Intelligent detection and control system for driving behaviors and habits

Abstract

The invention discloses an intelligent detection and control system for driving behaviors and habits, and relates to the technical field of automobile intelligent driving and safety monitoring. Comprising an infrared camera module which is installed at a vehicle driving position and used for capturing action images of legs and feet of a driver; the heating sheet module increases the detection precision of the camera in a low-light environment by increasing heat to the legs and the feet of the driver; and the image processor module is responsible for receiving the video images and carrying out deep learning algorithm processing so as to identify time, speed, depth or distance indexes of pedal treading of the driver. By monitoring and analyzing behaviors of the driver in real time, nonstandard driving operation can be found and corrected in time, and accidents caused by misoperation or intentional operation of the driver are reduced; compared with a traditional DMS, the problem that the DMS cannot observe the feet of the driver and cannot obtain the real operation of acceleration and deceleration of the driver is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of automobile intelligent driving and safety monitoring, and in particular to an intelligent detection and control system for driving behavior and habits. Background Art

[0002] Driving safety has always been the focus of social attention. According to statistics, a large number of traffic accidents are related to improper operation or irregular behavior of drivers. Traditional driving monitoring systems mainly rely on on-board cameras to capture external traffic conditions, but have limited monitoring and evaluation capabilities for the driver's own behavior. For example, DMS detects driver habits, generally through cameras to detect face, head, hands and other behaviors, and uses artificial intelligence algorithms to judge behavior; however, it is not possible to know whether the driver has accidentally stepped on the brake or accelerator.

[0003] After searching, the Chinese patent application number CN202021564102.3 discloses a bus driver status monitoring and dangerous driving monitoring system, including a heart rate monitoring bracelet, a DMS infrared camera, a DMS controller ECU, a bus vehicle control unit VCU, a heart rate comparator and a central processing module. The monitoring system in the above patent has the problem of not being able to monitor the driver's feet well. Summary of the invention

[0004] The purpose of the present invention is to solve the shortcomings of the prior art and to propose an intelligent detection and control system for driving behavior and habits.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions: An intelligent detection and control system for driving behavior and habits, comprising: Infrared camera module: installed at the driving position of the vehicle to capture the action images of the driver's legs and feet; Heater module: The heater module increases the camera's detection accuracy in low-light environments by adding heat to the driver's legs and feet; Image processor module: responsible for receiving video images and processing them with deep learning algorithms to identify the time, speed, depth or distance indicators of the driver's pedaling; Network transmission system: transmit image recognition information and vehicle information to the backend server in real time; Backend server and driver behavior management system: evaluate the matching degree of human-machine data through big data analysis, and provide analysis reports to vehicle managers; Vehicle-computer interface: capable of extracting acceleration and deceleration vehicle information and integrating it with image recognition data for analysis; Vehicle central control system: receives feedback from the backend server and reminds the driver to regulate operating behavior through audio, video, information and instructions.

[0006] Preferably, the heating plate module is installed at the driving position near the pedal or on the pedal, and transmits heat to the driver's legs through far-infrared waves.

[0007] Preferably: the image processor module comprises: Image preprocessing unit: performs noise reduction, enhancement and scale standardization on the original images collected by the camera; Feature extraction unit: uses computer vision algorithms to detect and identify the driver's leg and foot movements and extract key features; Behavior classification unit: applies machine learning algorithms to classify the extracted features and determine whether the driver's operation conforms to the preset safe driving mode; When the behavior classification unit performs classification judgment, the following steps are specifically included: S11: Feature extraction: Extract the driver's leg and foot motion features from the preprocessed image; represent the extracted features as a multi-dimensional feature vector ,in Indicates The value of a feature; S12: Model selection: Support vector machine is used for classification, and its optimization goal is expressed as: ; ; in, is the normal vector of the hyperplane, is the intercept, is the slack variable, is the regularization parameter, Yes Functions that map to high-dimensional spaces; S13: Training process: Use the labeled training data set to train the classifier; the training data set contains feature vectors and corresponding category labels ,in Indicates the category of the i-th sample; S14: Model application: The feature vector extracted in real time is input into the trained classifier to obtain the classification result of the driver's operation; if it is judged that it does not conform to the preset safe driving mode, a corresponding warning signal is generated.

[0008] Preferably: the network transmission system comprises: Data encoding module: responsible for encoding the driving behavior data generated by the image processor and the vehicle status data collected by the vehicle-machine interface to ensure the security and integrity of the data during transmission; Data transmission module: uses the vehicle communication network to transmit the encoded data to the background server in real time, and supports data upload and download; Data decoding module: After the backend server receives the data, it decodes it and restores it to a processable format.

[0009] Preferably: the background server and the driver behavior management system include: Data storage module: used to store the received driving behavior and vehicle status data; Data analysis module: Use machine learning and deep learning algorithms to analyze stored data, extract driving behavior characteristics, and evaluate the standardization and safety of driver operations; User management module: records driver identity information and driving habit files; Feedback generation module: Generates specific feedback reports and management suggestions based on data analysis results; The data analysis module comprises: Data Mining Unit: Use statistical learning and data mining techniques to discover hidden patterns and association rules in driving behavior data; Model training unit: Use historical data to train deep learning models to predict and evaluate driver behavior safety; Real-time analysis unit: processes incoming driving behavior data in real time, responds quickly and provides immediate behavior scoring and feedback.

[0010] Preferably, the data analysis module, when performing analysis and prediction, specifically includes the following steps: S21: Data preparation: Collect driving behavior history data, including driver operation data, vehicle status data, and corresponding safety assessment results; S22: Deep learning model: A hybrid architecture of convolutional neural network and recurrent neural network is selected for analysis, including: convolutional layer and recurrent layer. The convolutional layer extracts local features through filters. The formula is: ; ; Among them, * represents the convolution operation, and Respectively represent The convolution kernel weights and bias terms of the layer; The recurrent layer processes sequence data, and its state update formula is: ; ;

[0011] in, represents the hidden state at time t, is the input at time t, , and Represent the weight matrices from input to hidden layer, from hidden layer to hidden layer, and from hidden layer to output layer, respectively. and is the bias term; S23: Training process: Use historical data to train the model; S24: Real-time prediction: Input the real-time collected driving behavior data into the trained model to quickly derive the safety score of the current driving behavior and provide necessary feedback.

[0012] Preferably: the user management module includes: Identity authentication unit: manages driver authentication and authorization, ensuring data security and privacy; File maintenance unit: stores and maintains user's personal data and driving history records; Permission control unit: restricts access to and operations on sensitive data based on user roles and permission settings.

[0013] Preferably: the vehicle-machine interface includes: Vehicle data acquisition module: responsible for collecting real-time data from various sensors and control units of the vehicle; Data preprocessing module: cleans and formats the collected raw data, removes noise and extracts key information; Data fusion module: combines processed vehicle data with driving behavior data to form a comprehensive data set.

[0014] Preferably: the vehicle central control system comprises: Interactive interface module: provides an operating interface for the driver, displaying real-time information, warnings and prompts; Voice and video output module: convey information to the driver via voice or video; Control instruction execution module: automatically adjusts the vehicle's control parameters or directly controls it according to the analysis results and instructions of the background server.

[0015] Preferably, the control instruction execution module, when performing control, comprises the following steps: S31: Control signal analysis: analyzing the control instructions from the backend server or the driver input and converting them into specific control signals; S32: System coordination: Coordinate the braking system, steering system, and throttle control as needed, and distribute control signals to corresponding actuators; S33: Action execution: The actuator performs an action according to the received control signal; The vehicle central control system, when performing safety monitoring, includes the following steps: S41: Status monitoring: real-time monitoring of key status parameters of the vehicle; S42: Abnormality judgment: When the monitored status parameters exceed the normal operating range, it is judged that an abnormality occurs; S43: Emergency handling: Once an abnormality is detected, immediately switch back to manual control mode and warn the driver through the vehicle's central control system.

[0016] The beneficial effects of the present invention are: 1. The present invention can timely discover and correct irregular driving operations by real-time monitoring and analyzing the driver's behavior, thereby reducing accidents caused by driver's misoperation or intentional operation. Compared with the traditional DMS, the present invention solves the problem that DMS cannot observe the driver's feet and cannot obtain the driver's actual acceleration and deceleration operations.

[0017] 2. The present invention can determine the movements of the driver's legs and feet through image recognition; by combining the driver's movements with vehicle information, such as acceleration, deceleration, etc., it can determine whether the driver's behavior is normal; at the same time, it can determine whether the state of the vehicle itself is normal.

[0018] 3. The present invention can warm the driver's legs and feet through far-infrared wave heating; at the same time, it can solve the problem of unclear camera capture when the temperature is low or the driver is wearing too thick clothes; the clear infrared images of the driver's legs and feet after far-infrared wave heating can enhance the detection precision and accuracy of the infrared camera. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A flow chart of classification and judgment by a behavior classification unit of an intelligent detection and control system for driving behavior and habits proposed by the present invention; Figure 2 A flow chart for analyzing and predicting a data analysis module of an intelligent detection and control system for driving behavior and habits proposed by the present invention. DETAILED DESCRIPTION

[0020] The technical solution of the present invention is further described in detail below in conjunction with specific implementation methods.

[0021] Example

[0022] An intelligent detection and control system for driving behavior and habits, comprising: Infrared camera module: installed at the driving position of the vehicle to capture the action images of the driver's legs and feet; Heating module: The heating module increases the heat to the driver's legs and feet to improve the detection accuracy of the camera in low-light environments. The heating module is installed near the pedals at the driver's seat or on the pedals to transfer heat to the driver's legs through far-infrared waves. Image processor module: responsible for receiving video images and performing deep learning algorithm processing to identify key indicators such as the time, speed, depth or distance of the driver's pedaling; Network transmission system: transmit image recognition information and vehicle information to the backend server in real time; Backend server and driver behavior management system: evaluate the matching degree of human-machine data through big data analysis, and provide analysis reports to vehicle managers; Vehicle-computer interface: can extract vehicle information such as acceleration and deceleration, and integrate and analyze it with image recognition data; Vehicle central control system: Receives feedback from the background server and reminds the driver to standardize operating behavior through audio, video, information, and instructions. When necessary, it sends control instructions to take over some driving functions.

[0023] Wherein, the image processor module includes: Image preprocessing unit: performs noise reduction, enhancement and scale standardization on the original images collected by the camera to improve image quality for subsequent analysis; Feature extraction unit: uses computer vision algorithms to detect and identify the driver's leg and foot movements, extracting key features such as movement range, speed, and frequency; Behavior classification unit: Apply machine learning algorithms to classify the extracted features and determine whether the driver's operation complies with the preset safe driving mode.

[0024] The behavior classification unit performs classification judgment, which specifically includes the following steps: S11: Feature extraction: Extract the driver's leg and foot motion features from the preprocessed image; represent the extracted features as a multi-dimensional feature vector ,in Indicates The value of a feature; S12: Model selection: Support vector machine is used for classification, and its optimization goal is expressed as: ; ; in, is the normal vector of the hyperplane, is the intercept, is the slack variable, is the regularization parameter, Yes Functions that map to high-dimensional spaces; S13: Training process: Use the labeled training data set to train the classifier; the training data set contains feature vectors and corresponding category labels ,in Indicates the category of the i-th sample; S14: Model application: The feature vector extracted in real time is input into the trained classifier to obtain the classification result of the driver's operation; if it is judged that it does not conform to the preset safe driving mode, a corresponding warning signal is generated.

[0025] Wherein, the network transmission system comprises: Data encoding module: responsible for encoding the driving behavior data generated by the image processor and the vehicle status data collected by the vehicle-machine interface to ensure the security and integrity of the data during transmission; Data transmission module: uses the vehicle communication network (such as WiFi, 4G / 5G, Bluetooth, etc.) to transmit the encoded data to the background server in real time, and supports data upload and download; Data decoding module: After the backend server receives the data, it decodes it and restores it to a processable format for subsequent analysis.

[0026] Wherein, the background server and the driver behavior management system include: Data storage module: used to store the received driving behavior and vehicle status data, providing sufficient data support for big data analysis; Data analysis module: Use machine learning and deep learning algorithms to analyze stored data, extract driving behavior characteristics, and evaluate the standardization and safety of driver operations; User management module: records driver identity information and driving habit files to facilitate targeted management and training; Feedback Generation Module: Generates specific feedback reports and management suggestions based on data analysis results. This information can be used to guide drivers, adjust vehicle settings or perform maintenance.

[0027] Wherein, the data analysis module includes: Data Mining Unit: Use statistical learning and data mining techniques to discover hidden patterns and association rules in driving behavior data; Model training unit: Use historical data to train deep learning models so that they can accurately predict and evaluate the safety of driver behavior; Real-time analysis unit: processes incoming driving behavior data in real time, responds quickly and provides immediate behavior scoring and feedback.

[0028] The data analysis module, when performing analysis and prediction, specifically includes the following steps: S21: Data preparation: Collect driving behavior history data, including driver operation data, vehicle status data, and corresponding safety assessment results; S22: Deep learning model: A hybrid architecture of convolutional neural network and recurrent neural network is selected for analysis, including: convolutional layer and recurrent layer. The convolutional layer extracts local features through filters. The formula is: ; ; Among them, * represents the convolution operation, and Respectively represent The convolution kernel weights and bias terms of the layer; The recurrent layer processes sequence data, and its state update formula is: ; ; in, represents the hidden state at time t, is the input at time t, , and Represent the weight matrices from input to hidden layer, from hidden layer to hidden layer, and from hidden layer to output layer, respectively. and is the bias term; S23: Training process: Use historical data to train the model; S24: Real-time prediction: Input the real-time collected driving behavior data into the trained model to quickly derive the safety score of the current driving behavior and provide necessary feedback.

[0029] Wherein, the user management module includes: Identity authentication unit: responsible for managing the driver's identity authentication and authorization, ensuring data security and privacy; File maintenance unit: stores and maintains user's personal data and driving history records; Permission control unit: restricts access to and operations on sensitive data based on user roles and permission settings.

[0030] Wherein, the feedback generation module includes: Report generation unit: automatically generates detailed driving behavior reports based on data analysis results, including statistics, charts and improvement suggestions; Early warning unit: When potential unsafe behavior is detected, real-time early warning information is generated and transmitted to the vehicle's central control system.

[0031] Wherein, the vehicle-machine interface includes: Vehicle data acquisition module: responsible for collecting real-time data from various sensors and control units of the vehicle, such as speed, fuel consumption, braking status, etc.; Data preprocessing module: cleans and formats the collected raw data, removes noise and extracts key information; Data fusion module: combines processed vehicle data with driving behavior data to form a comprehensive data set for further analysis.

[0032] Wherein, the vehicle data collection module includes: Sensor data acquisition unit: obtains real-time data from various sensors of the vehicle (such as speed sensor, gyroscope, fuel consumption meter); Vehicle status monitoring unit: real-time monitoring of the basic status of the vehicle, such as vehicle speed, engine speed, brake status, etc.; Data synchronization unit: ensures that the collected data is synchronized with the actual status of the vehicle to avoid information delay or confusion.

[0033] Wherein, the vehicle central control system includes: Interactive interface module: provides an operating interface for the driver, displaying real-time information, warnings and prompts; Voice and video output module: Give instructions or warnings to the driver through voice or video when necessary to improve the efficiency of information transmission; Control instruction execution module: automatically adjusts the vehicle's control parameters or directly controls certain functions, such as emergency braking, lane keeping, etc., based on the analysis results and instructions of the background server.

[0034] Wherein, the control instruction execution module, when performing control, includes the following steps: S31: Control signal analysis: analyzing the control instructions from the backend server or the driver input and converting them into specific control signals; S32: System coordination: distribute appropriate control signals to corresponding actuators according to different control systems (such as braking system, steering system, throttle control, etc.) that need to be coordinated; S33: Action execution: The actuator performs actions according to the received control signal, such as adjusting the braking force, changing the steering angle, etc., to achieve complex driving assistance operations.

[0035] The vehicle central control system, when performing safety monitoring, includes the following steps: S41: Status monitoring: real-time monitoring of key status parameters of the vehicle, such as vehicle speed, engine status, brake status, etc.; S42: Abnormality judgment: When the monitored status parameters exceed the normal operating range, it is judged that an abnormality occurs; S43: Emergency handling: Once an abnormality is detected, immediately switch back to manual control mode and warn the driver through the vehicle's central control system; if necessary, initiate emergency braking or other safety protection measures.

[0036] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. An intelligent detection and control system for driving behavior and habits, characterized in that: include: Infrared camera module: installed at the driving position of the vehicle to capture the action images of the driver's legs and feet; Heater module: The heater module increases the camera's detection accuracy in low-light environments by adding heat to the driver's legs and feet; Image processor module: responsible for receiving video images and processing them with deep learning algorithms to identify the time, speed, depth or distance indicators of the driver's pedaling; Network transmission system: transmit image recognition information and vehicle information to the backend server in real time; Backend server and driver behavior management system: evaluate the matching degree of human-machine data through big data analysis, and provide analysis reports to vehicle managers; Vehicle-computer interface: capable of extracting acceleration and deceleration vehicle information and integrating it with image recognition data for analysis; Vehicle central control system: receives feedback from the backend server and reminds the driver to regulate operating behavior through audio, video, information and instructions.

2. The intelligent detection and control system for driving behavior and habits according to claim 1, characterized in that: The heating plate module is installed at the driving position near the pedal or on the pedal, and transmits heat to the driver's legs through far infrared waves.

3. The intelligent detection and control system for driving behavior and habits according to claim 1, characterized in that: The image processor module comprises: Image preprocessing unit: performs noise reduction, enhancement and scale standardization on the original images collected by the camera; Feature extraction unit: uses computer vision algorithms to detect and identify the driver's leg and foot movements and extract key features; Behavior classification unit: applies machine learning algorithms to classify the extracted features and determine whether the driver's operation conforms to the preset safe driving mode; When the behavior classification unit performs classification judgment, the following steps are specifically included: S11: Feature extraction: Extract the driver's leg and foot motion features from the preprocessed image; represent the extracted features as a multi-dimensional feature vector ,in Indicates The value of a feature; S12: Model selection: Support vector machine is used for classification, and its optimization goal is expressed as: ; ; in, is the normal vector of the hyperplane, is the intercept, is the slack variable, is the regularization parameter, Yes Functions that map to high-dimensional spaces; S13: Training process: Use the labeled training data set to train the classifier; the training data set contains feature vectors and corresponding category labels ,in Indicates the category of the i-th sample; S14: Model application: The feature vector extracted in real time is input into the trained classifier to obtain the classification result of the driver's operation; if it is judged that it does not conform to the preset safe driving mode, a corresponding warning signal is generated.

4. The intelligent detection and control system for driving behavior and habits according to claim 1, characterized in that: The network transmission system comprises: Data encoding module: responsible for encoding the driving behavior data generated by the image processor and the vehicle status data collected by the vehicle-machine interface to ensure the security and integrity of the data during transmission; Data transmission module: uses the vehicle communication network to transmit the encoded data to the background server in real time, and supports data upload and download; Data decoding module: After the backend server receives the data, it decodes it and restores it to a processable format.

5. The intelligent detection and control system for driving behavior and habits according to claim 1, characterized in that: The background server and driver behavior management system include: Data storage module: used to store the received driving behavior and vehicle status data; Data analysis module: Use machine learning and deep learning algorithms to analyze stored data, extract driving behavior characteristics, and evaluate the standardization and safety of driver operations; User management module: records driver identity information and driving habit files; Feedback generation module: Generates specific feedback reports and management suggestions based on data analysis results; The data analysis module comprises: Data Mining Unit: Use statistical learning and data mining techniques to discover hidden patterns and association rules in driving behavior data; Model training unit: Use historical data to train deep learning models to predict and evaluate driver behavior safety; Real-time analysis unit: processes incoming driving behavior data in real time, responds quickly and provides immediate behavior scoring and feedback.

6. The intelligent detection and control system for driving behavior and habits according to claim 5, characterized in that: The data analysis module, when performing analysis and prediction, specifically includes the following steps: S21: Data preparation: Collect driving behavior history data, including driver operation data, vehicle status data, and corresponding safety assessment results; S22: Deep learning model: A hybrid architecture of convolutional neural network and recurrent neural network is selected for analysis, including: convolutional layer and recurrent layer. The convolutional layer extracts local features through filters. The formula is: ; ; Among them, * represents the convolution operation, and Respectively represent The convolution kernel weights and bias terms of the layer; The recurrent layer processes sequence data, and its state update formula is: ; ; in, represents the hidden state at time t, is the input at time t, , and Represent the weight matrices from input to hidden layer, from hidden layer to hidden layer, and from hidden layer to output layer, respectively. and is the bias term; S23: Training process: Use historical data to train the model; S24: Real-time prediction: Input the real-time collected driving behavior data into the trained model to obtain the safety score of the current driving behavior.

7. The intelligent detection and control system for driving behavior and habits according to claim 6, characterized in that: The user management module comprises: Identity authentication unit: manages driver authentication and authorization, ensuring data security and privacy; File maintenance unit: stores and maintains user's personal data and driving history records; Permission control unit: restricts access to and operations on sensitive data based on user roles and permission settings.

8. The intelligent detection and control system for driving behavior and habits according to claim 1, characterized in that: The vehicle-machine interface includes: Vehicle data acquisition module: responsible for collecting real-time data from various sensors and control units of the vehicle; Data preprocessing module: cleans and formats the collected raw data, removes noise and extracts key information; Data fusion module: combines processed vehicle data with driving behavior data to form a comprehensive data set.

9. The intelligent detection and control system for driving behavior and habits according to claim 1, characterized in that: The vehicle central control system comprises: Interactive interface module: provides an operating interface for the driver, displaying real-time information, warnings and prompts; Voice and video output module: convey information to the driver via voice or video; Control instruction execution module: automatically adjusts the vehicle's control parameters or directly controls it according to the analysis results and instructions of the background server.

10. An intelligent detection and control system for driving behavior and habits according to claim 9, characterized in that: The control instruction execution module, when performing control, comprises the following steps: S31: Control signal analysis: analyzing the control instructions from the backend server or the driver input and converting them into specific control signals; S32: System coordination: Coordinate the braking system, steering system, and throttle control as needed, and distribute control signals to corresponding actuators; S33: Action execution: The actuator performs an action according to the received control signal; The vehicle central control system, when performing safety monitoring, includes the following steps: S41: Status monitoring: real-time monitoring of key status parameters of the vehicle; S42: Abnormality judgment: When the monitored status parameters exceed the normal operating range, it is judged that an abnormality occurs; S43: Emergency handling: Once an abnormality is detected, immediately switch back to manual control mode and warn the driver through the vehicle's central control system.

Images