yes, proceed to step S440: start the dangerous driving warning mechanism.

[0155] Specifically, the warning signal can take many forms, including but not limited to visual alarms, auditory alarms and tactile feedback. Visual alarms can display warning information through the vehicle's dashboard; auditory alarms can issue warning sounds through the vehicle's audio system; tactile feedback can be fed back to the driver through seat vibration or steering wheel.

[0156] In some embodiments, Fig.11 As shown, the distracted driving monitoring includes steps S510 to S540.

[0157] Step S510: Acquire the driver's head image through the cockpit camera.

[0158] Specifically, distracted driving behavior mainly detects whether the driver does not focus on the area directly ahead for a long time while driving.

[0159] Step S520: Calculate the driver's head posture angle based on the driver's head image.

[0160] Specifically, the image algorithm module processes the collected images, detects the face target frame in the image, and sends the detected face image to the deep learning model. The model can output the pitch angle (Pitch) and yaw angle (Yaw) of the driver's head. These angle data reflect the offset of the driver's head relative to the front.

[0161] Step S530: Based on the head posture angle and the head posture angle duration, determine whether the driver is distracted driving. If so, proceed to step S540.

[0162] Specifically, the system determines whether the driver is distracted driving based on the calculated head posture angles and the duration of these angles. The system presets the safety threshold of the head posture angles and the duration threshold of these angles deviating from the safety threshold. When the driver's head posture angle exceeds the safety threshold and the duration exceeds the preset time threshold, the system will determine that the driver is distracted.

[0163] The criteria for distracted driving include: Pitch ≥ Pitch up or Pitch≥Pitch down or Yaw≥Yaw left or Yaw≥Yaw right , and T ≥ T t .

[0164] Pitch up Pitch angle head-up threshold obtained by calibration; Pitch down The pitch angle descent threshold obtained by calibration; Yaw left The calibrated left turn threshold of the deflection angle; Yaw right The right turn threshold of the deflection angle obtained by calibration; T t The distraction alarm time threshold.

[0165] Step S540: Activate the distracted driving warning mechanism.

[0166] For example, Fig.12 As shown, the logic of the cockpit camera driver monitoring function is fully demonstrated.

[0167] In one implementation of the present application, an embodiment of the present application provides a vehicle-mounted camera control device, such as Fig.13 As shown, the vehicle-mounted camera control device 1300 includes a first acquisition module 1301 , a first start module 1302 , a second acquisition module 1303 , a high-speed judgment module 1304 , a third acquisition module 1305 and a second start module 1306 .

[0168] The first acquisition module 1301 is used to acquire the power state and door state of the target vehicle, wherein the power state includes a power-on state and a power-off state, and the door state includes an unlocked state and a locked state; the first start module 1302 is used to determine whether the door state is a locked state when the target vehicle is in a power-off state, and if so, to start the debris detection function of the cabin camera of the target vehicle; the second acquisition module 1303 is used to acquire the speed of the target vehicle when the target vehicle is in a power-on state; the high-speed judgment module 1304 is used to determine whether the speed of the target vehicle reaches or exceeds a preset speed threshold, and if so, to determine that the target vehicle is in a high-speed driving state; the third acquisition module 1305 is used to acquire the duration of the high-speed driving of the target vehicle; the second start module 1306 is used to start the driver monitoring function of the cabin camera if the duration of the high-speed driving of the target vehicle exceeds a preset first time threshold.

[0169] The modules / units described as separate components may or may not be physically separated, and the components displayed as modules / units may or may not be physical modules, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules / units may be selected according to actual needs to achieve the purpose of the embodiments of the present application. For example, the functional modules / units in the various embodiments of the present application may be integrated into one processing module, or each module / unit may exist physically separately, or two or more modules / units may be integrated into one module / unit.

[0170] In one implementation of the present application, Fig.14 As shown, the embodiment of the present application provides a vehicle-mounted visual perception system, including:

[0171] The visual sensor unit includes a front-view camera and a cockpit camera; the front-view camera is used to collect forward driving images; the cockpit camera is used to collect images inside the vehicle cabin, wherein the cockpit camera is used to monitor the driver's status when the vehicle is powered on and driving; and is used to detect objects left in the vehicle cabin when the vehicle is powered off and locked.

[0172] Specifically, Fig.15 As shown, Fig.16 The figure shows the hardware installation diagram of the front view camera and the cockpit camera. The front view camera is embedded and fixed in front of the controller unit and installed on the front windshield; the cockpit camera is embedded and fixed behind the controller unit and installed on the front windshield.

[0173] The controller unit is connected to the visual sensor unit in communication, and includes a SOC module and an MCU module, wherein the MCU module is used to control the SOC module to power on, receive vehicle data parameters and transmit the vehicle data parameters to the SOC module, and perform safety monitoring; the SOC module includes an image acquisition module, an image algorithm module, a control algorithm module and a data transmission module; wherein,

[0174] The image acquisition module is used to control the camera of the visual sensor unit to acquire images;

[0175] The image algorithm module is used to identify and detect the collected images;

[0176] The control algorithm module is used to adjust the algorithm function according to the vehicle data parameters and image recognition results;

[0177] The data transmission module is used to transmit the vehicle cabin image taken by the cabin camera and the result of detecting the remains in the cabin to an external device when the vehicle is electrically locked.

[0178] In the above implementation, a cockpit camera is used to meet the functional requirements of the driver monitoring system (DMS) and the object detection system (OMS). This design reduces the number of visual sensors required and uses a single camera to monitor the driver's status and detect objects left in the cabin, thereby reducing hardware cost and complexity. The cockpit camera and the front-view camera share the same controller unit for data processing, reducing the need for additional controller units and further reducing material costs and assembly complexity. In addition, the data sensed by the front-view and cockpit cameras are directly sent to the controller unit for processing without the need for additional relay or conversion equipment, which improves the efficiency and reliability of data transmission and simplifies the system maintenance and fault diagnosis process.

[0179] In one implementation of the present application, an embodiment of the present application provides an electronic device, such as Fig.17 As shown, the electronic device includes one or more processors and one or more memories; the memories store computer-readable codes, and when the computer-readable codes are executed by the one or more processors, the vehicle-mounted camera control as described above is implemented. In addition, the electronic device may also include conventional electronic devices such as I / O interfaces and communication modules, which will not be elaborated here.

[0180] The descriptions of the processes or structures corresponding to the above-mentioned figures have different emphases. For parts that are not described in detail in a certain process or structure, please refer to the relevant descriptions of other processes or structures.

[0181] In one implementation of the present application, an embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the vehicle-mounted camera control as described above is implemented.

[0182] A person of ordinary skill in the art can understand that all or part of the steps in the method for implementing the above-mentioned embodiment can be completed by instructing a processor through a program, and the program can be stored in a computer-readable storage medium, and the storage medium is a non-transitory medium, such as a random access memory, a read-only memory, a flash memory, a hard disk, a solid-state hard disk, a magnetic tape, a floppy disk, an optical disc, and any combination thereof. The above-mentioned storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or a data center that includes one or more available media. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a tape), an optical medium (for example, a digital video disc (DVD)), or a semiconductor medium (for example, a solid-state disk (SSD)), etc.

[0183] In an implementation of the present application, the present application embodiment may also provide a computer program product, the computer program product including one or more computer instructions. When the computer instructions are loaded and executed on a computing device, the process or function described in the embodiment of the present application is generated in whole or in part. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from a website site, a computer or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer or data center. When the computer program product is executed by a computer, the computer executes the method described in the aforementioned method embodiment. The computer program product may be a software installation package, and in the case where the aforementioned method is required, the computer program product may be downloaded and executed on a computer.

[0184] In the several embodiments provided in the present application, it should be understood that the disclosed system, device or method can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of modules / units is only a logical function division, and there may be other division methods in actual implementation, such as multiple modules or units can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or modules or units, which can be electrical, mechanical or other forms.

[0185] Those of ordinary skill in the art should further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0186] The above embodiments are merely illustrative of the principles and effects of the present application and are not intended to limit the present application. Anyone familiar with the technology may modify or change the above embodiments without violating the spirit and scope of the present application. Therefore, all equivalent modifications or changes made by a person of ordinary skill in the art without departing from the spirit and technical ideas disclosed in the present application shall still be covered by the claims of the present application.

Claims

1. A vehicle camera control method, characterized in that: include: Acquire the power state and door state of the target vehicle, wherein the power state includes a power-on state and a power-off state, and the door state includes an unlocked state and a locked state; When the target vehicle is in a power-off state, determining whether the door state is a locked state, and if so, activating a leftover object detection function of a cabin camera of the target vehicle; When the target vehicle is in a powered-on state, obtaining the speed of the target vehicle; Determine whether the speed of the target vehicle reaches or exceeds a preset speed threshold, and if so, determine that the target vehicle is in a high-speed driving state, and obtain the duration of the high-speed driving of the target vehicle; If the duration of the target vehicle's high-speed driving exceeds a preset first time threshold, the driver monitoring function of the cockpit camera is activated.

2. The vehicle-mounted camera control method according to claim 1, characterized in that: The step of starting the leftover object detection function of the cabin camera of the target vehicle includes: Regularly acquiring an interior image of the target vehicle through the cockpit camera; The in-vehicle image is processed based on an image algorithm to determine whether there is a left-behind object in the target vehicle; if so, the in-vehicle image and an alarm message are sent to the user; if not, only the in-vehicle image is sent to the user.

3. The vehicle-mounted camera control method according to claim 1, characterized in that: The driver monitoring function of the cockpit camera includes one or more combinations of fatigue status monitoring, dangerous driving behavior monitoring and distracted driving monitoring.

4. The vehicle-mounted camera control method according to claim 3, characterized in that: The fatigue status monitoring includes: Acquiring the driver's facial image in real time through the cockpit camera; Based on the facial image of the driver, obtaining the eyelid closure degree and mouth opening degree of the driver; Set the eyelid closure fatigue judgment condition and yawn fatigue judgment condition respectively; Within a preset time, the number of times the driver's eyelid closure degree meets the eyelid closure fatigue judgment condition and the number of times the mouth opening degree meets the yawn fatigue judgment condition are counted; Based on the statistical results, it is determined whether the driver is in a fatigue state. If so, the fatigue driving warning mechanism is activated.

5. The vehicle-mounted camera control method according to claim 3, characterized in that: It also includes monitoring the driver's fatigue status using different sensitivity modes: Obtaining road condition information of the target vehicle; Based on the road condition information, dynamically adjust the frame rate of the cockpit camera capturing the driver's image; Based on the frame rate of the driver's image captured by the cockpit camera, different sensitivity modes are selected to monitor the driver's fatigue state; when the frame rate of the driver's image captured by the cockpit camera is lower than a preset frame rate threshold, the high sensitivity mode is adopted to monitor the driver's fatigue state; when the frame rate of the driver's image captured by the cockpit camera is higher than the preset frame rate threshold, the low sensitivity mode is adopted to monitor the driver's fatigue state.

6. The vehicle-mounted camera control method according to claim 5, characterized in that: The determining whether the driver is in a fatigue state based on the statistical results includes: In the low sensitivity mode, when the number of times that the counted eyelid closure degree satisfies the eyelid closure fatigue judgment condition reaches a first threshold, it is determined that the driver is in a fatigue state; In the high sensitivity mode, when the number of times that the counted eyelid closure degree satisfies the eyelid closure fatigue judgment condition reaches a second threshold, it is determined that the driver is in a fatigue state; wherein the first threshold is greater than the second threshold; In the low sensitivity mode, when the number of times that the mouth opening and closing degrees that meet the yawn fatigue judgment condition are counted reaches a third threshold, it is determined that the driver is in a fatigue state; In the high-sensitivity mode, when the number of times the mouth opening and closing degrees that meet the yawn fatigue judgment condition are counted reaches the fourth threshold, it is determined that the driver is in a fatigue state; wherein the third threshold is greater than the fourth threshold.

7. The vehicle-mounted camera control method according to claim 5, characterized in that: The dynamically adjusting the frame rate of the cockpit camera collecting the driver image based on the road condition information includes: When the target vehicle is in a complex road condition, reducing the frame rate of the cockpit camera collecting the driver's image; When the target vehicle is in a simple road condition, the frame rate of the cockpit camera for collecting the driver's image is increased.

8. The vehicle-mounted camera control method according to claim 3, characterized in that: The dangerous driving behavior monitoring includes: Acquiring a head image of the driver through the cockpit camera; Based on the driver's head image, obtaining the driver's face target frame and the driving interference object target frame; Based on the spatial intersection of the face target frame and the driving-interfering object target frame and the intersection duration, determining whether the driver is engaging in dangerous driving behavior; If so, the dangerous driving warning mechanism is activated.

9. The vehicle-mounted camera control method according to claim 3, characterized in that: The distracted driving monitoring includes: Acquiring a head image of the driver through the cockpit camera; Calculating the driver's head posture angle based on the driver's head image; Based on the head posture angle and the head posture angle duration, it is determined whether the driver is distracted driving; if so, a distracted driving warning mechanism is activated.

10. A vehicle-mounted camera control device, characterized in that: include: A first acquisition module, used to acquire a power state and a door state of a target vehicle, wherein the power state includes a power-on state and a power-off state, and the door state includes an unlocked state and a locked state; A first starting module, used for determining whether the door state is a locked state when the target vehicle is in a powered-off state, and if so, starting a leftover object detection function of a cabin camera of the target vehicle; A second acquisition module is used to acquire the speed of the target vehicle when the target vehicle is in a powered-on state; A high-speed judgment module is used to judge whether the speed of the target vehicle reaches or exceeds a preset speed threshold, and if so, it is determined that the target vehicle is in a high-speed driving state; A third acquisition module is used to acquire the duration of the high-speed driving of the target vehicle; The second starting module is used to start the driver monitoring function of the cockpit camera if the duration of the high-speed driving of the target vehicle exceeds a preset first time threshold.

11. A vehicle-mounted visual perception system, characterized in that: include: A visual sensor unit, including a forward-looking camera and a cabin camera; The front-view camera is used to collect forward driving images; The cockpit camera is used to collect images in the vehicle cabin, wherein the cockpit camera is used to monitor the driver's status when the vehicle is powered on and driving; and is used to detect objects left in the vehicle cabin when the vehicle is powered off and locked; A controller unit, which is in communication with the visual sensor unit and includes a SOC module and an MCU module, wherein the MCU module is used to control the SOC module to power on, receive vehicle data parameters and transmit the vehicle data parameters to the SOC module, and perform safety monitoring; The SOC module includes an image acquisition module, an image algorithm module, a control algorithm module and a data transmission module; wherein, The image acquisition module is used to control the camera of the visual sensor unit to acquire images; The image algorithm module is used to identify and detect the collected images; The control algorithm module is used to adjust the algorithm function according to the vehicle data parameters and image recognition results; The data transmission module is used to transmit the vehicle cabin image taken by the cabin camera and the result of detecting the remains in the cabin to an external device when the vehicle is electrically locked.

12. An electronic device, characterized in that: The electronic device comprises: Processor and memory; Wherein, the memory is used to store computer programs; The processor is used to execute the computer program stored in the memory so that the electronic device executes the vehicle camera control method according to any one of claims 1 to 9.

13. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by an electronic device, the vehicle camera control method described in any one of claims 1 to 9 is implemented.

Images