Intelligent visual cabin monitoring method based on multiple cameras and vehicle

Abstract

The invention discloses an intelligent visual cabin monitoring method based on multiple cameras and a vehicle, and belongs to the field of automobile intelligent cabins, and the method comprises the steps: obtaining the face information and the head posture of a driver through an IR camera, executing a driver state monitoring strategy according to the face information and the head posture of the driver, and obtaining a judgment result and alarm information, the judgment result and the alarm information are sent to a corresponding module of the whole vehicle; the condition in a cockpit is obtained through the wide-angle camera, a cockpit monitoring strategy is executed according to the condition in the cockpit to obtain the expression of a passenger or the state of an object, and the expression of the passenger or the state of the object is sent to a corresponding module of the whole vehicle. According to the invention, a cabin monitoring technology using a plurality of cameras as sensors is adopted, driver detection, passenger detection and cabin detection can be carried out, behaviors such as distraction and fatigue of a driver which are not beneficial to driving safety can be monitored, whether children, pets, wallets and the like are left in the cabin is checked, and financial loss of the driver is avoided.

Description

Technical Field

[0001] The invention discloses a multi-camera-based intelligent visual cockpit monitoring method and a vehicle, belonging to the field of automobile intelligent cockpits. Background Art

[0002] As cars have entered the new energy era from traditional energy, car cockpits have also experienced a transformation from mechanical cockpits to electronic cockpits, and then to today's smart cockpits. In the era of smart cockpits, cars are defined as the third mobile space. In addition to entertainment and interactive functions, health needs in the car space are increasingly valued by OEMs and consumers. The connotation of healthy smart cockpits is becoming more and more rich, including safety and health monitoring functions, nursing and health care functions, etc. It can be said that the first half of the car cockpit is intelligence, and the second half is safety and health.

[0003] At present, the cockpit systems of some cars have begun to monitor the driver's physiological state, but monitoring the physiological state alone cannot fully identify the driver's current driving state. Summary of the invention

[0004] In view of the defects of the prior art, the present invention proposes an intelligent visual cockpit monitoring method and vehicle based on multiple cameras. By adopting the in-cabin monitoring technology using multiple cameras as sensors, the status of the driver and passengers and what is left in the cockpit are monitored to improve the safety and comfort of the car and enhance the driving experience.

[0005] The technical solution of the present invention is as follows:

[0006] According to a first aspect of an embodiment of the present invention, there is provided a multi-camera-based intelligent visual cockpit monitoring method, comprising:

[0007] Acquire the driver's facial information and head posture through the IR camera, execute the driver status monitoring strategy according to the driver's facial information and head posture to obtain the determination result and alarm information, and send the determination result and alarm information to the corresponding module of the whole vehicle;

[0008] The situation in the cockpit is acquired through the wide-angle camera, and a cockpit monitoring strategy is executed according to the situation in the cockpit to obtain the expression of the occupant or the state of the object, and the expression of the occupant or the state of the object is sent to the corresponding module of the whole vehicle.

[0009] Preferably, the driver status monitoring strategy includes:

[0010] Obtain a facial feature image of the current driver, and input the facial feature image of the current driver into a Resnet-50-v2 algorithm model to obtain feature points of the current driver's face;

[0011] The LSTM algorithm performs model recognition based on feature points to obtain the driver status, and obtains a determination result and alarm information based on the driver status.

[0012] Preferably, the cockpit monitoring strategy includes:

[0013] Obtain a facial expression image of the current driver in the cockpit, and input the facial expression image of the current driver into a Resnet-50-v2 algorithm model to obtain feature points of the current driver's face;

[0014] The facial feature points of the current driver are used through the starGAN model to obtain the driver's expression.

[0015] Preferably, the cockpit monitoring strategy includes:

[0016] The current state of objects in the cockpit is obtained, and the current state of objects in the cockpit is preprocessed and input into the optimized pruning network to extract deep features. At the same time, the CoogLeNet network is trained using the pruning algorithm to extract deep features. Finally, the optimized pruned CoogLeNet network is combined with a classifier to realize different object recognition to obtain the object state.

[0017] Preferably, the driver state includes at least a distracted state or a fatigued state.

[0018] Preferably, the state of the object at least includes: a relic state or a child's forgotten state.

[0019] According to a second aspect of an embodiment of the present invention, there is provided an intelligent visual cockpit monitoring device based on multiple cameras, comprising:

[0020] A state monitoring module, used to obtain the driver's facial information and head posture through the IR camera, execute the driver's state monitoring strategy according to the driver's facial information and head posture to obtain a determination result and an alarm message, and send the determination result and the alarm message to a corresponding module of the vehicle;

[0021] The cockpit monitoring module is used to obtain the situation in the cockpit through the wide-angle camera, execute the cockpit monitoring strategy according to the situation in the cockpit to obtain the expression of the occupant or the state of the object, and send the expression of the occupant or the state of the object to the corresponding module of the whole vehicle.

[0022] Preferably, the status monitoring module is used to:

[0023] Obtain a facial feature image of the current driver, and input the facial feature image of the current driver into a Resnet-50-v2 algorithm model to obtain feature points of the current driver's face;

[0024] The LSTM algorithm performs model recognition based on feature points to obtain the driver status, and obtains a determination result and alarm information based on the driver status.

[0025] According to a third aspect of an embodiment of the present invention, there is provided a vehicle, comprising:

[0026] one or more processors;

[0027] a memory for storing the one or more processor-executable instructions;

[0028] Wherein, the one or more processors are configured to:

[0029] Execute the method described in the first aspect of the embodiment of the present invention.

[0030] According to a fourth aspect of an embodiment of the present invention, a non-temporary computer-readable storage medium is provided. When instructions in the storage medium are executed by a processor of a terminal, the terminal is enabled to execute the method described in the first aspect of the embodiment of the present invention.

[0031] According to a fifth aspect of an embodiment of the present invention, an application product is provided. When the application product is running on a terminal, the terminal executes the method described in the first aspect of the embodiment of the present invention.

[0032] The beneficial effects of the present invention are:

[0033] The present invention provides a multi-camera-based intelligent visual cockpit monitoring method and a vehicle, which adopts a cabin monitoring technology using multiple cameras as sensors, and can perform driver detection, passenger detection, and cabin detection, and can monitor driver distraction, fatigue, and other behaviors that are not conducive to driving safety; it can also realize cabin monitoring to check whether there are children, pets, wallets, etc. left in the cabin, so as to avoid financial losses of the driver; it can also customize the in-car environment, and can control the in-car atmosphere lights, music, and temperature according to the status of the passengers, so as to improve the comfortable travel environment; it can also use facial emotion recognition technology to allow the IMS system to automatically save happy moments in the journey. When the system detects a happy and laughing facial expression, the in-cabin camera can save instant photos, so that passengers can keep good memories.

[0034] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 The present invention is a schematic diagram of electrical connections of an intelligent visual cockpit monitoring system based on multiple cameras according to an exemplary embodiment.

[0036] Figure 2The present invention is a flowchart of a method for intelligent visual cockpit monitoring based on multiple cameras according to an exemplary embodiment.

[0037] Figure 3 It is a schematic diagram of the model structure of the starGAN model based on which the expression recognition function in the intelligent visual cockpit monitoring method based on multiple cameras is based, according to an exemplary embodiment.

[0038] Figure 4 It is a schematic diagram of a deep neural network structure based on which an object recognition function in a multi-camera based intelligent visual cockpit monitoring method is based, according to an exemplary embodiment.

[0039] Figure 5 The present invention is a flowchart of an object recognition function in a multi-camera based intelligent visual cockpit monitoring method according to an exemplary embodiment.

[0040] Figure 6 The present invention is a schematic block diagram of a structure of an intelligent visual cockpit monitoring device based on multiple cameras according to an exemplary embodiment.

[0041] Figure 7 The present invention is a schematic block diagram of a vehicle structure according to an exemplary embodiment. DETAILED DESCRIPTION

[0042] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0043] In the description of the present invention, it should be noted that the terms "center", "up", "down", "left", "right", "vertical", "horizontal", "inside", "outside", etc. indicate directions or positional relationships based on the directions or positional relationships shown in the accompanying drawings. They are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific direction, be constructed and operated in a specific direction. Therefore, they cannot be understood as limitations on the present invention.

[0044] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0045] The embodiment of the present invention provides a multi-camera based intelligent visual cockpit monitoring method implemented by a multi-camera based intelligent visual cockpit monitoring system, such as Figure 1 As shown, it includes an IR camera and a wide-angle camera electrically connected to the ECU controller respectively, and the vehicle is connected to the CANbus, ETH module, MCU Debugging and power module, wherein:

[0046] The IR camera uses one ordinary IR camera (DMC) for driver status monitoring and in-cabin face recognition, and the wide-angle camera uses one wide-angle camera (OMC) for intelligent cockpit monitoring.

[0047] The power module includes KL30, KL15 and GND. According to the design, the power management design of the intelligent cockpit monitoring system is carried out to meet the system's vehicle power supply, sleep, wake-up, CAN diagnosis and other strategic requirements. The CAN communication module implements signal interaction with the vehicle's CAN bus based on the selected CAN transceiver chip and MCU, and then feeds back the face recognition monitoring results, driver status monitoring results, expression recognition results, and whether there are children, pets, or specific items left in the cockpit in the intelligent cockpit monitoring system to the vehicle's CAN bus, and implements human-computer interaction functions such as corresponding warnings, provision, and playing corresponding music.

[0048] The ETH module performs overall data transmission, local software upgrade, OTA and other data record transmission of the intelligent monitoring system according to the selected ETH IC. The MCU Debugging module performs overall communication, diagnosis and fault handling of the system according to the design.

[0049] Embodiment 1: Figure 2 A method for intelligent visual cockpit monitoring based on multiple cameras is shown according to an exemplary embodiment, comprising:

[0050] Step 101, obtain the driver's facial information and head posture through the IR camera, execute the driver status monitoring strategy according to the driver's facial information and head posture to obtain the judgment result and alarm information, and send the judgment result and alarm information to the corresponding module of the whole vehicle. The specific content is as follows:

[0051] The driver's facial information and head posture are obtained by the IR camera, and a driver status monitoring strategy is implemented based on the driver's facial information and head posture. The specific steps of the driver status monitoring strategy are as follows:

[0052] Obtain the facial feature image of the current driver, and input the facial feature image of the current driver into the Resnet-50-v2 algorithm model to obtain the feature points of the current driver's face;

[0053] The LSTM algorithm performs model recognition based on feature points to obtain the driver status, and obtains a determination result and alarm information based on the driver status.

[0054] The judgment result and alarm information are transmitted to the vehicle CAN bus through the MCU. The corresponding module of the vehicle uses auditory + visual methods to remind the driver that he is currently distracted or fatigued after receiving the signal on the CAN bus, and needs to shift his sight back to the road or stop to rest.

[0055] Step 102, obtaining the situation in the cockpit through the wide-angle camera, executing the cockpit monitoring strategy according to the situation in the cockpit to obtain the occupant's expression or the state of the object, and sending the occupant's expression or the state of the object to the corresponding module of the vehicle, the specific contents are as follows:

[0056] The cockpit situation is acquired through the wide-angle camera, and the cockpit monitoring strategy is executed according to the cockpit situation. The specific steps of the cockpit monitoring strategy are as follows:

[0057] Obtain the facial expression image of the current driver in the cockpit, and input the facial expression image of the current driver into the Resnet-50-v2 algorithm model to obtain the feature points of the current driver's face;

[0058] The facial feature points of the current driver are used through the starGAN model to obtain the passenger's expression.

[0059] The starGAN model analyzes and compares the feature points extracted by the feature network containing 5 convolutional layers, thereby achieving the purpose of facial expression recognition and age recognition. The recognition functions include but are not limited to the number of people, age groups, positions of people, expressions of people, monitoring whether children, pets, valuables are left in the cockpit, and identifying specific items. Figure 3 As shown in the figure, it includes a generator and a judger. The generator learns the parameters of the feature points to generate a 2D image that can be recognized by the algorithm. In the design of the neural network, the generator draws on the residual block structure used by CycleGAN; the judger uses the PatchGANs structure with LeakyPeLU as the activation function to ensure image quality, as shown in the figure. Figure 4 shown.

[0060] The passenger's expression is transmitted to the vehicle's CAN bus through the MCU. The corresponding module of the vehicle obtains the result on the CAN bus. When the passenger's expression is "sad", soothing music is played and the interior ambient light is adjusted to the driver's favorite color, so that people and the car can empathize with each other.

[0061] The above cockpit monitoring strategy also includes:

[0062] The current state of objects in the cockpit is obtained, and the current state of objects in the cockpit is preprocessed and input into the optimized pruning network to extract deep features. The Inception module is used to deepen the network learning depth. At the same time, the pruning algorithm is used to train the CoogLeNet network to extract deep features. Finally, the optimized pruned CoogLeNet network is combined with the classifier to realize different object recognition to obtain the object state, such as Figure 5 As shown, the object status includes at least: the state of relics or the state of children's forgotten. Relics include pets (cats, dogs), mobile phones, books, wallets, computers, etc. At the same time, it provides users or fleet managers with real-time cockpit video during vehicle driving to improve vehicle safety.

[0063] Embodiment 2: Figure 6 An intelligent visual cockpit monitoring device based on multiple cameras is shown according to an exemplary embodiment, and is characterized by comprising:

[0064] The state monitoring module 210 is used to obtain the driver's facial information and head posture through the IR camera, execute the driver's state monitoring strategy according to the driver's facial information and head posture to obtain the determination result and alarm information, and send the determination result and alarm information to the corresponding module of the whole vehicle;

[0065] The cockpit monitoring module 220 is used to obtain the situation in the cockpit through the wide-angle camera, execute the cockpit monitoring strategy according to the situation in the cockpit to obtain the occupant's expression or the state of the object, and send the occupant's expression or the state of the object to the corresponding module of the whole vehicle.

[0066] The status monitoring module 210 is used to:

[0067] Obtain a facial feature image of the current driver, and input the facial feature image of the current driver into a Resnet-50-v2 algorithm model to obtain feature points of the current driver's face;

[0068] The LSTM algorithm performs model recognition based on feature points to obtain the driver status, and obtains a determination result and alarm information based on the driver status.

[0069] This application uses in-cabin monitoring technology that uses multiple cameras as sensors. It can perform driver detection, passenger detection, and in-cabin detection, and can monitor driver distraction, fatigue, and other behaviors that are not conducive to driving safety; it can also implement in-cabin monitoring to check whether there are children, pets, wallets, etc. left in the cabin to avoid financial losses to the driver; it can also customize the in-car environment, and can control the in-car ambient lighting, music, and temperature according to the status of the passengers to enhance the comfort of the travel environment; it can also use facial emotion recognition technology to allow the IMS system to automatically save happy moments during the journey. When the system detects facial expressions of happiness and laughter, the in-cabin camera can save instant photos to allow passengers to retain good memories.

[0070] Embodiment three: Figure 6 300 is a block diagram of a vehicle 300 provided in an embodiment of the present application. For example, the vehicle 300 may be a hybrid vehicle, a non-hybrid vehicle, an electric vehicle, a fuel cell vehicle, or other types of vehicles. The vehicle 300 may be an autonomous vehicle, a semi-autonomous vehicle, or a non-autonomous vehicle. The vehicle 300 may also be equipped with a brake-by-wire system.

[0071] Reference Figure 6 , the vehicle 300 may include various subsystems, for example, an infotainment system 310, a perception system 320, a decision control system 330, a drive system 340, and a computing platform 350. The vehicle 300 may also include more or fewer subsystems, and each subsystem may include multiple components. In addition, each subsystem and each component of the vehicle 300 may be interconnected by wire or wireless means.

[0072] In some embodiments, the infotainment system 310 may include a communication system, an entertainment system, and a navigation system, among others.

[0073] The perception system 320 may include several sensors for sensing information about the environment around the vehicle 300. For example, the perception system 320 may include a global positioning system (the global positioning system may be a GPS system, or a Beidou system or other positioning systems), an inertial measurement unit (IMU), a laser radar, a millimeter wave radar, an ultrasonic radar, and a camera.

[0074] The decision control system 330 may include a computing system, a vehicle controller, a steering system, a throttle, and a braking system.

[0075] The drive system 340 may include components that provide power movement for the vehicle 300. In one embodiment, the drive system 340 may include an engine, an energy source, a transmission system, and wheels. The engine may be one or a combination of an internal combustion engine, an electric motor, and an air compression engine. The engine is capable of converting energy provided by the energy source into mechanical energy.

[0076] Some or all functions of the vehicle 300 are controlled by a computing platform 350. The computing platform 350 may include at least one processor 351 and a memory 352, and the processor 351 may execute instructions 353 stored in the memory 352.

[0077] The processor 351 may be any conventional processor, such as a commercially available CPU. The processor may also include a graphics processor (Graphic Process Unit, GPU), a field programmable gate array (Field Programmable Gate Array, FPGA), a system on chip (System on Chip, SOC), an application specific integrated circuit (Application Specific Integrated Circuit, ASIC) or a combination thereof.

[0078] The memory 352 may be implemented by any type of volatile or nonvolatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0079] In addition to instructions 353 , memory 352 may also store data, such as road maps, route information, vehicle location, direction, speed, etc. The data stored in memory 352 may be used by computing platform 350 .

[0080] In the embodiment of the present disclosure, the processor 351 can execute instruction 353 to complete all or part of the steps of the above-mentioned multi-camera-based intelligent visual cockpit monitoring method.

[0081] Embodiment 4: In an exemplary embodiment, a computer-readable storage medium is also provided, on which a computer program is stored. When the program is executed by a processor, an intelligent visual cockpit monitoring method based on multiple cameras as provided in all the inventive embodiments of the present application is implemented.

[0082] Any combination of one or more computer-readable media may be used. A computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, device, or device.

[0083] Computer-readable signal media may include a data signal propagated in baseband or as part of a carrier wave, which carries a computer-readable program code. Such propagated data signals may take a variety of forms, including, but not limited to, electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0084] Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0085] Computer program code for performing the operations of the present invention may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0086] Embodiment 5: In an exemplary embodiment, an application product is also provided, including one or more instructions, which can be executed by the processor 351 of the above-mentioned device to complete the above-mentioned multi-camera-based intelligent visual cockpit monitoring method.

[0087] Although the embodiments of the present invention have been disclosed above, they are not limited to the applications listed in the specification and the embodiments. It can be fully applied to various fields suitable for the present invention. For those familiar with the art, additional modifications can be easily realized. Therefore, without departing from the general concept defined by the claims and equivalent scope, the present invention is not limited to the specific details and the illustrations shown and described here.

Claims

1. An intelligent visual cockpit monitoring method based on multiple cameras, applied to an intelligent cockpit monitoring system, comprising an IR camera and a wide-angle camera electrically connected to an ECU controller respectively, characterized in that: include: Acquire the driver's facial information and head posture through the IR camera, execute the driver status monitoring strategy according to the driver's facial information and head posture to obtain the determination result and alarm information, and send the determination result and alarm information to the corresponding module of the whole vehicle; The situation in the cockpit is acquired through the wide-angle camera, and a cockpit monitoring strategy is executed according to the situation in the cockpit to obtain the expression of the occupant or the state of the object, and the expression of the occupant or the state of the object is sent to the corresponding module of the whole vehicle.

2. The multi-camera based intelligent visual cockpit monitoring method according to claim 1, characterized in that: The driver status monitoring strategy includes: Obtain a facial feature image of the current driver, and input the facial feature image of the current driver into a Resnet-50-v2 algorithm model to obtain feature points of the current driver's face; The LSTM algorithm performs model recognition based on feature points to obtain the driver status, and obtains a determination result and alarm information based on the driver status.

3. The multi-camera based intelligent visual cockpit monitoring method according to claim 1, characterized in that: The cockpit monitoring strategy includes: Obtain a facial expression image of the current driver in the cockpit, and input the facial expression image of the current driver into a Resnet-50-v2 algorithm model to obtain feature points of the current driver's face; The facial feature points of the current driver are used through the starGAN model to obtain the driver's expression.

4. The multi-camera based intelligent visual cockpit monitoring method according to claim 1, characterized in that: The cockpit monitoring strategy includes: The current state of objects in the cockpit is obtained, and the current state of objects in the cockpit is preprocessed and input into the optimized pruning network to extract deep features. At the same time, the CoogLeNet network is trained using the pruning algorithm to extract deep features. Finally, the optimized pruned CoogLeNet network is combined with a classifier to realize different object recognition to obtain the object state.

5. The multi-camera based intelligent visual cockpit monitoring method according to claim 2, characterized in that: The driver state includes at least: a distracted state or a fatigue state.

6. The multi-camera based intelligent visual cockpit monitoring method according to claim 4, characterized in that: The object state at least includes: a relic state or a child forgotten state.

7. An intelligent visual cockpit monitoring device based on multiple cameras, characterized in that: include: A state monitoring module, used to obtain the driver's facial information and head posture through the IR camera, execute the driver's state monitoring strategy according to the driver's facial information and head posture to obtain a determination result and an alarm message, and send the determination result and the alarm message to a corresponding module of the vehicle; The cockpit monitoring module is used to obtain the situation in the cockpit through the wide-angle camera, execute the cockpit monitoring strategy according to the situation in the cockpit to obtain the expression of the occupant or the state of the object, and send the expression of the occupant or the state of the object to the corresponding module of the whole vehicle.

8. The multi-camera based intelligent visual cockpit monitoring device according to claim 7, characterized in that: The state monitoring module is used to: Obtain a facial feature image of the current driver, and input the facial feature image of the current driver into a Resnet-50-v2 algorithm model to obtain feature points of the current driver's face; The LSTM algorithm performs model recognition based on feature points to obtain the driver status, and obtains a determination result and alarm information based on the driver status.

9. A vehicle, characterized in that: include: processor; a memory for storing processor-executable instructions; Wherein, the processor is configured to execute the steps of the multi-camera based intelligent visual cockpit monitoring method as described in any one of claims 1 to 6.

10. A computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the program instructions are executed by a processor, the steps of the multi-camera based intelligent visual cockpit monitoring method described in any one of claims 1 to 6 are implemented.

Images