A dangerous driving detection method and system based on big data

Abstract

The present invention discloses a dangerous driving detection method and system based on big data, and relates to the technical field of automobile safety detection and analysis. The method comprises the following steps: analyzing the pre-processed vehicle violation data and the pre-processed vehicle abnormal driving data to obtain the vehicle driving evaluation coefficient, analyzing the pre-processed vital sign capture data to obtain the driving concentration evaluation coefficient, comparing the driving concentration evaluation coefficient with the threshold of the impact of the environment on driving concentration, comparing the vehicle's dangerous driving comprehensive evaluation coefficient with the vehicle's dangerous driving comprehensive evaluation threshold, and regulating the vehicle through an adjustment module according to the comparison result. The present invention regulates the vehicle through the adjustment module, thereby achieving the effect of improving the versatility of the vehicle driving evaluation system for different situations, and solving the problem of insufficient versatility of correcting the vehicle driving evaluation system for handling different environmental situations in the prior art.

Description

Technical Field

[0001] The present invention relates to the technical field of automobile safety detection and analysis, and in particular to a dangerous driving detection method and system based on big data. Background Art

[0002] With the development of modern cities, transportation has become an indispensable part of people's daily lives. However, during long-term driving or driving at night, drivers are prone to fatigue, which leads to slow brain reactions and reduced response speed to unexpected road situations, thus causing traffic accidents. More seriously, the proportion of serious and major traffic accidents caused by fatigue driving has reached more than 40%, and fatigue driving has become the second largest cause of traffic accidents.

[0003] Existing assisted driving systems focus too much on the vehicle and road environment, while ignoring the driver's own status. By introducing this system, driving safety can be greatly improved, thereby fundamentally preventing traffic accidents. Traffic accidents can cause property losses, and in serious cases, even endanger the lives of drivers and pedestrians. Fatigue driving is different from drunk driving and speeding. The detection and prevention of fatigue driving is extremely difficult and cannot be detected in time by general instruments. Therefore, the driver's fatigue driving behavior seriously threatens the personal safety of drivers and pedestrians.

[0004] As for the problem of being unable to detect drivers in real time, with breakthroughs in artificial intelligence, machine learning, and machine vision recognition, as well as the intelligence of vehicle-mounted hardware, the academic community has proposed a new research direction based on machine vision and portrait capture algorithms to collect real-time images of drivers to determine whether the driver has abnormal behavior. Therefore, this project uses image recognition technology in machine vision, uses a vehicle-mounted camera to shoot the driver's face, uses OpenMV to run the trained neural network model, and detects fatigue driving on the driver's face, which can effectively prevent fatigue driving, especially in occupations such as long-distance truck drivers. At the same time, sensor technology is also used to detect the air quality and alcohol concentration in the car, and when necessary, the driver is reminded through warnings and takes corresponding measures to ensure driving safety.

[0005] For example, the in-vehicle safety detection and processing system disclosed in the patent application with publication number: CN114235042A includes: a temperature and humidity sensor, a water pressure sensor, a machine vision module, a processing device and a controller. The controller is used to: determine whether there are stranded people and infants in the car through the machine vision module after the engine is turned off, and automatically monitor the environmental data in the car through the temperature and humidity sensor. When the children in the car are in danger of suffocation or dehydration, the control processing device handles the problem of the children being stranded in the car; after the water pressure sensor monitors that the water pressure reaches a certain threshold for a long time, the control processing device handles the problem of the vehicle falling into the water; the machine vision module identifies the driver's driving condition, and when it is monitored that the driver has a dangerous behavior state, the control processing device handles the problem of dangerous driving of the driver in the car.

[0006] For example, the patent application with announcement number: CN118443078A discloses an online monitoring system and method for tank trucks, including: calculating the driving abnormality coefficient of the tank truck through abnormal vehicle behavior, abnormal road danger and abnormal driver; monitoring the state of the tank truck through fuel factors and tank factors, and calculating the abnormal state coefficient of the tank truck; starting from the danger level of the tank truck and the environment in which it is located, obtaining the threat radius of the tank truck according to the flash point and flammable and explosive characteristics of the fuel, quantifying and unifying the personnel and vehicles within the threat radius, and calculating the environmental threat coefficient of the tank truck in real time; calculating its transportation risk coefficient through the driving abnormality coefficient, tank state abnormality coefficient and environmental threat coefficient of the tank truck.

[0007] However, in the process of implementing the technical solution of the invention in the embodiments of the present application, the present application found that the above technology has at least the following technical problems:

[0008] In the prior art, the vehicle driving system can only detect speeding and overloading, while the environment will affect the driver data collection, the vehicle and its vehicle driving evaluation system, resulting in the problem of insufficient versatility of the vehicle driving evaluation system in handling different environmental conditions. Summary of the invention

[0009] The embodiments of the present application provide a dangerous driving detection method and system based on big data, thereby solving the problem in the prior art of correcting the insufficient versatility of the vehicle driving evaluation system in handling different environmental conditions, and achieving the effect of improving the versatility of the vehicle driving evaluation system in handling different situations.

[0010] The embodiment of the present application provides a dangerous driving detection method based on big data, comprising the following steps: collecting vehicle violation data, vehicle abnormal driving data and vital sign capture data through an on-board diagnostic system, cleaning, denoising and normalizing the vehicle violation data, vehicle abnormal driving data and vital sign capture data to obtain pre-processed vehicle violation data, pre-processed vehicle abnormal driving data and pre-processed vital sign capture data; removing the outliers of the pre-processed vehicle violation data, pre-processed vehicle abnormal driving data and pre-processed vital sign capture data according to the big data, analyzing the pre-processed vehicle violation data and the pre-processed vehicle abnormal driving data to obtain a vehicle driving evaluation coefficient, and performing noise reduction on the pre-processed vital sign capture data. The data is captured for analysis to obtain a driving concentration evaluation coefficient; the threshold value of the impact of the environment on driving concentration is obtained from a database, the driving concentration evaluation coefficient is compared with the threshold value of the impact of the environment on driving concentration, and the vehicle is regulated through an adjustment module according to the comparison result of the driving concentration evaluation coefficient and the threshold value of the impact of the environment on driving concentration; the vehicle driving evaluation coefficient and the driving concentration evaluation coefficient are comprehensively evaluated to obtain a comprehensive evaluation coefficient of dangerous driving of the vehicle, the comprehensive evaluation threshold of dangerous driving of the vehicle is obtained from a database, the comprehensive evaluation coefficient of dangerous driving of the vehicle is compared with the comprehensive evaluation threshold of dangerous driving of the vehicle, and the vehicle is regulated through an adjustment module according to the comparison result of the comprehensive evaluation coefficient of dangerous driving of the vehicle and the comprehensive evaluation threshold of dangerous driving of the vehicle.

[0011] Furthermore, the specific collection process of collecting vehicle violation data, abnormal vehicle driving data and vital sign capture data through the on-board diagnostic system is as follows: when the vehicle is started, the on-board diagnostic system is automatically initialized, monitors the vehicle's driving speed in real time, and compares it with the speed limit of the road, records the number of speeding times, records the number of red light runs through GPS positioning and the traffic rules database, and monitors in real time through the front camera whether the vehicle is driving in the opposite direction. The vehicle violation data includes the number of speeding times, the number of red light runs and whether the vehicle is driving in the opposite direction; the vital sign capture data is collected through the breathing sensor on the seat, the camera and the grip sensor on the steering wheel. The vital sign capture data includes the breathing rate, the number of head movements and the grip of the steering wheel; an optical sensor is installed inside the vehicle, and the visibility, current light intensity and atmospheric pressure are captured in real time through the optical sensor. The vital sign capture data includes visibility, current light intensity and atmospheric pressure.

[0012] Furthermore, the specific process of removing outliers of preprocessed vehicle violation data, preprocessed vehicle abnormal driving data and preprocessed vital sign capture data based on big data is as follows: arrange the preprocessed vehicle violation data, preprocessed vehicle abnormal driving data and preprocessed vital sign capture data in order from large to small and use a clustering algorithm to identify outliers of the preprocessed vehicle violation data, preprocessed vehicle abnormal driving data and preprocessed vital sign capture data, mark the outliers as outliers, replace the outliers with the average value of the previous and next data, and if there is no average value of the previous and next data, directly delete the outliers marked as outliers.

[0013] Furthermore, the preprocessed vehicle violation data and the preprocessed vehicle abnormal driving data are analyzed to obtain the specific analysis process of the vehicle driving evaluation coefficient: extract features from the preprocessed vehicle violation data and the preprocessed vehicle abnormal driving data, divide the allocation weight factors of the preprocessed vehicle violation data and the preprocessed vehicle abnormal driving data, combine the preprocessed vehicle violation data and the preprocessed vehicle abnormal driving data according to the weight factors of the preprocessed vehicle violation data and the preprocessed vehicle abnormal driving data to obtain the vehicle driving evaluation coefficient, and apply the vehicle driving evaluation coefficient to the real-time monitoring system to perform real-time evaluation of the vehicle driving status.

[0014] Furthermore, the specific analysis process of analyzing the preprocessed vital sign capture data is: extracting features from the preprocessed vital sign capture data, assigning weight factors to the parameters in the preprocessed vital sign capture data, and obtaining a driving concentration evaluation coefficient based on the weight factors of the parameters in the preprocessed vital sign capture data and combining the parameters in the preprocessed vital sign capture data; applying the driving concentration evaluation coefficient to the real-time monitoring system to perform real-time evaluation of the vehicle's driving concentration.

[0015] Furthermore, the specific comparison process of comparing the driving concentration evaluation coefficient with the threshold of environmental impact on driving concentration is as follows: if the driving concentration evaluation coefficient fluctuates, the system increases the sensitivity of the camera and the sensor through the adjustment module; if the driving concentration evaluation coefficient does not fluctuate, the threshold of environmental impact on driving concentration is obtained from the database, and the driving concentration evaluation coefficient is compared with the threshold of environmental impact on driving concentration; if the driving concentration evaluation coefficient is less than the threshold of environmental impact on driving concentration, the system analyzes the pre-processed vital sign capture data in real time; if the driving concentration evaluation coefficient is greater than or equal to the threshold of environmental impact on driving concentration, the system activates the seat vibration reminder and issues a voice reminder through the adjustment module, and further evaluates the driving concentration evaluation coefficient in combination with the vehicle driving evaluation coefficient.

[0016] Furthermore, the specific comparison process of comparing the vehicle's dangerous driving comprehensive assessment coefficient with the vehicle's dangerous driving comprehensive assessment threshold is as follows: when the driving concentration assessment coefficient does not fluctuate, the vehicle driving assessment coefficient and the driving concentration assessment coefficient are comprehensively evaluated to obtain the vehicle's dangerous driving comprehensive assessment coefficient, the vehicle's dangerous driving comprehensive assessment threshold is obtained from the database, the vehicle's dangerous driving comprehensive assessment coefficient is compared with the vehicle's dangerous driving comprehensive assessment threshold, if the vehicle's dangerous driving comprehensive assessment coefficient is less than the vehicle's dangerous driving comprehensive assessment threshold, the vehicle's driving is monitored and evaluated in real time, if the vehicle's dangerous driving comprehensive assessment coefficient is greater than or equal to the vehicle's dangerous driving comprehensive assessment threshold, the system automatically activates driving assistance functions including lane keeping assist and automatic emergency braking assist through the adjustment module, and compares the vehicle's dangerous driving comprehensive assessment coefficient with the vehicle's dangerous driving comprehensive assessment threshold again, if the vehicle's dangerous driving comprehensive assessment coefficient is greater than or equal to the vehicle's dangerous driving comprehensive assessment threshold, the system connects to the vehicle's intelligent driving system through the adjustment module to take over the car, stops the vehicle on the side of the road, sends a push to the cloud center through the app, and contacts the emergency contacts stored in advance.

[0017] Furthermore, the specific control method of controlling the vehicle through the adjustment module is: when the vehicle is detected to be speeding, the system reduces the speed by limiting the throttle response or enabling the cruise control system. When sudden braking is detected, the system activates the anti-lock braking system or the dynamic stability control system to optimize braking stability. For lane departure behavior, the system makes slight corrections to the steering wheel through the electric power steering system to help the vehicle return to the correct lane. If the driving concentration assessment coefficient is greater than or equal to the threshold of the environment's impact on driving concentration, the system activates the seat vibration reminder and issues a voice reminder through the adjustment module. If the vehicle's comprehensive dangerous driving assessment coefficient is greater than or equal to the vehicle's comprehensive dangerous driving assessment threshold, the system connects to the vehicle's intelligent driving system through the adjustment module to take over the car, stops the vehicle on the side of the road, and sends a push to the cloud center through the app to contact the emergency contacts stored in advance.

[0018] Furthermore, the specific method for obtaining the comprehensive evaluation coefficient of dangerous driving of a vehicle is as follows: ; ; In the formula, The comprehensive evaluation coefficient of dangerous driving of vehicles under each dangerous driving time monitoring period is used to evaluate the impact of vehicle driving and driving concentration on dangerous driving under each dangerous driving time monitoring period. The vehicle driving evaluation coefficient under each vehicle driving time monitoring section is used to evaluate the impact of vehicle violations and abnormal vehicle driving on vehicle driving under each vehicle driving time monitoring section. The driving concentration evaluation coefficient under each driving concentration time monitoring segment is used to evaluate the impact of the driving visual capture data under each driving concentration time monitoring segment on driving concentration. Expressed as the weight factor of the vehicle driving evaluation coefficient, Expressed as the weight factor of the driving focus evaluation coefficient, Expressed as The abnormal weather index under each abnormal weather time monitoring period is used to evaluate the impact of abnormal weather under each abnormal weather time monitoring period on dangerous driving control.

[0019] The embodiment of the present application provides a dangerous driving detection system based on big data, which includes a data collection module, a data analysis module, a threshold comparison module and an adjustment module: the data collection module is used to collect vehicle violation data, vehicle abnormal driving data and vital sign capture data through an on-board diagnostic system, clean, denoise and normalize the vehicle violation data, vehicle abnormal driving data and vital sign capture data to obtain pre-processed vehicle violation data, pre-processed vehicle abnormal driving data and pre-processed vital sign capture data; the data analysis module is used to remove the abnormal values ​​of the pre-processed vehicle violation data, pre-processed vehicle abnormal driving data and pre-processed vital sign capture data according to big data, analyze the pre-processed vehicle violation data and pre-processed vehicle abnormal driving data to obtain a vehicle driving evaluation coefficient, and analyze the pre-processed vital sign capture data to obtain a driving concentration evaluation coefficient; the adjustment module is used to control the vehicle through the adjustment module according to the comparison result of the vehicle dangerous driving comprehensive evaluation coefficient and the vehicle dangerous driving comprehensive evaluation threshold.

[0020] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0021] 1. By comparing the comprehensive evaluation coefficient of dangerous driving of the vehicle with the comprehensive evaluation threshold of dangerous driving of the vehicle, the vehicle is regulated by the adjustment module according to the comparison result of the comprehensive evaluation coefficient of dangerous driving of the vehicle and the comprehensive evaluation threshold of dangerous driving of the vehicle, thereby achieving the effect of improving the versatility of the vehicle driving evaluation system for different situations, and effectively solving the problem of insufficient versatility of the vehicle driving evaluation system for handling different environmental situations in the prior art.

[0022] 2. By arranging the preprocessed vehicle violation data, the preprocessed vehicle abnormal driving data and the preprocessed vital sign capture data in order from large to small and using a clustering algorithm to identify the outliers of the preprocessed vehicle violation data, the preprocessed vehicle abnormal driving data and the preprocessed vital sign capture data, marking the outliers as outliers, and replacing the outliers with the average value of the data before and after, the effect of improving the data accuracy is achieved, and the problem of insufficient data accuracy in the prior art is effectively solved.

[0023] 3. By analyzing the pre-processed vehicle violation data and the pre-processed vehicle abnormal driving data, the vehicle driving evaluation coefficient is obtained, thereby improving the accuracy of adjusting one's driving behavior according to the evaluation results. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 A flowchart of a dangerous driving detection method based on big data provided in an embodiment of the present application;

[0025] Figure 2 An image of an abnormal weather index in a dangerous driving detection method based on big data provided in an embodiment of the present application;

[0026] Figure 3 A schematic diagram of the structure of a dangerous driving detection system based on big data provided in an embodiment of the present application. DETAILED DESCRIPTION

[0027] The embodiments of the present application provide a dangerous driving detection method and system based on big data, thereby solving the problem in the prior art of the vehicle driving evaluation system's lack of versatility in handling different environmental conditions. The vehicle's comprehensive dangerous driving evaluation coefficient is compared with the vehicle's comprehensive dangerous driving evaluation threshold, and the vehicle is regulated through an adjustment module according to the comparison result of the vehicle's comprehensive dangerous driving evaluation coefficient and the vehicle's comprehensive dangerous driving evaluation threshold, thereby achieving the effect of improving the versatility of the vehicle driving evaluation system for different situations.

[0028] The technical solution in the embodiment of the present application is to solve the above problem and correct the problem that the vehicle driving evaluation system is not versatile enough to handle different environmental conditions. The overall idea is as follows:

[0029] By collecting vehicle violation data, vehicle abnormal driving data and vital sign capture data and preprocessing them, the abnormal values ​​of the collected vehicle violation data, vehicle abnormal driving data and vital sign capture data are corrected or deleted, the preprocessed vehicle violation data and the preprocessed vehicle abnormal driving data are analyzed to obtain a vehicle driving evaluation coefficient, the preprocessed vital sign capture data are analyzed to obtain a driving concentration evaluation coefficient, the driving concentration evaluation coefficient is compared with the threshold of the impact of the environment on the driving concentration, and the vehicle is regulated through an adjustment module according to the comparison result of the driving concentration evaluation coefficient and the threshold of the impact of the environment on the driving concentration, the vehicle driving evaluation coefficient and the driving concentration evaluation coefficient are comprehensively evaluated to obtain a comprehensive evaluation coefficient of vehicle dangerous driving, the comprehensive evaluation coefficient of vehicle dangerous driving is compared with the comprehensive evaluation threshold of vehicle dangerous driving, and the vehicle is regulated through an adjustment module according to the comparison result of the comprehensive evaluation coefficient of vehicle dangerous driving and the comprehensive evaluation threshold of vehicle dangerous driving, thereby achieving the effect of improving the versatility of the vehicle driving evaluation system for different situations.

[0030] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0031] like Figure 1 As shown, it is a flow chart of a dangerous driving detection method based on big data provided by an embodiment of the present application. The method is applied to a dangerous driving detection system based on big data. The method comprises the following steps: collecting vehicle violation data, vehicle abnormal driving data and vital sign capture data through an on-board diagnostic system, cleaning, denoising and normalizing the vehicle violation data, vehicle abnormal driving data and vital sign capture data to obtain pre-processed vehicle violation data, pre-processed vehicle abnormal driving data and pre-processed vital sign capture data; removing the outliers of the pre-processed vehicle violation data, pre-processed vehicle abnormal driving data and pre-processed vital sign capture data according to the big data, analyzing the pre-processed vehicle violation data and the pre-processed vehicle abnormal driving data to obtain vehicle violation data. The vehicle driving assessment coefficient is obtained, and the pre-processed vital sign capture data is analyzed to obtain the driving concentration assessment coefficient; the threshold value of the impact of the environment on driving concentration is obtained from the database, the driving concentration assessment coefficient is compared with the threshold value of the impact of the environment on driving concentration, and the vehicle is regulated through the adjustment module according to the comparison result of the driving concentration assessment coefficient and the threshold value of the impact of the environment on driving concentration; the vehicle driving assessment coefficient and the driving concentration assessment coefficient are comprehensively evaluated to obtain the comprehensive assessment coefficient of dangerous driving of the vehicle, the comprehensive assessment threshold of dangerous driving of the vehicle is obtained from the database, the comprehensive assessment coefficient of dangerous driving of the vehicle is compared with the comprehensive assessment threshold of dangerous driving of the vehicle, and the vehicle is regulated through the adjustment module according to the comparison result of the comprehensive assessment coefficient of dangerous driving of the vehicle and the comprehensive assessment threshold of dangerous driving of the vehicle.

[0032] In this embodiment, the on-board diagnostic system is connected to the data collection module. The on-board diagnostic system includes sensors, cameras, and a recorder of external environment data. The sensors include brake sensors, acceleration sensors, speed sensors, steering angle sensors, and GPS sensors. The cameras include front cameras, built-in cameras, and visual tracking systems. The external environment includes light sensors and meteorological sensors. The data recorder is connected to the on-board diagnostic system to transmit the above-collected data to the data recorder. The vehicle diagnostic system provides a basis for a dangerous driving detection system based on big data, and is integrated with the Internet to improve the detection efficiency of a dangerous driving detection system based on big data.

[0033] Furthermore, the specific collection process of collecting vehicle violation data, abnormal vehicle driving data and vital sign capture data through the on-board diagnostic system is as follows: when the vehicle is started, the on-board diagnostic system is automatically initialized, monitors the vehicle's driving speed in real time, and compares it with the speed limit of the road, records the number of speeding times, records the number of red light runs through GPS positioning and the traffic rules database, and monitors in real time through the front camera whether the vehicle is driving in the opposite direction. The vehicle violation data includes the number of speeding times, the number of red light runs and whether the vehicle is driving in the opposite direction; the vital sign capture data is collected through the breathing sensor on the seat, the camera and the grip sensor on the steering wheel. The vital sign capture data includes the breathing rate, the number of head movements and the grip of the steering wheel; an optical sensor is installed inside the vehicle, and the visibility, current light intensity and atmospheric pressure are captured in real time through the optical sensor. The vital sign capture data includes visibility, current light intensity and atmospheric pressure.

[0034] In this embodiment, when the vehicle is started, the on-board diagnostic system is automatically initialized, and all sensors and cameras are ready. The system's speed sensor monitors the vehicle's driving speed in real time and compares it with the built-in road speed limit database, records the number of speeding times, monitors whether the vehicle passes through the intersection when the light is red through GPS positioning or the front camera, and records the number of red light runs. The front camera monitors the vehicle's driving direction in real time to determine whether it is driving in the opposite direction. The system monitors the acceleration and deceleration of the vehicle. When it exceeds a preset threshold, it records the number of sudden acceleration or sudden braking. Combined with the vehicle's steering system and speed data, it records the number of sharp turns. The vehicle's internal sensors capture the driver's visibility, current light intensity, and atmospheric pressure in real time to assess the driver's fatigue state. When the driver realizes that his behavior is being monitored, he may pay more attention to complying with traffic rules and improve his driving habits.

[0035] Furthermore, the specific process of removing outliers of preprocessed vehicle violation data, preprocessed vehicle abnormal driving data and preprocessed vital sign capture data based on big data is as follows: arrange the preprocessed vehicle violation data, preprocessed vehicle abnormal driving data and preprocessed vital sign capture data in order from large to small and use a clustering algorithm to identify outliers of the preprocessed vehicle violation data, preprocessed vehicle abnormal driving data and preprocessed vital sign capture data, mark the outliers as outliers, replace the outliers with the average value of the previous and next data, and if there is no average value of the previous and next data, directly delete the outliers marked as outliers.

[0036] In this embodiment, a clustering algorithm is applied to perform cluster analysis on the preprocessed data. The clustering algorithm includes a K-means algorithm and a DBSCAN algorithm. For example, the K-means algorithm first determines the number of clusters of the K-means clustering algorithm, randomly selects k preprocessed vehicle violation data points as initial cluster centers, calculates the distance from each preprocessed vehicle violation data point to each cluster center, and assigns the preprocessed vehicle violation data point to the nearest cluster center. The center point of each cluster is recalculated, usually taking the mean of all points in the cluster. Repeat the steps of assigning the preprocessed vehicle violation data point to the cluster center and updating the cluster center until the stop condition is met (such as the change of the cluster center is less than or reaches a preset number of iterations). After the clustering is completed, for each cluster, the distance from all points to its cluster center is calculated, and the average value of the distance from all points to the cluster center is used as a threshold. If the distance from a point to its cluster center is greater than the set threshold, the point is marked as an outlier. By identifying and replacing outliers, noise and abnormal values ​​in the data can be eliminated, and the overall quality and reliability of the data can be improved.

[0037] Furthermore, the preprocessed vehicle violation data and the preprocessed vehicle abnormal driving data are analyzed to obtain the specific analysis process of the vehicle driving evaluation coefficient: extract features from the preprocessed vehicle violation data and the preprocessed vehicle abnormal driving data, divide the allocation weight factors of the preprocessed vehicle violation data and the preprocessed vehicle abnormal driving data, combine the preprocessed vehicle violation data and the preprocessed vehicle abnormal driving data according to the weight factors of the preprocessed vehicle violation data and the preprocessed vehicle abnormal driving data to obtain the vehicle driving evaluation coefficient, and apply the vehicle driving evaluation coefficient to the real-time monitoring system to perform real-time evaluation of the vehicle driving status.

[0038] In this embodiment, the specific method for obtaining the vehicle driving evaluation coefficient is:

[0039] ;

[0040] In the formula, Expressed as The vehicle driving evaluation coefficient under each vehicle driving time monitoring section is used to evaluate the impact of vehicle violations and abnormal vehicle driving on vehicle driving under each vehicle driving time monitoring section. Several vehicle driving time monitoring points are set for vehicle monitoring. , Indicates the total number of vehicle driving time monitoring points. 20 vehicle driving time monitoring points are regarded as a driving time monitoring segment. Expressed as The number of speeding times in each vehicle driving time monitoring section, Expressed as The number of red light violations in each vehicle driving time monitoring segment, Expressed as Whether the vehicle is driving in the wrong direction during the driving time monitoring period, if so, If you don't go against =1, Expressed as The number of sudden accelerations in each vehicle driving time monitoring segment, Expressed as The number of emergency brakes in each vehicle driving time monitoring section, Expressed as The number of sharp turns in each vehicle driving time monitoring segment, Represented as the weight factor of the preprocessed vehicle violation data, It is represented as the weight factor of the preprocessed abnormal driving data of the vehicle.

[0041] Using physical relationship analysis, the more times you run a red light, the smaller the vehicle driving evaluation coefficient is. When driving in the opposite direction, the vehicle driving evaluation coefficient is directly , when not retrograde, =1, the vehicle is directly controlled according to the calculated vehicle driving evaluation coefficient.

[0042] The weight factors of the preprocessed vehicle violation data and the weight factors of the preprocessed vehicle abnormal driving data can be obtained through the dynamic information database, which represents the proportion of the preprocessed vehicle violation data and the preprocessed vehicle abnormal driving data in the vehicle driving evaluation coefficient. The weight factors of the preprocessed vehicle violation data and the weight factors of the preprocessed vehicle abnormal driving data can also be obtained through mapping relationships. For example, a mapping set of vehicle violation data and vehicle abnormal driving data and their corresponding weights is established respectively through the relationship between the vehicle violation data and vehicle abnormal driving data in the historical data and the dynamic information. The weight factors of the preprocessed vehicle violation data and the weight factors of the preprocessed vehicle abnormal driving data corresponding to the mapping set are obtained by inputting real-time vehicle violation data and vehicle abnormal driving data.

[0043] Furthermore, the specific analysis process of analyzing the preprocessed vital sign capture data is: extracting features from the preprocessed vital sign capture data, assigning weight factors to the parameters in the preprocessed vital sign capture data, and obtaining a driving concentration evaluation coefficient based on the weight factors of the parameters in the preprocessed vital sign capture data and combining the parameters in the preprocessed vital sign capture data; applying the driving concentration evaluation coefficient to the real-time monitoring system to perform real-time evaluation of the vehicle's driving concentration.

[0044] In this embodiment, features are extracted from the preprocessed vital sign capture data, including breathing frequency, number of head movements, and grip strength on the steering wheel. The driving concentration evaluation coefficient is obtained according to the weight factors of the parameters in the preprocessed vital sign capture data combined with the parameters in the preprocessed vital sign capture data. The specific method for obtaining the driving concentration evaluation coefficient is:

[0045] ;

[0046] In the formula, Expressed as The driving concentration evaluation coefficient under each driving concentration time monitoring segment is used to evaluate the impact of the driver's physical signs on driving concentration under each driving concentration time monitoring segment. Several driving concentration time monitoring points are set for the physical sign capture process, and 20 driving concentration time monitoring points are regarded as a driving concentration time monitoring segment. , Indicates the total number of driving focus time monitoring segments, Expressed as The driver's breathing rate in each driving concentration time monitoring segment, Expressed as The driver's breathing rate in each driving concentration time monitoring segment, It is expressed as the driver's breathing rate during normal driving in a driving concentration time monitoring period. Expressed as The number of driver head movements in each driving focus time monitoring segment, Expressed as The number of driver head movements in each driving focus time monitoring segment, It is expressed as the number of head movements of the driver during normal driving in a driving concentration time monitoring segment. Expressed as The driver's grip on the steering wheel during the driving concentration time monitoring period, Expressed as The driver's grip on the steering wheel during the driving concentration time monitoring period, It is expressed as the driver's grip on the steering wheel during normal driving in a driving concentration time monitoring period. It is represented as the weight factor of the driver's breathing rate in the vital sign capture data, It is expressed as the weight factor of the driver's head movement frequency in the vital signs capture data, It is represented as the weight factor of the driver’s grip on the steering wheel in the vital signs capture data, Represents a natural constant.

[0047] Using physical relationship analysis, breathing frequency will have a negative impact on the driver's driving concentration. The driver's breathing rate in the first driving concentration time monitoring period is greater than that in the If the driver's breathing rate in a driving concentration time monitoring segment is lower than that in a driving concentration time monitoring segment, it means that the driver may be in a fatigue state, and the driving concentration evaluation coefficient in this driving concentration time monitoring segment is reduced.

[0048] In a specific embodiment, the driver's breathing frequency during normal driving in a driving concentration time monitoring segment, the number of driver's head movements during normal driving in a driving concentration time monitoring segment, and the driver's grip on the steering wheel during normal driving in a driving concentration time monitoring segment can be directly obtained through historical data, and the average processing results of the driver's breathing frequency during normal driving in a driving concentration time monitoring segment, the number of driver's head movements during normal driving in a driving concentration time monitoring segment, and the driver's grip on the steering wheel during normal driving in a driving concentration time monitoring segment are taken as the driver's breathing frequency during normal driving in a driving concentration time monitoring segment, the number of driver's head movements during normal driving in a driving concentration time monitoring segment, and the driver's grip on the steering wheel during normal driving in a driving concentration time monitoring segment.

[0049] Using physical relationship analysis, the larger the driving concentration assessment coefficient is, the more focused the vehicle driver is, and the larger the vehicle driving assessment coefficient is. When the driving concentration assessment coefficient is smaller, it means that the less focused the vehicle driver is, and the smaller the vehicle driving assessment coefficient is.

[0050] The weight factor of light intensity in the vital sign capture data, the weight factor of ambient atmospheric pressure in the vital sign capture data and the weight factor of ambient visibility in the vital sign capture data can be obtained through the dynamic information database, which represents the proportion of light intensity, ambient atmospheric pressure and ambient visibility in the vital sign capture data in the vehicle driving evaluation coefficient. The weight factor of light intensity in the vital sign capture data, the weight factor of ambient atmospheric pressure in the vital sign capture data and the weight factor of ambient visibility in the vital sign capture data can also be obtained through mapping relationships. For example, a mapping set of light intensity, ambient atmospheric pressure and ambient visibility and their corresponding weights is established respectively through the relationship between vehicle violation data and vehicle abnormal driving data in historical data and dynamic information. The weight factor of light intensity in the vital sign capture data, the weight factor of ambient atmospheric pressure in the vital sign capture data and the weight factor of ambient visibility in the vital sign capture data in the mapping set corresponding to the real-time light intensity, ambient atmospheric pressure and ambient visibility are obtained.

[0051] Furthermore, the specific comparison process of comparing the driving concentration evaluation coefficient with the threshold of environmental impact on driving concentration is as follows: if the driving concentration evaluation coefficient fluctuates, the system increases the sensitivity of the camera and the sensor through the adjustment module; if the driving concentration evaluation coefficient does not fluctuate, the threshold of environmental impact on driving concentration is obtained from the database, and the driving concentration evaluation coefficient is compared with the threshold of environmental impact on driving concentration; if the driving concentration evaluation coefficient is less than the threshold of environmental impact on driving concentration, the system analyzes the pre-processed vital sign capture data in real time; if the driving concentration evaluation coefficient is greater than or equal to the threshold of environmental impact on driving concentration, the system activates the seat vibration reminder and issues a voice reminder through the adjustment module, and further evaluates the driving concentration evaluation coefficient in combination with the vehicle driving evaluation coefficient.

[0052] In this embodiment, the system monitors the change of the driving concentration evaluation coefficient in real time, compares two adjacent driving concentration evaluation coefficients, and determines that the difference between the ratio and 1 is greater than 0.1, and determines that it is fluctuating, otherwise it is determined that it is not fluctuating. If the coefficient fluctuation is detected, the system automatically increases the sensitivity of the camera and the sensor through the adjustment module to obtain more accurate vital sign capture data. If the driving concentration evaluation coefficient is stable, the system obtains the preset threshold value of the impact of the environment on driving concentration from the database, and compares the current driving concentration evaluation coefficient with the threshold value. If the driving concentration evaluation coefficient is less than the threshold value, the system analyzes the pre-processed vital sign capture data in real time to determine whether the vital sign capture data is accurate. If the driving concentration evaluation coefficient is greater than or equal to the threshold value, the system starts the seat vibration reminder and issues a voice reminder through the adjustment module to warn the driver that the harsh environment has a negative impact on the accuracy of the vital sign capture data. Regardless of whether the driving concentration evaluation coefficient fluctuates and whether the driving concentration evaluation coefficient is greater than the threshold value, the system combines the driving concentration evaluation coefficient with the vehicle driving evaluation coefficient to perform a more comprehensive vehicle driving state evaluation. By adjusting the sensitivity of the camera and the sensor, the system can more accurately monitor the driver's concentration state.

[0053] Furthermore, the specific comparison process of comparing the vehicle's dangerous driving comprehensive assessment coefficient with the vehicle's dangerous driving comprehensive assessment threshold is as follows: when the driving concentration assessment coefficient does not fluctuate, the vehicle driving assessment coefficient and the driving concentration assessment coefficient are comprehensively evaluated to obtain the vehicle's dangerous driving comprehensive assessment coefficient, the vehicle's dangerous driving comprehensive assessment threshold is obtained from the database, the vehicle's dangerous driving comprehensive assessment coefficient is compared with the vehicle's dangerous driving comprehensive assessment threshold, if the vehicle's dangerous driving comprehensive assessment coefficient is less than the vehicle's dangerous driving comprehensive assessment threshold, the vehicle's driving is monitored and evaluated in real time, if the vehicle's dangerous driving comprehensive assessment coefficient is greater than or equal to the vehicle's dangerous driving comprehensive assessment threshold, the system automatically activates driving assistance functions including lane keeping assist and automatic emergency braking assist through the adjustment module, and compares the vehicle's dangerous driving comprehensive assessment coefficient with the vehicle's dangerous driving comprehensive assessment threshold again, if the vehicle's dangerous driving comprehensive assessment coefficient is greater than or equal to the vehicle's dangerous driving comprehensive assessment threshold, the system connects to the vehicle's intelligent driving system through the adjustment module to take over the car, stops the vehicle on the side of the road, sends a push to the cloud center through the app, and contacts the emergency contacts stored in advance.

[0054] In this embodiment, when the vehicle driving assessment coefficient and the driving concentration assessment coefficient are stable, the system conducts a comprehensive assessment of the two coefficients to obtain a comprehensive assessment coefficient of dangerous driving of the vehicle. The system obtains a preset comprehensive assessment threshold of dangerous driving of the vehicle from the database, and compares the comprehensive assessment coefficient of dangerous driving of the vehicle with the threshold. If the comprehensive assessment coefficient of dangerous driving of the vehicle is less than the threshold, the system continues to monitor and evaluate the vehicle driving in real time to prevent the threshold comparison from being an error. If the comprehensive assessment coefficient of dangerous driving of the vehicle is greater than or equal to the threshold, the system automatically activates driving assistance functions such as lane keeping assist and automatic emergency braking assist through the adjustment module. The system compares the comprehensive assessment coefficient of dangerous driving of the vehicle with the threshold again. If the coefficient is still greater than or equal to the threshold, the system connects the intelligent driving system of the vehicle through the adjustment module, takes over the car and stops the vehicle on the roadside. The system sends a push to the cloud center through the app and notifies the emergency contact. Through real-time monitoring and automatic intervention, the system can significantly improve driving safety and reduce the occurrence of traffic accidents.

[0055] Furthermore, the specific control method of controlling the vehicle through the adjustment module is: when the vehicle is detected to be speeding, the system reduces the speed by limiting the throttle response or enabling the cruise control system. When sudden braking is detected, the system activates the anti-lock braking system or the dynamic stability control system to optimize braking stability. For lane departure behavior, the system makes slight corrections to the steering wheel through the electric power steering system to help the vehicle return to the correct lane. If the driving concentration assessment coefficient is greater than or equal to the threshold of the environment's impact on driving concentration, the system activates the seat vibration reminder and issues a voice reminder through the adjustment module. If the vehicle's comprehensive dangerous driving assessment coefficient is greater than or equal to the vehicle's comprehensive dangerous driving assessment threshold, the system connects to the vehicle's intelligent driving system through the adjustment module to take over the car, stops the vehicle on the side of the road, and sends a push to the cloud center through the app to contact the emergency contacts stored in advance.

[0056] In this embodiment, when the system detects that the vehicle is speeding, the system automatically limits the throttle response so that even if the driver steps on the accelerator, the vehicle will not accelerate too violently, or activates the cruise control system to automatically adjust the vehicle speed to the set safe speed. When an emergency braking behavior is detected, the system automatically activates the anti-lock braking system to prevent the wheels from locking and improve braking efficiency, or activates the dynamic stability control system to optimize braking stability and prevent the vehicle from losing control. When the vehicle deviates from the lane, the system slightly corrects the steering wheel through the electric power steering system to assist the driver in guiding the vehicle back to the correct lane. If the driving concentration assessment coefficient is greater than or equal to the threshold of the impact of the environment on driving concentration, the system starts the seat vibration reminder to remind the driver to pay attention through physical touch, and issues a voice reminder to remind the driver to stay focused through hearing. If the vehicle's comprehensive evaluation coefficient of dangerous driving is greater than or equal to the comprehensive evaluation threshold of dangerous driving, the system connects to the vehicle's intelligent driving system through the adjustment module, takes over the car operation, and safely stops the vehicle on the roadside. It sends a push to the cloud center through the app to notify the emergency and contact the emergency contacts stored in advance to request assistance. When the system detects dangerous driving behavior, it can automatically take measures to intervene without the driver's manual operation, which significantly reduces the risk of traffic accidents.

[0057] Furthermore, the specific method for obtaining the comprehensive evaluation coefficient of dangerous driving of a vehicle is as follows: ; ; In the formula, The comprehensive evaluation coefficient of dangerous driving of vehicles under each dangerous driving time monitoring period is used to evaluate the impact of vehicle driving and driving concentration on dangerous driving under each dangerous driving time monitoring period. The vehicle driving evaluation coefficient under each vehicle driving time monitoring section is used to evaluate the impact of vehicle violations and abnormal vehicle driving on vehicle driving under each vehicle driving time monitoring section. The driving concentration evaluation coefficient under each driving concentration time monitoring segment is used to evaluate the impact of the driving visual capture data on driving concentration under each driving concentration time monitoring segment. Expressed as the weight factor of the vehicle driving evaluation coefficient, Expressed as the weight factor of the driving focus evaluation coefficient, Expressed as The abnormal weather index under each abnormal weather time monitoring period is used to evaluate the impact of abnormal weather under each abnormal weather time monitoring period on dangerous driving control.

[0058] In this embodiment, several dangerous driving time monitoring points are set for comprehensive evaluation of dangerous driving of vehicles. , It indicates the total number of dangerous driving time monitoring points, and 20 dangerous driving time monitoring points are regarded as a dangerous driving time monitoring section.

[0059] In this embodiment, the weight factor of the vehicle driving evaluation coefficient and the weight factor of the driving concentration evaluation coefficient can be obtained through the dynamic information database, and a mapping set of the vehicle driving evaluation coefficient and the driving concentration evaluation coefficient and their corresponding weights is established respectively through the relationship between the vehicle driving evaluation coefficient and the driving concentration evaluation coefficient in the historical data and the dynamic information. By inputting the real-time vehicle driving evaluation coefficient and the driving concentration evaluation coefficient, the corresponding preprocessed weight factor of the vehicle driving evaluation coefficient and the weight factor of the driving concentration evaluation coefficient in the mapping set are obtained.

[0060] The specific method for obtaining the abnormal weather index is as follows:

[0061] ;

[0062] In the formula, Expressed as The abnormal weather index under each abnormal weather time monitoring segment is used to evaluate the impact of abnormal weather on dangerous driving control under each abnormal weather time monitoring segment. Several abnormal weather time monitoring points are set for vehicle monitoring, and 20 abnormal weather time monitoring points are regarded as a spatial monitoring segment. , Indicates the total number of abnormal weather time monitoring segments, Expressed as The light intensity during the abnormal weather monitoring period is Expressed as Atmospheric pressure during the abnormal weather monitoring period, Expressed as Road visibility during the abnormal weather monitoring period.

[0063] Using physical relationship analysis, the stronger the light intensity, the higher the visibility. In this formula, the light intensity is negatively correlated with the abnormal weather index, while in the driving concentration evaluation coefficient formula, the light intensity is positively correlated with the driving concentration evaluation coefficient. The abnormal weather index will also affect the light intensity in the driving concentration evaluation coefficient formula. The abnormal weather index will also affect the system's judgment on speeding, sudden acceleration, sudden braking and sharp turns. If the abnormal weather index is higher, the judgment range of speeding, sudden acceleration, sudden braking and sharp turns should be narrower. For example, when the abnormal weather index is 5, the threshold for judging speeding is 60km / h. If the abnormal weather index is 10, the threshold for judging speeding is 50km / h.

[0064] In a specific embodiment, examples of abnormal weather index data are shown in the following table.

[0065] Table 1 Abnormal weather index data example table

[0066]

[0067] pass Figure 2 From the data in Table 1, it can be seen that when the atmospheric pressure and road visibility remain constant, the greater the light intensity, the lower the abnormal weather index.

[0068] like Figure 3 As shown, it is a structural schematic diagram of a dangerous driving detection system based on big data provided by an embodiment of the present application. The dangerous driving detection system based on big data provided by an embodiment of the present application includes: a data collection module, a data analysis module, a threshold comparison module and an adjustment module: the data collection module is used to collect vehicle violation data, vehicle abnormal driving data and vital sign capture data through the on-board diagnostic system, and clean, denoise and normalize the vehicle violation data, vehicle abnormal driving data and vital sign capture data to obtain pre-processed vehicle violation data, pre-processed vehicle abnormal driving data and pre-processed vital sign capture data; the data analysis module is used to remove the pre-processed vehicle violation data, pre-processed vehicle abnormal driving data and pre-processed vital sign capture data according to the big data. The preprocessed vehicle violation data and the preprocessed vehicle abnormal driving data are analyzed to obtain the vehicle driving evaluation coefficient, and the preprocessed vital sign capture data are analyzed to obtain the driving concentration evaluation coefficient; Threshold comparison module: obtains the threshold value of the impact of the environment on driving concentration from the database, compares the driving concentration evaluation coefficient with the threshold value of the impact of the environment on driving concentration, obtains the vehicle dangerous driving comprehensive evaluation threshold from the database, and compares the vehicle dangerous driving comprehensive evaluation coefficient with the vehicle dangerous driving comprehensive evaluation threshold; Adjustment module: used to control the vehicle through the adjustment module according to the comparison result of the vehicle dangerous driving comprehensive evaluation coefficient and the vehicle dangerous driving comprehensive evaluation threshold.

[0069] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0070] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0071] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0072] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0073] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0074] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A dangerous driving detection method based on big data, characterized in that: The following steps are involved: Collecting vehicle violation data, vehicle abnormal driving data and vital sign capture data through an on-board diagnostic system, cleaning, denoising and normalizing the vehicle violation data, vehicle abnormal driving data and vital sign capture data to obtain pre-processed vehicle violation data, pre-processed vehicle abnormal driving data and pre-processed vital sign capture data; According to the big data, outliers of the pre-processed vehicle violation data, the pre-processed vehicle abnormal driving data and the pre-processed vital signs captured data are removed, the pre-processed vehicle violation data and the pre-processed vehicle abnormal driving data are analyzed to obtain the vehicle driving evaluation coefficient, and the pre-processed vital signs captured data are analyzed to obtain the driving concentration evaluation coefficient; Obtaining a threshold value of the impact of the environment on driving concentration from a database, comparing a driving concentration evaluation coefficient with the threshold value of the impact of the environment on driving concentration, and regulating the vehicle through a regulating module according to the comparison result of the driving concentration evaluation coefficient and the threshold value of the impact of the environment on driving concentration; Comprehensively evaluate the vehicle driving assessment coefficient and the driving concentration assessment coefficient to obtain the vehicle dangerous driving comprehensive assessment coefficient, obtain the vehicle dangerous driving comprehensive assessment threshold from the database, compare the vehicle dangerous driving comprehensive assessment coefficient with the vehicle dangerous driving comprehensive assessment threshold, and regulate the vehicle through the adjustment module according to the comparison result of the vehicle dangerous driving comprehensive assessment coefficient and the vehicle dangerous driving comprehensive assessment threshold; The specific analysis process of analyzing the preprocessed vehicle violation data and the preprocessed vehicle abnormal driving data to obtain the vehicle driving evaluation coefficient is as follows: extracting features from the preprocessed vehicle violation data and the preprocessed vehicle abnormal driving data, dividing the allocation weight factors of the preprocessed vehicle violation data and the preprocessed vehicle abnormal driving data, combining the preprocessed vehicle violation data and the preprocessed vehicle abnormal driving data according to the weight factors of the preprocessed vehicle violation data and the preprocessed vehicle abnormal driving data to obtain the vehicle driving evaluation coefficient, and applying the vehicle driving evaluation coefficient to the real-time monitoring system to perform real-time evaluation of the vehicle driving status; The specific analysis process of analyzing the preprocessed vital sign capture data is as follows: extracting features from the preprocessed vital sign capture data, assigning weight factors to parameters in the preprocessed vital sign capture data, obtaining a driving concentration evaluation coefficient based on the weight factors of the parameters in the preprocessed vital sign capture data combined with the parameters in the preprocessed vital sign capture data, applying the driving concentration evaluation coefficient to the real-time monitoring system, and performing real-time evaluation of the vehicle's driving concentration; The specific method for obtaining the driving concentration evaluation coefficient is as follows: ; In the formula, Expressed as The driving concentration evaluation coefficient under each driving concentration time monitoring segment is used to evaluate the impact of the driver's physical signs on driving concentration under each driving concentration time monitoring segment. Several driving concentration time monitoring points are set for the physical sign capture process, and 20 driving concentration time monitoring points are regarded as a driving concentration time monitoring segment. , Indicates the total number of driving focus time monitoring segments, Expressed as The driver's breathing rate in each driving concentration time monitoring segment, Expressed as The driver's breathing rate in each driving concentration time monitoring segment, It is expressed as the driver's breathing rate during normal driving in a driving concentration time monitoring period. Expressed as The number of driver head movements in each driving focus time monitoring segment, Expressed as The number of driver head movements in each driving focus time monitoring segment, It is expressed as the number of driver's head movements during normal driving in a driving concentration time monitoring segment. Expressed as The driver's grip on the steering wheel during the driving concentration time monitoring period, Expressed as The driver's grip on the steering wheel during the driving concentration time monitoring period, It is expressed as the driver's grip on the steering wheel during normal driving in a driving concentration time monitoring period. It is represented as the weight factor of the driver's breathing rate in the vital sign capture data, It is expressed as the weight factor of the driver's head movement frequency in the vital signs capture data, It is represented as the weight factor of the driver’s grip on the steering wheel in the vital signs capture data, Represents a natural constant.

2. A dangerous driving detection method based on big data as claimed in claim 1, characterized in that: The specific collection process of collecting vehicle violation data, vehicle abnormal driving data and vital sign capture data through the on-board diagnostic system is as follows: When the vehicle is started, the on-board diagnostic system automatically initializes, monitors the vehicle's speed in real time, compares it with the speed limit of the road, records the number of speeding, records the number of red light violations through GPS positioning and traffic rules database, and monitors in real time whether the vehicle is driving in the wrong direction through the front camera. The vehicle violation data includes the number of speeding, the number of red light violations, and whether the vehicle is driving in the wrong direction. Vital sign capture data is collected through the breathing sensor on the seat, the camera and the grip sensor on the steering wheel. The vital sign capture data includes breathing rate, number of head movements and grip force on the steering wheel; Optical sensors are installed inside the vehicle to capture visibility, current light intensity and atmospheric pressure in real time. The vital signs captured data include visibility, current light intensity and atmospheric pressure.

3. The dangerous driving detection method based on big data as claimed in claim 1, characterized in that: The specific process of removing the abnormal values ​​of the pre-processed vehicle violation data, the pre-processed vehicle abnormal driving data and the pre-processed vital sign capture data according to the big data is as follows: The preprocessed vehicle violation data, the preprocessed vehicle abnormal driving data and the preprocessed vital sign capture data are arranged in order from large to small, and a clustering algorithm is used to identify outliers of the preprocessed vehicle violation data, the preprocessed vehicle abnormal driving data and the preprocessed vital sign capture data, and the outliers are marked as outliers. The outliers are replaced with the average value of the previous and next data. If there is no average value of the previous and next data, the outliers marked as outliers are directly deleted.

4. The dangerous driving detection method based on big data as claimed in claim 1, characterized in that: The specific method for obtaining the vehicle driving evaluation coefficient is: ； In the formula, Expressed as The vehicle driving evaluation coefficient under each vehicle driving time monitoring section is used to evaluate the impact of vehicle violations and abnormal vehicle driving on vehicle driving under each vehicle driving time monitoring section. Several vehicle driving time monitoring sections are set for vehicle monitoring. , Indicates the total number of vehicle driving time monitoring segments, and 20 vehicle driving time monitoring points are regarded as a driving time monitoring segment. Expressed as The number of speeding times in each vehicle driving time monitoring section, Expressed as The number of red light violations in each vehicle driving time monitoring segment, Expressed as Whether the vehicle is driving in the wrong direction during the driving time monitoring period, if so, If you don't go against =1, Expressed as The number of sudden accelerations in each vehicle driving time monitoring segment, Expressed as The number of emergency brakes in each vehicle driving time monitoring section, Expressed as The number of sharp turns in each vehicle driving time monitoring segment, Represented as the weight factor of the preprocessed vehicle violation data, It is represented as the weight factor of the preprocessed abnormal driving data of the vehicle.

5. The dangerous driving detection method based on big data as claimed in claim 1, characterized in that: The specific comparison process of comparing the driving concentration evaluation coefficient with the threshold of the impact of the environment on driving concentration is as follows: If the driving concentration assessment coefficient fluctuates, the system will increase the sensitivity of the camera and sensor through the adjustment module. If the driving concentration assessment coefficient does not fluctuate, the threshold value of the impact of the environment on driving concentration is obtained from the database, and the driving concentration assessment coefficient is compared with the threshold value of the impact of the environment on driving concentration. If the driving concentration assessment coefficient is less than the threshold value of the impact of the environment on driving concentration, the system will analyze the pre-processed vital sign capture data in real time. If the driving concentration assessment coefficient is greater than or equal to the threshold value of the impact of the environment on driving concentration, the system will activate the seat vibration reminder and issue a voice reminder through the adjustment module, and further evaluate the driving concentration assessment coefficient in combination with the vehicle driving assessment coefficient.

6. The dangerous driving detection method based on big data as claimed in claim 1, characterized in that: The specific comparison process of comparing the vehicle dangerous driving comprehensive assessment coefficient with the vehicle dangerous driving comprehensive assessment threshold is as follows: When the driving concentration assessment coefficient does not fluctuate, the vehicle driving assessment coefficient and the driving concentration assessment coefficient are comprehensively evaluated to obtain the vehicle dangerous driving comprehensive assessment coefficient, the vehicle dangerous driving comprehensive assessment threshold is obtained from the database, the vehicle dangerous driving comprehensive assessment coefficient is compared with the vehicle dangerous driving comprehensive assessment threshold, if the vehicle dangerous driving comprehensive assessment coefficient is less than the vehicle dangerous driving comprehensive assessment threshold, the vehicle driving is monitored and evaluated in real time, if the vehicle dangerous driving comprehensive assessment coefficient is greater than or equal to the vehicle dangerous driving comprehensive assessment threshold, the system automatically activates the driving assistance functions including lane keeping assist and automatic emergency braking assist through the adjustment module, and compares the vehicle dangerous driving comprehensive assessment coefficient with the vehicle dangerous driving comprehensive assessment threshold again, if the vehicle dangerous driving comprehensive assessment coefficient is greater than or equal to the vehicle dangerous driving comprehensive assessment threshold, the system connects to the vehicle's intelligent driving system through the adjustment module to take over the car, stops the vehicle on the side of the road, sends a push to the cloud center through the app, and contacts the emergency contacts stored in advance.

7. The dangerous driving detection method based on big data as claimed in claim 1, characterized in that: The specific control method of controlling the vehicle by the control module is: When the vehicle is detected to be speeding, the system reduces the speed by limiting the throttle response or enabling the cruise control system. When sudden braking is detected, the system activates the anti-lock braking system or the dynamic stability control system to optimize braking stability. For lane departure behavior, the system makes slight corrections to the steering wheel through the electric power steering system to help the vehicle return to the correct lane. If the driving concentration assessment coefficient is greater than or equal to the threshold of the environment's impact on driving concentration, the system activates the seat vibration reminder and issues a voice reminder through the adjustment module. If the vehicle's comprehensive dangerous driving assessment coefficient is greater than or equal to the vehicle's comprehensive dangerous driving assessment threshold, the system connects to the vehicle's intelligent driving system through the adjustment module to take over the car, stop the vehicle on the side of the road, and send a push to the cloud center through the app to contact the emergency contacts stored in advance.

8. The dangerous driving detection method based on big data as claimed in claim 1, characterized in that: The specific method for obtaining the comprehensive evaluation coefficient of dangerous driving of the vehicle is as follows: ； ; In the formula, Expressed as The comprehensive evaluation coefficient of dangerous driving of vehicles under each dangerous driving time monitoring period is used to evaluate the impact of vehicle driving and driving concentration on dangerous driving under each dangerous driving time monitoring period. Expressed as The vehicle driving evaluation coefficient under each vehicle driving time monitoring section is used to evaluate the impact of vehicle violations and abnormal vehicle driving on vehicle driving under each vehicle driving time monitoring section. Expressed as The driving concentration evaluation coefficient under each driving concentration time monitoring segment is used to evaluate the impact of the driving visual capture data on driving concentration under each driving concentration time monitoring segment. Expressed as the weight factor of the vehicle driving evaluation coefficient, Expressed as the weight factor of the driving focus evaluation coefficient, The abnormal weather index under each abnormal weather time monitoring period is used to evaluate the impact of abnormal weather under each abnormal weather time monitoring period on dangerous driving control.

9. A system using the dangerous driving detection method based on big data as claimed in any one of claims 1 to 8, characterized in that: Including data collection module, data analysis module, threshold comparison module and adjustment module: Data collection module: used to collect vehicle violation data, vehicle abnormal driving data and vital sign capture data through the on-board diagnostic system, clean, remove noise and normalize the vehicle violation data, vehicle abnormal driving data and vital sign capture data to obtain pre-processed vehicle violation data, pre-processed vehicle abnormal driving data and pre-processed vital sign capture data; Data analysis module: used to remove the abnormal values ​​of the pre-processed vehicle violation data, the pre-processed vehicle abnormal driving data and the pre-processed vital sign capture data according to the big data, analyze the pre-processed vehicle violation data and the pre-processed vehicle abnormal driving data to obtain the vehicle driving evaluation coefficient, and analyze the pre-processed vital sign capture data to obtain the driving concentration evaluation coefficient; Threshold comparison module: obtain the threshold of the impact of the environment on driving concentration from the database, compare the driving concentration assessment coefficient with the threshold of the impact of the environment on driving concentration, obtain the comprehensive assessment threshold of dangerous driving of the vehicle from the database, and compare the comprehensive assessment coefficient of dangerous driving of the vehicle with the comprehensive assessment threshold of dangerous driving of the vehicle; Adjustment module: used to adjust the vehicle through the adjustment module according to the comparison result of the vehicle's dangerous driving comprehensive assessment coefficient and the vehicle's dangerous driving comprehensive assessment threshold.

Images