An AI-based automobile safety driving monitoring system

Abstract

The application relates to the technical field of automobile safe driving monitoring, in particular to an automobile safe driving monitoring system based on AI, which comprises an information collection unit, which is used to collect the line-of-sight observation track of a driver, the posture of the driver driving a vehicle and the environmental image information in the vehicle driving process and the sound information of the surrounding environment in the vehicle driving process; an information preprocessing unit, which is used to preprocess the information collected by the information collection unit; an AI management unit, which is used to analyze and adjust the driving state of the driver according to the information after preprocessing; and a prewarning unit, which is used to issue a prewarning when the driving state of the driver does not meet preset standards. The application improves the monitoring precision of the driving state of the driver.

Description

Technical Field

[0001] This invention relates to the field of automotive safe driving monitoring technology, and in particular to an AI-based automotive safe driving monitoring system. Background Technology

[0002] With the continuous increase in car ownership and people's growing emphasis on safe driving, the market demand for automotive safety monitoring systems continues to grow. This is especially true in the commercial vehicle and taxi sectors, where the requirements for monitoring driver behavior and ensuring vehicle safety are even more stringent.

[0003] In recent years, artificial intelligence technology has made significant progress in various fields, providing strong technical support for the development of automotive safety driving monitoring systems. Through technologies such as deep learning and image recognition, the system can more accurately identify poor driving behaviors such as driver fatigue and distraction, as well as environmental information such as roads, traffic signs, vehicles, and pedestrians. The widespread application of high-precision sensing devices such as cameras, radar, and lidar enables vehicles to acquire information about their surrounding environment in real time, providing accurate data input for AI algorithms.

[0004] AI-based vehicle safety monitoring systems are gradually becoming an important part of modern automotive technology. These systems integrate advanced sensors, computers, and controllers to monitor driver behavior and vehicle operating status in real time, thereby improving driving safety and reducing traffic accident rates.

[0005] Chinese Patent Publication No. CN110406540A discloses a bus driver safety monitoring system, comprising an information receiving module, a control processing module, and an information transmission module. The information receiving module connects to external devices via a wireless network, including a smart bracelet, a thermal imaging camera, an infrared camera, and a vehicle safety indicator. The control processing module includes a data analysis module, an image processing module, a logic judgment module, and a status storage module. It processes data from the information receiving module according to its source and stores the current alert level. Alert levels are accumulated or eliminated based on different triggering conditions. The information transmission module receives information and control signals from the control processing module and connects to a bus passenger transport center and the public security department via a wireless network. This invention monitors the safety status of bus drivers, focusing on whether the driver has been attacked, and has significant practical implications for public transportation safety.

[0006] Therefore, the bus driver safety monitoring system has the following problems: The invention determines the warning status level by setting various modules and accumulates or eliminates the warning status according to different triggering conditions. For the warning status that cannot be eliminated, it connects to the bus passenger transport center and the public security department through a wireless network for processing. However, the invention has low accuracy in monitoring the driver's line of sight deviation and cannot determine whether the driver has observed the surrounding environment of the vehicle in a sufficient manner during the driving process, thus it cannot determine whether the driver's driving state is in a safe state. Summary of the Invention

[0007] To address this issue, the present invention provides an AI-based vehicle safety driving monitoring system to overcome the problem of low accuracy in monitoring driver line-of-sight deviation in existing technologies.

[0008] To achieve the above objectives, the present invention provides an AI-based vehicle safety driving monitoring system, characterized in that it includes:

[0009] The information acquisition unit includes an image information acquisition component and an audio information acquisition component. The image information acquisition component is used to acquire the driver's line of sight, the driver's posture while driving the vehicle, and environmental image information during the vehicle's driving process. The audio information acquisition component is used to record the sound information of the surrounding environment during the vehicle's driving process.

[0010] An information preprocessing unit, connected to the information acquisition unit, is used to preprocess the information acquired by the information acquisition unit, including an image information preprocessing component and an audio information preprocessing component.

[0011] The AI ​​management unit is connected to the information preprocessing unit and is used to determine whether the current driving state meets the preset standard based on the line of sight overlap, or to make a second determination on whether the current driving state meets the preset standard based on the driver's driving time.

[0012] The warning unit, which is connected to the AI ​​management unit, is used to issue warning information to the driver based on the judgment result of the AI ​​management unit.

[0013] Furthermore, the AI ​​management unit establishes a Cartesian coordinate system based on the preprocessed single-frame vehicle environment image information to form a preset line-of-sight observation trajectory, compares it with the real-time collected line-of-sight observation trajectory of the driver, and records the comparison result as the first preset line-of-sight overlap threshold.

[0014] Furthermore, the AI ​​management unit determines the type of road condition based on the road condition evaluation coefficient, wherein the types of road conditions include complex road conditions, standard road conditions, and simple road conditions.

[0015] Furthermore, the formula for calculating the road condition evaluation coefficient is as follows:

[0016]

[0017] Wherein, P represents the road condition evaluation coefficient, α represents the influence coefficient of sound volume, dB represents the sound decibel value of the environment in which the vehicle is located, dB0 represents the sound limit of the environment in which the vehicle is located, β represents the influence coefficient of sound type, T represents the sound type of the environment in which the vehicle is located, and T0 represents all sound types in the city. Herein, dB and T are obtained based on the sound information of the surrounding environment recorded during the vehicle's driving process.

[0018] Furthermore, the AI ​​management unit determines, based on the line-of-sight overlap, whether the driver's driving state within a single detection cycle meets a preset standard for that cycle.

[0019] If the line-of-sight overlap is greater than or equal to the first preset line-of-sight overlap threshold, the AI ​​management unit determines that the driver's driving state meets the preset standard.

[0020] If the line-of-sight overlap is less than the first preset line-of-sight overlap threshold and greater than or equal to the second preset line-of-sight overlap threshold, the AI ​​management unit will make a second determination on whether the driver's driving status meets the preset standard based on the driver's driving time.

[0021] If the line-of-sight overlap is less than the second preset line-of-sight overlap threshold, the AI ​​management unit determines that the driver's driving state does not meet the preset standard and controls the warning unit to issue a warning.

[0022] Furthermore, the AI ​​management unit makes a secondary determination based on the driver's driving time to determine whether the driver's driving status within the detection period meets the preset standard.

[0023] If the driving time is greater than or equal to the preset driving time threshold, the AI ​​management unit corrects the second preset line-of-sight overlap threshold based on the line-of-sight overlap difference.

[0024] If the driving time is less than a preset driving time threshold, the AI ​​management unit will determine whether the driver's driving state meets the preset standard three times based on the driver's driving posture and the driver's reaction time.

[0025] Furthermore, the AI ​​management unit sets several threshold adjustment methods for the second preset line-of-sight overlap threshold based on the line-of-sight overlap difference, and each threshold adjustment method has a different adjustment range for the second preset line-of-sight overlap threshold.

[0026] Furthermore, the AI ​​management unit determines that if the cumulative driving time of the driving postures appearing in the historical driving trips that match the current driver's driving posture does not meet the preset standard, it controls the warning unit to issue a warning.

[0027] Furthermore, the AI ​​management unit determines whether the driving state meets the preset standard three times based on the comparison result between the driver's reaction time and the preset reaction time threshold, and determines the type of road condition if the condition does not meet the standard.

[0028] Furthermore, the AI ​​management unit determines the correction method for the second preset line-of-sight overlap threshold based on the type of road condition.

[0029] Compared with the prior art, the beneficial effects of the present invention are as follows: by setting up an information collection unit to collect information on the driver's driving state, line of sight observation trajectory and the surrounding environment of the vehicle, setting up an information preprocessing unit to preprocess the information collected by the information collection unit to improve the processing efficiency and accuracy of the AI ​​management unit, setting up an AI management unit to analyze and adjust according to the preprocessed information to ensure the driver's safe driving, and setting up an early warning unit to issue an early warning to the driver based on the analysis results of the AI ​​management unit, the monitoring accuracy of the driver's driving state is improved, and the driver's safety during vehicle driving is improved.

[0030] Furthermore, the AI ​​management unit of the present invention records the ratio of the preset line-of-sight observation trajectory to the driver's real-time line-of-sight observation trajectory as the line-of-sight overlap degree, calculates the average line-of-sight overlap degree based on historical safe driving states and records it as the first preset line-of-sight overlap degree, and sets the driver's safe driving state based on historical safe driving trips, thereby improving the monitoring accuracy of the driver's driving state and improving the driver's safety during vehicle driving.

[0031] Furthermore, the AI ​​management unit of this invention calculates the road condition evaluation coefficient based on the preprocessed sound information, and determines the road condition of the current driving process based on the road condition evaluation coefficient, providing more accurate environmental information for monitoring the driver's driving status, thereby improving the monitoring accuracy of the driver's driving status and improving the driver's safety during vehicle driving.

[0032] Furthermore, the AI ​​management unit of this invention evaluates road conditions based on the different influence weights of sound volume and type, thereby improving the accuracy of road condition evaluation. Based on this, it improves the accuracy of monitoring the driver's driving status and enhances the driver's safety during vehicle operation.

[0033] Furthermore, the AI ​​management unit of the present invention determines whether the driver's driving state meets the preset standard within the detection period based on the line-of-sight overlap degree, and distinguishes the driver's driving state using the preset line-of-sight overlap degree threshold, thereby improving the monitoring accuracy of the driver's driving state and improving the driver's safety during vehicle driving.

[0034] Furthermore, the AI ​​management unit of this invention makes a secondary determination of whether the driver's driving status meets the preset standard based on the driver's driving time, thereby further improving the monitoring accuracy of the driver's status and thus improving the driver's safety during vehicle driving.

[0035] Furthermore, the AI ​​management unit of the present invention increases the second preset line-of-sight overlap threshold when the driving time exceeds the preset driving time, and determines the increase of the second preset line-of-sight overlap threshold based on the line-of-sight overlap difference, thereby improving the monitoring accuracy of the driver's driving state and improving the driver's safety during vehicle driving.

[0036] Furthermore, the AI ​​management unit of this invention controls the information collection unit to collect the driver's driving posture when the driving time is less than or equal to the preset driving time, and judges whether the current driving process meets the standard based on the historical driving process, thereby improving the monitoring accuracy of the driver's driving status and improving the driver's safety during vehicle driving.

[0037] Furthermore, the AI ​​management unit of the present invention monitors the driver's reaction time when the driver drives the vehicle in the current driving posture for less than a preset cumulative time, and determines whether the current driving process meets the preset standard based on the reaction time monitoring results, thereby improving the monitoring accuracy of the driver's driving state and improving the driver's safety during vehicle driving.

[0038] Furthermore, the AI ​​management unit of the present invention determines the adjustment range of the second preset line-of-sight overlap threshold based on the road conditions under which the driver is driving, thereby improving the monitoring accuracy of the driver's driving state and enhancing the driver's safety during vehicle driving. Attached Figure Description

[0039] Figure 1 This is a structural block diagram of an AI-based vehicle safety driving monitoring system according to an embodiment of the present invention;

[0040] Figure 2 This is a flowchart illustrating how to determine whether a driver's driving state meets a preset standard, according to an embodiment of the present invention.

[0041] Figure 3 This is a flowchart illustrating the process of correcting the second preset line-of-sight overlap threshold in an embodiment of the present invention;

[0042] Figure 4 This is a flowchart illustrating a secondary determination of whether the driver's driving state meets a preset standard, according to an embodiment of the present invention. Detailed Implementation

[0043] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0044] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0045] It should be noted that in the description of this invention, the terms "upper", "lower", "left", "right", "inner", "outer", etc., which indicate directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and is not intended to indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of this invention.

[0046] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0047] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0048] Please see Figure 1 , Figure 2 , Figure 3 as well as Figure 4 The diagrams shown are: a structural block diagram of an AI-based vehicle safety driving monitoring system according to an embodiment of the present invention; a flowchart of a process for determining whether a driver's driving state meets a preset standard according to an embodiment of the present invention; a flowchart of a process for correcting the second preset line-of-sight overlap threshold according to an embodiment of the present invention; and a flowchart of a process for a second determination of whether the driver's driving state meets a preset standard according to an embodiment of the present invention.

[0049] The AI-based vehicle safety driving monitoring system described in this embodiment of the invention includes:

[0050] The information acquisition unit includes an image information acquisition component and an audio information acquisition component. The image information acquisition component is used to acquire the driver's line of sight, the driver's posture while driving the vehicle, and environmental image information during the vehicle's operation. The audio information acquisition component is located outside the vehicle and is used to record the sound information of the surrounding environment during the vehicle's operation.

[0051] An information preprocessing unit, which is connected to the information acquisition unit, includes an image information preprocessing component and an audio information preprocessing component. The image information preprocessing component is used to extract frames from the video acquired by the image information acquisition component to obtain several image information and extract image features from the image information. The audio information preprocessing component is used to perform noise reduction and filtering on the audio information acquired by the audio information acquisition component.

[0052] The AI ​​management unit is connected to the information preprocessing unit and is used to determine whether the current driving state meets the preset standard based on the line of sight overlap, or to make a second determination on whether the current driving state meets the preset standard based on the driver's driving time, and to issue a warning message for driving states that do not meet the preset standard.

[0053] The early warning unit, which is connected to the AI ​​management unit, is used to issue early warning information to the driver based on the analysis results of the AI ​​management unit.

[0054] Specifically, the image information acquisition component is an infrared camera, which is not limited to any particular type, and is used to collect the surrounding environment, the driver's line of sight, and the driver's posture while driving the vehicle.

[0055] Specifically, the sound information acquisition component consists of an audio sensor and an amplification circuit. The specific model and location are not limited, and it is used to collect sound information during vehicle operation.

[0056] Specifically, the location of the image information acquisition component is not limited, but the location of the component should ensure that the image information of the driver's line of sight, the vehicle's surrounding environment, and the driver's driving posture is taken as the origin from the position corresponding to the center of the steering wheel.

[0057] Specifically, the process by which the AI ​​management unit calculates the first preset line-of-sight overlap threshold based on the preprocessed single-frame vehicle environment image information includes:

[0058] The AI ​​management unit sets the detection cycle to 1 second for a single driving state.

[0059] Based on the vehicle environment image information preprocessed within any detection cycle, the AI ​​management unit establishes a rectangular coordinate system with the position corresponding to the center of the vehicle's steering wheel as the origin, and marks the image features around the vehicle with position information to form a preset line of sight observation trajectory.

[0060] The AI ​​management unit controls the image information acquisition component to record the driver's line of sight observation trajectory at the time node corresponding to the vehicle environment image information;

[0061] The AI ​​management unit records the ratio of the gaze observation trajectory to the preset gaze observation trajectory as the gaze overlap degree;

[0062] The AI ​​management unit calculates the average value of the line-of-sight overlap based on historical safe driving trips and records the average value as the first preset line-of-sight overlap threshold.

[0063] Specifically, the calculation process for the average line-of-sight overlap degree is as follows:

[0064] The ratio of the sum of the line-of-sight overlap rates collected in all detection cycles during a single safe driving trip to the driving time is recorded as the average line-of-sight overlap rate for a single trip.

[0065] The average line-of-sight overlap is the ratio of the sum of the average single line-of-sight overlap of historical safe driving trips to the number of historical safe driving trips.

[0066] Specifically, the AI ​​management unit determines the decibel level and sound type of the vehicle's environment based on the pre-processed sound information, and calculates a road condition evaluation coefficient based on the obtained decibel level and sound type.

[0067] If the road condition evaluation coefficient is greater than or equal to the first preset coefficient of 0.9, the AI ​​management unit determines that the road condition is a complex road condition.

[0068] If the road condition evaluation coefficient is less than the first preset coefficient of 0.9 and greater than or equal to the second preset coefficient of 0.6, then the AI ​​management unit determines that the road condition is a standard road condition.

[0069] If the road condition evaluation coefficient is less than the second preset coefficient of 0.6, the AI ​​management unit determines the road condition to be a simple road condition.

[0070] Specifically, the formula for calculating the road condition evaluation coefficient is as follows:

[0071]

[0072] Where P represents the road condition evaluation coefficient, α represents the influence coefficient of sound volume (0.4), dB represents the sound decibel value of the environment in which the vehicle is located, dB0 represents the sound limit of the environment in which the vehicle is located, β represents the influence coefficient of sound type (0.6), T represents the sound type of the environment in which the vehicle is located, and T0 represents all 70 sound types in the city.

[0073] Specifically, the noise limits for the vehicle's environment, such as in areas requiring special quiet, like sanatoriums, upscale villa areas, and luxury hotel areas, are 50 decibels during the day and 40 decibels at night; for areas along main urban roads, inland waterways, and main and secondary railway lines, the limits are 70 decibels during the day and 55 decibels at night, with no specific restrictions.

[0074] Specifically, the AI ​​management unit determines, based on the line-of-sight overlap, whether the driver's driving status meets a preset standard within a single detection cycle.

[0075] If the line-of-sight overlap is greater than or equal to the first preset line-of-sight overlap threshold of 90%, the AI ​​management unit determines that the driver's driving status meets the preset standard within the detection period.

[0076] If the line-of-sight overlap is less than the first preset line-of-sight overlap threshold of 90% and greater than or equal to the second preset line-of-sight overlap threshold of 85%, the AI ​​management unit will make a second determination based on the driver's driving time to determine whether the driver's driving status meets the preset standard within the detection period.

[0077] If the line-of-sight overlap is less than 85% of the second preset line-of-sight overlap threshold, the AI ​​management unit determines that the driver's driving status does not meet the preset standard within the detection period and controls the warning unit to issue a warning.

[0078] Specifically, the AI ​​management unit makes a secondary determination based on the driver's driving time to determine whether the driver's driving status within the detection period meets preset standards.

[0079] If the driving time is greater than or equal to the preset driving time threshold of 1 hour, the AI ​​management unit corrects the second preset line-of-sight overlap threshold by 85% based on the line-of-sight overlap difference.

[0080] If the driving time is less than the preset driving time threshold of 1 hour, the AI ​​management unit will determine whether the driver's driving state meets the preset standard three times based on the driver's driving posture and the driver's reaction time.

[0081] Specifically, the process by which the AI ​​management unit corrects the second preset line-of-sight overlap threshold based on the line-of-sight overlap difference includes:

[0082] If the difference in line-of-sight overlap is less than or equal to 3% of the first preset difference, the AI ​​management unit uses a first correction coefficient of 1.02 to correct the second preset line-of-sight overlap threshold.

[0083] If the difference in line-of-sight overlap is greater than the first preset difference of 3% and less than or equal to the second preset difference of 5%, then the AI ​​management unit uses a second correction coefficient of 1.03 to correct the second preset line-of-sight overlap threshold.

[0084] If the difference in line-of-sight overlap is greater than the second preset difference by 5%, the AI ​​management unit uses a third correction coefficient of 1.05 to correct the second preset line-of-sight overlap threshold to 85%.

[0085] Specifically, the AI ​​management unit controls the image information acquisition component to collect the driver's driving posture within the current detection period, and retrieves the cumulative driving time of driving the vehicle in the same posture from historical driving trips.

[0086] If the cumulative driving time is greater than or equal to the preset cumulative time of 20 hours, the AI ​​management unit determines whether the driving state meets the preset standard based on the driver's reaction time.

[0087] If the cumulative driving time is less than the preset cumulative driving time of 20 hours, the AI ​​management unit determines that the driving status does not meet the preset standard and controls the warning unit to issue a warning.

[0088] Specifically, the AI ​​management unit determines the driver's driving posture by controlling the image acquisition component to acquire the driver's driving posture during the current driving journey, establishing a Cartesian coordinate system with the position corresponding to the center point of the steering wheel as the origin, marking the coordinates of the driver's left hand, left elbow, right hand, right elbow and head, and matching them with the coordinate positions corresponding to the driver's driving posture in the driver's historical driving journey. If the position coordinates match completely, it is determined to be the same driving posture.

[0089] Specifically, the process by which the AI ​​management unit determines whether the driving state meets preset standards based on the driver's reaction time includes:

[0090] If the reaction time exceeds the preset reaction time threshold of 3 seconds, the AI ​​management unit determines that the current driving state does not meet the preset standard and controls the warning unit to issue a warning.

[0091] If the reaction time is less than or equal to the preset reaction time threshold of 3 seconds, the AI ​​management unit determines that the current driving state meets the preset standard and determines the correction method of the second preset line-of-sight overlap threshold of 85% based on the road conditions.

[0092] Specifically, the process by which the AI ​​management unit determines that the current driving state meets the preset standard and determines the correction method for the first preset line-of-sight overlap threshold based on the road conditions includes:

[0093] If the road condition is a simple road condition, the AI ​​management unit determines to use the first correction coefficient of 1.02 to correct the second preset line-of-sight overlap threshold of 85%.

[0094] If the road conditions are standard, the AI ​​management unit determines to use the second correction coefficient of 1.03 to correct the second preset line-of-sight overlap threshold of 85%.

[0095] If the road conditions are standard, the AI ​​management unit determines to use the third correction coefficient of 1.05 to correct the second preset line-of-sight overlap threshold of 85%.

[0096] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. An AI-based vehicle safety driving monitoring system, characterized in that, include: The information acquisition unit includes an image information acquisition component and an audio information acquisition component. The image information acquisition component is used to acquire the driver's line of sight, the driver's posture while driving the vehicle, and environmental image information during the vehicle's driving process. The audio information acquisition component is used to record the sound information of the surrounding environment during the vehicle's driving process. An information preprocessing unit, connected to the information acquisition unit, is used to preprocess the information acquired by the information acquisition unit, including an image information preprocessing component and an audio information preprocessing component. The AI ​​management unit is connected to the information preprocessing unit and is used to determine whether the current driving state meets the preset standard based on the line of sight overlap, or to make a second determination on whether the current driving state meets the preset standard based on the driver's driving time. A warning unit, which is connected to the AI ​​management unit, is used to issue warning information to the driver based on the judgment result of the AI ​​management unit; The AI ​​management unit establishes a rectangular coordinate system based on the preprocessed single-frame vehicle environment image information to form a preset line of sight observation trajectory, compares it with the real-time collected line of sight observation trajectory of the driver, and records the comparison result as the first preset line of sight overlap threshold. The AI ​​management unit determines, based on the line-of-sight overlap, whether the driver's driving status meets preset standards within a single detection cycle. If the line-of-sight overlap is greater than or equal to the first preset line-of-sight overlap threshold, the AI ​​management unit determines that the driver's driving state meets the preset standard. If the line-of-sight overlap is less than the first preset line-of-sight overlap threshold and greater than or equal to the second preset line-of-sight overlap threshold, the AI ​​management unit will make a second determination on whether the driver's driving status meets the preset standard based on the driver's driving time. If the line-of-sight overlap is less than the second preset line-of-sight overlap threshold, the AI ​​management unit determines that the driver's driving state does not meet the preset standard and controls the warning unit to issue a warning. The AI ​​management unit determines the type of road condition based on the road condition evaluation coefficient, which includes complex road conditions, standard road conditions, and simple road conditions. The formula for calculating the road condition evaluation coefficient is as follows: Wherein, P represents the road condition evaluation coefficient, α represents the influence coefficient of sound volume, dB represents the sound decibel value of the environment in which the vehicle is located, dB0 represents the sound limit of the environment in which the vehicle is located, β represents the influence coefficient of sound type, T represents the sound type of the environment in which the vehicle is located, and T0 represents all sound types in the city. Herein, dB and T are obtained based on the sound information of the surrounding environment recorded during the vehicle's driving process. The AI ​​management unit determines the correction method for the second preset line-of-sight overlap threshold based on the type of road condition.

2. The AI-based vehicle safety driving monitoring system according to claim 1, characterized in that, The AI ​​management unit makes a secondary determination based on the driver's driving time to determine whether the driver's driving status meets the preset standard within the detection period. If the driving time is greater than or equal to the preset driving time threshold, the AI ​​management unit corrects the second preset line-of-sight overlap threshold based on the line-of-sight overlap difference. If the driving time is less than a preset driving time threshold, the AI ​​management unit will determine whether the driver's driving state meets the preset standard three times based on the driver's driving posture and the driver's reaction time.

3. The AI-based vehicle safety driving monitoring system according to claim 2, characterized in that, The AI ​​management unit sets several threshold adjustment methods for the second preset line-of-sight overlap threshold based on the line-of-sight overlap difference, and each threshold adjustment method has a different adjustment range for the second preset line-of-sight overlap threshold.

4. The AI-based vehicle safety driving monitoring system according to claim 3, characterized in that, The AI ​​management unit determines that if the cumulative driving time of the driving postures appearing in the historical driving trips that match the current driver's driving posture does not meet the preset standard, it controls the warning unit to issue a warning.

5. The AI-based vehicle safety driving monitoring system according to claim 4, characterized in that, The AI ​​management unit determines whether the driving state meets the preset standard three times based on the comparison result between the driver's reaction time and the preset reaction time threshold, and determines the type of road condition if the driving state does not meet the standard.

Images