Safe driving assistance method and device based on student vision and posture monitoring
By combining multimodal fusion models of high-definition cameras, attitude sensors and infrared sensors, we can monitor students' visual and hand movements in real time, solving the problems of perspective limitation and misjudgment in the existing technology, and achieving accurate assessment of students' driving behavior and personalized feedback significantly reduces the risk of traffic accidents and improves the safety and effectiveness of driving training.
Patent Information
- Application Number
- CN202411229798.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-03
- Publication Date
- 2024-11-29
AI Technical Summary
The existing driving training system has problems such as the limitation of perspective angle and insufficient accuracy of image processing algorithms in monitoring students' vision and hand movements, resulting in misjudgment and unreliability, and cannot effectively reduce the risk of traffic accidents.
Using high-definition cameras, attitude sensors and infrared sensors combined with image processing technology, through multimodal fusion model and deep learning algorithm, students' vision, posture and hand movements are monitored in real time, and situational perception algorithms and personalized feedback mechanisms are combined to provide customized Reminders and training suggestions.
It realizes comprehensive and accurate monitoring of students' vision and posture, reduces the risk of traffic accidents, improves the effectiveness and safety of driving training, and provides a personalized and efficient driving training experience.
Smart Images

Figure CN119037442A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of automobile driving training safety technology, and in particular to a safe driving assistance method and device based on trainee vision and posture monitoring. Background Art
[0002] During driving training, novice trainees often make some common mistakes due to lack of practical driving experience and lack of concentration. For example, when making left and right turns, they often forget to check the rearview mirror to confirm safety; while driving, they may look down to check the gear position to engage the gear; even while driving, they may keep their hands on the gear position instead of on the steering wheel. These improper behaviors not only violate the basic principles of driving, but also greatly increase the risk of traffic accidents.
[0003] The relevant technical solutions currently available on the market mainly rely on cameras for posture monitoring. These camera-based visual monitoring systems usually install high-definition cameras inside the vehicle to capture the trainee's facial images in real time. Through image processing technology, the system can identify the direction of the trainee's gaze to determine whether they are looking at the rearview mirror or the road ahead. However, this type of system mainly focuses on visual monitoring and has relatively limited monitoring capabilities for hand movements. In addition, due to factors such as light changes or reflections from glasses, the system sometimes makes misjudgments, resulting in the inability to accurately identify the trainee's gaze direction.
[0004] The shortcomings of the existing technology are mainly reflected in two aspects: one is the limited viewing angle of the camera, and the other is the insufficient accuracy of the image processing algorithm. The limited viewing angle of the camera may cause some key actions to fail to be captured, while the insufficient accuracy of the image processing algorithm may lead to misjudgment, thus affecting the reliability of the system.
[0005] Therefore, it is particularly important to develop an intelligent driving training system that can monitor and correct these unsafe driving behaviors in real time. Such a system should not only be able to accurately identify the trainee's line of sight, but also have the ability to monitor hand movements to ensure that the trainee always maintains the correct driving posture and concentration during driving. In this way, the risk of traffic accidents can be significantly reduced and the effectiveness and safety of driving training can be improved. Summary of the invention
[0006] In view of the shortcomings of the current related existing technologies, the present invention provides a safe driving assistance method, device and equipment based on student vision and posture monitoring. The present invention can accurately and comprehensively monitor the student's vision and posture, strengthen the monitoring of hand movements, and thus provide a more personalized and efficient driving training experience. Ensure that the student continues to maintain the correct driving posture and concentrate during driving. Through this method, the risk of traffic accidents can be significantly reduced, and the effect and safety of driving training can be improved.
[0007] To achieve the above object, the first aspect of the present invention provides a safe driving assistance method based on student vision and posture monitoring, which at least includes the following steps:
[0008] S1. Monitor the students' vision and posture, and collect data information about the students' vision and posture;
[0009] S2. Perform data fusion processing on the collected student vision and posture data information to comprehensively judge the student's driving behavior;
[0010] S3, combining the vehicle driving environment and the student's driving task, using the situational awareness algorithm to determine whether the student's driving behavior is appropriate;
[0011] S4. When the system detects unsafe driving behavior of the trainee, the reminder mechanism is triggered.
[0012] Furthermore, in step S1, the student's vision and posture are monitored, including visual monitoring, posture monitoring and hand movement monitoring of the student; data information of the student's vision and posture is obtained, and the data information at least includes facial images, head posture and hand positions.
[0013] Furthermore, in visual monitoring of trainees, the implementation method is as follows: using the vehicle-mounted high-definition camera to capture the trainee's facial image in real time;
[0014] Analyze the captured facial images through image processing technology to identify the trainee's gaze direction;
[0015] By comparing the direction of sight with the driving status of the vehicle, it is determined whether the trainee is looking at the rearview mirror at a critical moment, or looking down at the block when performing a specific operation;
[0016] The algorithm is able to identify key features, including at least the position of the eyes and the gaze point of the pupil, so as to accurately determine whether the student is checking the rearview mirror when turning or looking down at the gear when shifting into gear.
[0017] Furthermore, in the posture monitoring of the trainees, the implementation method is as follows: using the posture sensor to collect the head posture of the trainees;
[0018] The gyroscope and accelerometer installed inside the vehicle monitor the changes in the trainee's head posture; the angle and acceleration of the trainee's head can be measured in real time;
[0019] At the same time, image recognition technology is used to further analyze the students’ head posture;
[0020] By comparing the normal driving posture with the current posture, it is determined whether the student has deviated from the normal field of vision for a long time. By setting thresholds or reference models, behaviors that deviate from the normal field of vision for a long time can be identified and warned.
[0021] Furthermore, in monitoring the hand movements of trainees, the implementation method is: using infrared sensors and cameras combined with image processing to monitor the position of trainees’ hands;
[0022] By setting a reasonable monitoring cycle and judgment logic, the behavior of keeping hands on the gear for a long time can be identified and warned; by monitoring the changes in hand position in real time, the system can determine whether the trainee keeps his hands on the gear for a long time while driving, which may cause the trainee's control of the vehicle to be unstable.
[0023] Furthermore, in step S2, the facial image captured by the camera is preprocessed, feature extracted, and sight line recognized using an image processing module; the head posture is analyzed using a posture recognition module; and the hand movement detection module is used to monitor the hand position;
[0024] Combine image data and sensor data, and use the multimodal fusion model in deep learning, including the attention mechanism and feature fusion layer, to comprehensively judge the student's driving behavior;
[0025] Use LSTM long short-term memory network model or transformer neural network model to process time series data, analyze students' driving behavior patterns over a period of time, predict potential unsafe behaviors, and intervene in advance;
[0026] It also includes real-time feature extraction and optimization; using dynamic convolutional neural networks or adaptive feature extraction algorithms to dynamically adjust feature extraction parameters according to the student's real-time driving status to better capture key information;
[0027] Generate more diverse training data using generative adversarial networks to enhance the model’s generalization capabilities, especially when dealing with rare or extreme driving behaviors.
[0028] Furthermore, it also includes using decision-making judgment methods to comprehensively evaluate whether there is unsafe driving behavior; wherein the decision-making judgment at least includes situational awareness decision-making, personalized feedback mechanism and interactive learning, wherein
[0029] Situation-aware decision making
[0030] Combine the vehicle driving environment and the student's driving task, and use the situational awareness algorithm to determine whether the student's driving behavior is appropriate; establish a risk assessment model based on machine learning, comprehensively consider the student's driving behavior, vehicle status and external environment, assess the risk level of the current driving behavior, and adjust the intensity and frequency of reminders accordingly;
[0031] Personalized feedback mechanism
[0032] By collecting students' driving behavior data and building personalized user portraits, including driving habits, reaction speed and concentration, we provide each student with customized reminders and training suggestions. Based on the students' progress and feedback, we dynamically adjust the content and methods of reminders, such as changing the tone and speed of voice prompts or adding visual prompts, to improve the effectiveness of reminders.
[0033] Interactive Learning
[0034] The driving training process is designed in a gamified form, and trainees are encouraged to improve their driving behavior through mechanisms such as points and rewards. The system can give instant feedback based on the trainees' performance and integrate safe driving knowledge points and skills into the game. After the training, trainees are provided with a detailed driving behavior analysis report, including the number and type of unsafe behaviors and improvement suggestions, to encourage trainees to reflect on themselves and actively improve.
[0035] Furthermore, it also includes a reminder mechanism; when the system detects that the student has unsafe driving behavior, the reminder mechanism will be triggered immediately; when the student does not observe the left and right rearview mirrors when turning left or right, the driving training system will broadcast a voice message "Please observe the left / right / rearview mirror when turning left / right, and make sure it is safe before turning"; when the student looks down at the gear position when shifting gears, the driving training system will broadcast a voice message "Don't look down when shifting gears"; when it is detected that the student's hand is always on the gear position, the driving training system will broadcast a voice message "Don't keep your hand on the gear position all the time"; and the intelligent driving training system will provide voice reminders to ensure that the student can receive the reminder information in time and correct the wrong behavior; the reminder content will be customized according to the specific wrong behavior to enhance the pertinence and effectiveness of the reminder.
[0036] Furthermore, it also includes a continuous monitoring mechanism; using the on-board camera or head-mounted device to capture the student's facial image in real time, and using image processing technology to identify the direction of the student's line of sight, to determine whether he or she is looking at the rearview mirror when turning, or looking down at the gear when shifting gears; the system will continuously monitor the student's driving behavior and repeat reminders when necessary; at the same time, the system can also record the student's driving behavior data to provide a reference for subsequent driving training and evaluation.
[0037] A second aspect of the present invention provides a device, which is used to execute the above method.
[0038] The present invention adopts the above technical solution, which has at least the following beneficial effects:
[0039] 1. Comprehensive monitoring of students' vision and posture: By integrating multiple technical means such as high-definition cameras, posture sensors and hand tracking modules, comprehensive monitoring of students' visual focus, head posture and hand movements can be achieved to ensure the comprehensiveness and accuracy of the monitoring range.
[0040] 2. Strengthen hand movement monitoring: Pay special attention to the changes in the trainees' hand position, use hand tracking technology or sensors installed near the gear position to monitor in real time whether the trainees keep their hands on the gear position for a long time, and issue reminders in time.
[0041] 3. Diversified reminder methods: Use multiple reminder methods such as sound, vision and touch, and flexibly choose according to the actual situation of the students and the driving environment to ensure the effective communication of reminder information.
[0042] 4. Possess adaptive capabilities: By introducing artificial intelligence algorithms and machine learning technologies, the system can learn and identify students’ driving habits and progress, dynamically adjust monitoring and reminder strategies, and provide a more personalized and efficient driving training experience. BRIEF DESCRIPTION OF THE DRAWINGS
[0043] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0044] Figure 1 The present invention is a safe driving assistance method process based on student vision and posture monitoring Figure 1 ;
[0045] Figure 2 The present invention is a safe driving assistance method process based on student vision and posture monitoring Figure 2 . DETAILED DESCRIPTION
[0046] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Instead, they are merely examples of devices and methods consistent with some aspects of the present invention as detailed in the appended claims.
[0047] like Figure 1 and Figure 2 As shown, the first aspect of this embodiment provides a safe driving assistance method based on student vision and posture monitoring, which at least includes the following steps:
[0048] S1. Monitor the students' vision and posture, and collect data information about the students' vision and posture;
[0049] S2. Perform data fusion processing on the collected student vision and posture data information to comprehensively judge the student's driving behavior;
[0050] S3, combining the vehicle driving environment and the student's driving task, using the situational awareness algorithm to determine whether the student's driving behavior is appropriate;
[0051] S4. When the system detects unsafe driving behavior of the trainee, the reminder mechanism is triggered.
[0052] In step S1 of this embodiment, the student's vision and posture are monitored, including vision monitoring, posture monitoring and hand movement monitoring of the student; data information of the student's vision and posture is obtained, and the data information at least includes facial images, head posture and hand positions.
[0053] In this embodiment, it should be supplemented that the vehicle-mounted high-definition camera has the characteristics of high definition, low latency and wide-angle field of view to ensure that the facial image of the trainee can be clearly captured. In the visual monitoring of the trainee, the implementation method is: using the vehicle-mounted high-definition camera to capture the facial image of the trainee in real time;
[0054] Analyze the captured facial images through image processing technology (such as machine learning and deep learning algorithms) to identify the direction of the trainee's gaze;
[0055] By comparing the line of sight with the vehicle's driving status (such as turning, going straight, etc.), determine whether the trainee is looking at the rearview mirror at critical moments (such as when turning), or looking down at the gear when performing specific operations (such as shifting gears);
[0056] The algorithms are able to identify key features, including at least the position of the eyes and the point of gaze of the pupil, to accurately determine whether the student is looking in the mirror when turning, or looking down at the gear when engaging a gear. The algorithms are able to identify key features, including the position of the eyes and the point of gaze of the pupil, to accurately determine whether the student is looking in the mirror when turning, or looking down at the gear when engaging a gear.
[0057] In this embodiment, it should be supplemented that the posture sensor (such as a gyroscope or an accelerometer) is used to measure the angle, acceleration and other parameters of the trainee's head in real time. In the posture monitoring of the trainee, the implementation method is: using the posture sensor to collect the trainee's head posture;
[0058] The gyroscope and accelerometer installed inside the vehicle monitor the changes in the trainee's head posture; the angle and acceleration of the trainee's head can be measured in real time;
[0059] At the same time, image recognition technology (such as facial feature point detection) is used to further analyze the student's head posture;
[0060] By comparing the normal driving posture with the current posture, it is determined whether the student has deviated from the normal field of vision for a long time. By setting thresholds or reference models, behaviors that deviate from the normal field of vision for a long time can be identified and warned.
[0061] In this embodiment, it should be supplemented that, in monitoring the hand movements of the trainees, the implementation method is: using hand tracking technology (infrared sensor and camera combined with image processing) to monitor the hand position of the trainees;
[0062] By setting a reasonable monitoring cycle and judgment logic, the behavior of keeping hands on the gear for a long time can be identified and warned; by monitoring the changes in hand position in real time, the system can determine whether the trainee keeps his hands on the gear for a long time while driving, which may cause the trainee's control of the vehicle to be unstable.
[0063] In this embodiment, it should be supplemented that in step S2, the image processing module is used to pre-process, extract features and identify sight lines of the facial image captured by the camera; the posture recognition module is used to analyze the head posture; and the hand movement detection module is used to monitor the hand position;
[0064] Multimodal fusion model: Combines image data (from high-definition cameras) and sensor data (from posture sensors) and uses multimodal fusion models in deep learning (such as attention mechanisms and feature fusion layers) to comprehensively judge the student's driving behavior;
[0065] Use LSTM long short-term memory network model or transformer neural network model to process time series data, analyze students' driving behavior patterns over a period of time, predict potential unsafe behaviors, and intervene in advance;
[0066] It also includes real-time feature extraction and optimization;
[0067] Dynamic feature extraction: Dynamic CNN or adaptive feature extraction algorithm is used to dynamically adjust the feature extraction parameters according to the student’s real-time driving status to better capture key information;
[0068] Generative adversarial networks (GANs) are used to generate more diverse training data and enhance the generalization ability of the model, especially when dealing with rare or extreme driving behaviors.
[0069] As an implementation method, this embodiment also includes using a decision-making judgment method to comprehensively evaluate whether there is unsafe driving behavior; wherein the decision-making judgment includes at least situational awareness decision-making, personalized feedback mechanism and interactive learning, wherein
[0070] Situation-aware decision making
[0071] The situational awareness algorithm is used to determine whether the student's driving behavior is appropriate, based on the vehicle's driving environment (such as road type and traffic conditions) and the student's driving tasks (such as turning, changing lanes, and parking). For example, in a complex traffic environment, the requirements for rearview mirror observation may be higher.
[0072] Risk assessment model: Establish a risk assessment model based on machine learning, comprehensively consider the student's driving behavior, vehicle status and external environment, evaluate the risk level of the current driving behavior, and adjust the intensity and frequency of reminders accordingly;
[0073] Personalized feedback mechanism
[0074] User portrait construction: By collecting students' driving behavior data, we build personalized user portraits, including driving habits, reaction speed, and concentration, and provide customized reminders and training suggestions for each student.
[0075] Dynamic adjustment strategy: According to the students’ progress and feedback, dynamically adjust the content and method of reminders, such as changing the tone and speed of voice prompts or adding visual prompts (such as HUD display) to improve the effectiveness of reminders.
[0076] Interactive Learning
[0077] Gamified learning: The driving training process is designed in a gamified form, and trainees are encouraged to improve their driving behavior through mechanisms such as points and rewards. The system can give instant feedback based on the trainees' performance and integrate safe driving knowledge points and skills into the game. Self-reflection and improvement: After the training, trainees are provided with a detailed driving behavior analysis report, including the number and type of unsafe behaviors and improvement suggestions, to encourage trainees to reflect on themselves and actively improve.
[0078] As an implementation method, this embodiment also includes a reminder mechanism; when the system detects that the student has unsafe driving behavior, the reminder mechanism will be triggered immediately; when the student does not observe the left and right rearview mirrors when turning left or right, the driving training system will broadcast a voice message "Please observe the left / right / rearview mirror when turning left / right, and make sure it is safe before turning"; when the student looks down at the gear position when shifting gears, the driving training system will broadcast a voice message "You cannot look down when shifting gears"; when it is detected that the student's hand is always on the gear position, the driving training system will broadcast a voice message "Don't keep your hand on the gear position all the time"; and the intelligent driving training system will provide voice reminders to ensure that the student can receive the reminder information in time and correct the wrong behavior; the reminder content will be customized according to the specific wrong behavior to enhance the pertinence and effectiveness of the reminder.
[0079] As an implementation method, this embodiment also includes a continuous monitoring mechanism; using a vehicle-mounted camera or a head-mounted device to capture the student's facial image in real time, and using image processing technology to identify the direction of the student's line of sight, it is determined whether the student is looking at the rearview mirror when turning, or looking down at the gear when shifting gears; the system will continuously monitor the student's driving behavior and repeat reminders when necessary; at the same time, the system can also record the student's driving behavior data to provide a reference for subsequent driving training and evaluation.
[0080] The second aspect of the present invention provides a device for executing the above method. The device includes at least key components such as a camera, a hand tracking module and a data processing unit. These components can be integrated into a vehicle-mounted system. Regarding the training vehicles equipped with the above device, these vehicles communicate with the device through an internal network to achieve real-time transmission and processing of data. At the same time, the vehicle can also be equipped with auxiliary equipment such as display screens and speakers to better display reminder information and enhance user experience.
[0081] The present invention can accurately and comprehensively monitor the trainee's vision and posture, strengthen the monitoring of hand movements, and thus provide a more personalized and efficient driving training experience. It ensures that the trainee can continue to maintain the correct driving posture and concentrate during driving. Through this method, the risk of traffic accidents can be significantly reduced, and the effect and safety of driving training can be improved.
[0082] Although the embodiments of the present invention have been shown and described above, it will be appreciated that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, substitute and vary the above embodiments within the scope of the present invention.
Claims
1. A safe driving assistance method based on student vision and posture monitoring, Features: At least the following steps are included: S1. Monitor the students' vision and posture, and collect data information about the students' vision and posture; S2. Perform data fusion processing on the collected student vision and posture data information to comprehensively judge the student's driving behavior; S3, combining the vehicle driving environment and the student's driving task, using the situational awareness algorithm to determine whether the student's driving behavior is appropriate; S4. When the system detects unsafe driving behavior of the trainee, the reminder mechanism is triggered.
2. The method according to claim 1, Features: In step S1, the vision and posture of the trainee are monitored, including vision monitoring, posture monitoring and hand movement monitoring of the trainee; data information of the vision and posture of the trainee is obtained, and the data information at least includes facial image, head posture and hand position.
3. The method according to claim 2, Features: In the visual monitoring of trainees, this is achieved by: using the on-board high-definition camera to capture the trainee’s facial image in real time; Analyze the captured facial images through image processing technology to identify the trainee's gaze direction; By comparing the direction of sight with the driving status of the vehicle, it is determined whether the trainee is looking at the rearview mirror at a critical moment, or looking down at the block when performing a specific operation; The algorithm is able to identify key features, including at least the position of the eyes and the gaze point of the pupil, so as to accurately determine whether the student is checking the rearview mirror when turning or looking down at the gear when shifting into gear.
4. The method according to claim 3, Features: In the posture monitoring of trainees, the implementation method is as follows: using posture sensors to collect the head posture of trainees; The gyroscope and accelerometer installed inside the vehicle monitor the changes in the trainee's head posture; the angle and acceleration of the trainee's head can be measured in real time; At the same time, image recognition technology is used to further analyze the students’ head posture; By comparing the normal driving posture with the current posture, it is determined whether the student has deviated from the normal field of vision for a long time. By setting thresholds or reference models, behaviors that deviate from the normal field of vision for a long time can be identified and warned.
5. The method according to claim 4, Features: In monitoring the hand movements of trainees, the implementation method is: using infrared sensors and cameras combined with image processing to monitor the position of trainees’ hands; By setting a reasonable monitoring cycle and judgment logic, the behavior of keeping hands on the gear for a long time can be identified and warned; by monitoring the changes in hand position in real time, the system can determine whether the trainee keeps his hands on the gear for a long time while driving, which may cause the trainee's control of the vehicle to be unstable.
6. The method according to claim 5, Features: In step S2, the facial image captured by the camera is preprocessed, feature extracted, and sight line recognized using an image processing module; the head posture is analyzed using a posture recognition module; Monitor hand position using hand motion detection module; Combine image data and sensor data, and use the multimodal fusion model in deep learning, including the attention mechanism and feature fusion layer, to comprehensively judge the student's driving behavior; Use LSTM long short-term memory network model or transformer neural network model to process time series data, analyze students' driving behavior patterns over a period of time, predict potential unsafe behaviors, and intervene in advance; It also includes real-time feature extraction and optimization; using dynamic convolutional neural networks or adaptive feature extraction algorithms to dynamically adjust feature extraction parameters according to the student's real-time driving status to better capture key information; Generate more diverse training data using generative adversarial networks to enhance the model’s generalization capabilities, especially when dealing with rare or extreme driving behaviors.
7. The method according to claim 6, Features: It also includes using decision-making judgment methods to comprehensively evaluate whether there is unsafe driving behavior; the judgment of decision-making at least includes situational awareness decision-making, personalized feedback mechanism and interactive learning, among which Situation-aware decision making Combine the vehicle driving environment and the student's driving task, and use the situational awareness algorithm to determine whether the student's driving behavior is appropriate; establish a risk assessment model based on machine learning, comprehensively consider the student's driving behavior, vehicle status and external environment, assess the risk level of the current driving behavior, and adjust the intensity and frequency of reminders accordingly; Personalized feedback mechanism By collecting students' driving behavior data and building personalized user portraits, including driving habits, reaction speed and concentration, we provide each student with customized reminders and training suggestions. Based on the students' progress and feedback, we dynamically adjust the content and methods of reminders, such as changing the tone and speed of voice prompts or adding visual prompts, to improve the effectiveness of reminders. Interactive Learning The driving training process is designed in a gamified form, and trainees are encouraged to improve their driving behavior through mechanisms such as points and rewards. The system can give instant feedback based on the trainees' performance and integrate safe driving knowledge points and skills into the game. After the training, trainees are provided with a detailed driving behavior analysis report, including the number and type of unsafe behaviors and improvement suggestions, to encourage trainees to reflect on themselves and actively improve.
8. The method according to claim 7, Features: It also includes a reminder mechanism; when the system detects that the trainee has unsafe driving behavior, the reminder mechanism will be triggered immediately; when the trainee does not observe the left and right rearview mirrors when turning left or right, the driving training system will broadcast a voice message "Please observe the left / right / rearview mirror when turning left / right, and make sure it is safe before turning"; when the trainee lowers his head to look at the gear position when shifting gears, the driving training system will broadcast a voice message "Don't lower your head when shifting gears"; when it is detected that the trainee's hand is always on the gear position, the driving training system will broadcast a voice message "Don't keep your hand on the gear position"; and the intelligent driving training system will provide voice reminders to ensure that the trainee can receive the reminder information in time and correct the wrong behavior; The reminder content will be customized according to the specific erroneous behavior to enhance the pertinence and effectiveness of the reminder.
9. The method according to claim 8, Features: It also includes a continuous monitoring mechanism; using the on-board camera or head-mounted device to capture the student's facial image in real time, and using image processing technology to identify the direction of the student's line of sight to determine whether he or she is looking at the rearview mirror when turning, or looking down at the gear when shifting gears; the system will continuously monitor the student's driving behavior and repeat reminders when necessary; at the same time, the system can also record the student's driving behavior data to provide a reference for subsequent driving training and evaluation.
10. A device, Features: The device is used to execute the method according to any one of claims 1 to 9.
Citation Information
Cited By
Auxiliary driving method, system and equipment based on inertial navigation and machine vision and medium
CN119239632A
Bad driving behavior feedback intervention method and device based on driving behavior big data
CN120116946A
Bad driving behavior feedback intervention method and device based on driving behavior big data
CN120116946B
