Intelligent driving behavior monitoring equipment for automobile driver

Abstract

The invention relates to the related technical field of driving behavior monitoring, and discloses an intelligent driving behavior monitoring device for an automobile driver, which comprises a monitoring device, the monitoring device comprises a device shell and a monitor, the device shell is fixedly connected with the monitor, and a sensor assembly is arranged on the outer side surface of the monitor. A data preprocessing module, a data processing module, a real-time early warning module and a communication module are arranged in the monitor; during use, driving behaviors and states of a driver are collected through the sensor assembly on the monitor, reasonable analysis and judgment are performed through the data preprocessing module and the data processing module, so that bad driving behaviors are confirmed, and finally the bad driving behaviors are notified through the real-time early warning module. The driver can be reminded of the bad driving behavior in time in the driving process, the bad driving habit is prevented from being developed, and the occurrence rate of traffic accidents caused by the bad driving behavior can be effectively reduced.

Description

Technical Field

[0001] The present invention relates to the technical field related to driving behavior monitoring, and in particular to an intelligent driving behavior monitoring device for automobile drivers. Background Art

[0002] Both new and old car owners have some bad driving habits, such as controlling the steering wheel with one hand, smoking while driving, etc. However, most car owners are not aware of these bad driving behaviors before an accident occurs, which increases the incidence of traffic accidents and seriously endangers their own and others' lives and property safety.

[0003] Existing monitoring technology usually only records some vehicle operation data, but ignores the supervision of driver behavior. Some drivers may exhibit bad driving behaviors during driving. If these driving behaviors are not corrected in time, they will easily develop bad driving habits, which will be particularly troublesome to correct later. These bad habits can easily cause traffic accidents and seriously endanger the life and property safety of themselves and others. Summary of the invention

[0004] In order to overcome the above-mentioned defects of the prior art, the present invention provides an intelligent driving behavior monitoring device for automobile drivers to solve the problems existing in the above-mentioned background technology.

[0005] The present invention provides the following technical solution: an intelligent driving behavior monitoring device for automobile drivers, comprising a monitoring device, the monitoring device comprising a device housing and a monitor, the device housing and the monitor being fixedly connected, a sensor assembly being arranged on the outer surface of the monitor, a data preprocessing module, a data processing module, a real-time warning module and a communication module being arranged inside the monitor, the data preprocessing module being capable of preprocessing data collected by the sensor, including image enhancement, noise removal and data normalization, the data processing module being used for classifying and processing the data, and judging the current driving behavior, the real-time warning module being used for warning the current illegal driving behavior, the communication module being used for transmitting the processed data to a remote server or a cloud platform in real time, obtaining the driver's driving information through the sensor assembly when in use, processing the data through the data preprocessing module and the data processing module, and judging the driver's behavior, and warning the real-time warning module after judging the illegal driving behavior, and uploading the collected information to the remote server or the cloud platform through the communication module.

[0006] Furthermore, the sensor component includes a visual sensor, an infrared sensor, a fill light, a photoresistor sensor and a timer. The visual sensor uses a high-definition camera to capture video data of the driver's facial expressions, head posture and limb movements. The infrared sensor is used to detect the driver's body temperature. The timer is used to record a single driving time. The photoresistor sensor is used to detect the brightness inside the car, and the fill light is used for fill light.

[0007] When in use, the visual sensor is used to capture the driver's facial expression, head posture and body movement video data. The facial expression, head posture and body movement are processed and captured by the data preprocessing module and the data processing module to determine the current state of the driver's facial expression, head posture and body movement, analyze the driver's psychological state and driving state, and judge whether the current driving behavior is illegal. When there is no facial expression, slow reaction or exaggerated facial expression, chaotic behavior, etc., it is judged as illegal driving behavior. At the same time, bad driving behaviors such as controlling the steering wheel with one hand, driving and talking on the phone, and fatigue driving are also monitored. During the detection process, the brightness inside the car is detected by the photoresistor sensor. When the brightness inside the car is low, the fill light is turned on for illumination to improve the clarity of the visual sensor. The infrared sensor is used to detect the driver's body temperature. The color change of the driver's face is captured by the visual sensor. The blood volume pulse wave signal is extracted using the rPPG algorithm. The extracted PPG signal is analyzed and processed to calculate the driver's heart rate. The current driver's physiological state is judged by the body temperature and heart rate.

[0008] Furthermore, a deep learning model based on convolutional neural network (CNN) and recurrent neural network (RNN) is established inside the data processing module to recognize the driver's facial expressions, head posture and body movements, and compare them with the internal database to determine whether the current behavior constitutes an illegal driving behavior.

[0009] Establish a convolutional neural network CNN deep learning model:

[0010] Input layer: receives raw image data, which are images of the driver's facial expressions, head postures, or body movements;

[0011] Convolutional layer: extracts features from input data through multiple convolution kernels. The convolution operation slides the convolution kernel on the input data to generate feature maps, which contain local feature information of the input data (facial expressions, head postures, and body movements);

[0012] Activation layer: usually uses the nonlinear activation function ReLU to increase the nonlinear expression ability of the network, so that the network can learn more complex features;

[0013] Pooling layer: used to reduce the spatial dimension of the feature map, reduce the amount of calculation, and retain important feature information;

[0014] Fully connected layer: Flattens the feature map and connects it to a fully connected neural network layer for classification or regression tasks. In the fully connected layer, each neuron is connected to all neurons in the previous layer, and the output is calculated through weights and biases.

[0015] Output layer: Outputs the final result of the network, such as classification labels or regression values. In the driver feature recognition task, the output layer may output the category of facial expression (such as smile, anger, etc.), the angle of head posture (such as pitch angle, yaw angle, etc.), or the type of body movement (such as waving, nodding, etc.).

[0016] Establish a recurrent neural network RNN ​​deep learning model:

[0017] Input sequence: The images or video frames of the driver’s facial expressions, head postures, or body movements are used as input sequences. These sequence data need to contain temporal continuity information.

[0018] Hidden layer: The hidden layer of RNN has short-term memory capability, which can receive the hidden layer state at the previous moment and the input data at the current moment, and calculate the hidden layer state at the current moment through the nonlinear function tanh;

[0019] Output layer: Calculates the output sequence based on the hidden layer state. In the driver feature recognition task, the output layer may output the driver feature category or regression value corresponding to each frame of image or video frame.

[0020] The specific process during use is as follows:

[0021] Facial expression recognition: CNN is used to extract feature information from facial expression images, and then RNN is used to capture the temporal relationship of expression changes, thereby accurately recognizing the driver's facial expression;

[0022] Head posture estimation: CNN is used to extract key point position information in the head image, such as joint points and feature points, and then RNN is used to model the changes in key point positions over time, thereby estimating the driver's head posture angle;

[0023] Body movement recognition: CNN is used to extract feature information from body movement images, and RNN is used to capture the temporal relationship of movement changes. The extracted features are classified in combination with the classifier model, thereby achieving accurate recognition of the driver's body movements.

[0024] Furthermore, the real-time warning module includes a voice playback unit and an interaction unit. The voice broadcast unit can broadcast the current illegal driving behavior through the speaker, and the interaction unit can be used to change the voice broadcast content. It can replace the voice that the driver is more familiar with to avoid the sudden appearance of strange voices during driving and thus affect the thinking. At the same time, using familiar voices, especially the voices of relatives, can better persuade the illegal driving behavior.

[0025] Furthermore, the communication module can efficiently compress video and image data to reduce the amount of data during storage and transmission, and at the same time use the RSA encryption algorithm to encrypt and protect the transmitted data to ensure the security and privacy of the data.

[0026] Furthermore, an external cover is provided between the device housing and the monitor, and the external cover is movably connected to the device housing. A connecting device is fixedly connected to the side of the device housing, and a vertical driving device is rotatably connected to the side of the connecting device. A motor and a gear set are provided inside the vertical driving device.

[0027] Furthermore, the bottom of the vertical driving device is fixedly connected to a connecting platform, the bottom of the connecting platform is rotatably connected to a horizontal driving device, and a motor is arranged inside the horizontal driving device.

[0028] Furthermore, the bottom of the horizontal driving device is fixedly connected to a connecting base, and a balancing device is rotatably connected between the horizontal driving device and the connecting base. The balancing device is provided with a plurality of counterweights, which can rotate under the action of inertia when the vehicle moves or stops, thereby dispersing the force and protecting the entire device. The bottom of the connecting base is fixedly connected to a suction cup, which is used to fix it when in use.

[0029] Beneficial effects:

[0030] 1. The intelligent driving behavior monitoring device for automobile drivers collects the driver's driving behavior and status through the sensor components on the monitor when in use, and performs reasonable analysis and judgment through the data preprocessing module and the data processing module to confirm the bad driving behavior, and finally notifies the bad driving behavior through the real-time warning module. It can timely remind the driver of the bad driving behavior during driving, avoid forming bad driving habits, and effectively reduce the incidence of traffic accidents caused by bad driving behavior. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 This is a schematic diagram of the overall structure of an intelligent driving behavior monitoring device for automobile drivers proposed by the present invention;

[0032] Figure 2This is a structural schematic diagram of a vehicle driver intelligent driving behavior monitoring device proposed by the present invention;

[0033] Figure 3 This is a module schematic diagram of a vehicle driver intelligent driving behavior monitoring device proposed by the present invention.

[0034] In the figure: 1. connecting base; 2. balancing device; 3. horizontal driving device; 4. connecting platform; 5. vertical driving device; 6. connecting device; 7. monitoring device; 701. device housing; 702. monitor; 703. external cover; 704. photoresistor sensor; 705. fill light; 706. visual sensor; 707. infrared sensor. DETAILED DESCRIPTION

[0035] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0036] Embodiment 1

[0037] See also Figure 1-Figure 3 , an intelligent driving behavior monitoring device for automobile drivers, including a monitoring device 7, the monitoring device 7 includes a device shell 701 and a monitor 702, the device shell 701 and the monitor 702 are fixedly connected, a sensor component is arranged on the outer surface of the monitor 702, and a data preprocessing module, a data processing module, a real-time warning module and a communication module are arranged inside the monitor 702, the data preprocessing module can preprocess the data collected by the sensor, including image enhancement, noise removal and data normalization, the data processing module is used to classify and process the data, and judge the current driving behavior, the real-time warning module is used to warn the current illegal driving behavior, and the communication module is used to transmit the processed data to a remote server or cloud platform in real time, when in use, the driver's driving information is obtained through the sensor component, the data is processed through the data preprocessing module and the data processing module, and the driver's behavior is judged, and after it is judged as an illegal driving behavior, a warning will be issued through the real-time warning module, and the collected information will be uploaded to the remote server or cloud platform through the communication module.

[0038] The sensor assembly includes a visual sensor 706, an infrared sensor 707, a fill light 705, a photoresistor sensor 704 and a timer. The visual sensor 706 uses a high-definition camera to capture the driver's facial expression, head posture and body movement video data. The infrared sensor 707 is used to detect the driver's body temperature. The timer is used to record a single driving time. The photoresistor sensor 704 is used to detect the brightness inside the car, and the fill light 705 is used for fill light.

[0039] When in use, the visual sensor 706 is used to capture the driver's facial expression, head posture and body movement video data. The facial expression, head posture and body movement are processed and captured by the data preprocessing module and the data processing module to determine the current state of the driver's facial expression, head posture and body movement, analyze the driver's psychological state and driving state, and judge whether the current driving behavior is illegal. When there is no facial expression, slow reaction or exaggerated facial expression, chaotic behavior, etc., it is judged as illegal driving behavior. At the same time, bad driving behaviors such as controlling the steering wheel with one hand, driving and talking on the phone, and fatigue driving are also monitored. During the detection process, the photoresistor sensor 704 is used to detect the brightness inside the car. When the brightness inside the car is low, the fill light 705 is turned on for lighting to improve the clarity of the visual sensor. The infrared sensor 707 is used to detect the driver's body temperature. The color change of the driver's face is captured by the visual sensor 706. The blood volume pulse wave signal is extracted using the rPPG algorithm. The extracted PPG signal is analyzed and processed to calculate the driver's heart rate. The current driver's physiological state is judged by the body temperature and heart rate.

[0040] Embodiment 2

[0041] See also Figure 1-Figure 3 A deep learning model based on convolutional neural network (CNN) and recurrent neural network (RNN) is established inside the data processing module to recognize the driver’s facial expressions, head posture and body movements, and compare them with the internal database to determine whether the current behavior constitutes an illegal driving behavior.

[0042] Establish a convolutional neural network CNN deep learning model:

[0043] Input layer: receives raw image data, which are images of the driver's facial expressions, head postures, or body movements;

[0044] Convolutional layer: extracts features from input data through multiple convolution kernels. The convolution operation slides the convolution kernel on the input data to generate feature maps, which contain local feature information of the input data (facial expressions, head postures, and body movements);

[0045] Activation layer: usually uses the nonlinear activation function ReLU to increase the nonlinear expression ability of the network, so that the network can learn more complex features;

[0046] Pooling layer: used to reduce the spatial dimension of the feature map, reduce the amount of calculation, and retain important feature information;

[0047] Fully connected layer: Flattens the feature map and connects it to a fully connected neural network layer for classification or regression tasks. In the fully connected layer, each neuron is connected to all neurons in the previous layer, and the output is calculated through weights and biases.

[0048] Output layer: Outputs the final result of the network, such as classification labels or regression values. In the driver feature recognition task, the output layer may output the category of facial expression (such as smile, anger, etc.), the angle of head posture (such as pitch angle, yaw angle, etc.), or the type of body movement (such as waving, nodding, etc.).

[0049] Establish a recurrent neural network RNN ​​deep learning model:

[0050] Input sequence: The images or video frames of the driver’s facial expressions, head postures, or body movements are used as input sequences. These sequence data need to contain temporal continuity information.

[0051] Hidden layer: The hidden layer of RNN has short-term memory capability, which can receive the hidden layer state at the previous moment and the input data at the current moment, and calculate the hidden layer state at the current moment through the nonlinear function tanh;

[0052] Output layer: Calculates the output sequence based on the hidden layer state. In the driver feature recognition task, the output layer may output the driver feature category or regression value corresponding to each frame of image or video frame.

[0053] The specific process during use is as follows:

[0054] Facial expression recognition: CNN is used to extract feature information from facial expression images, and then RNN is used to capture the temporal relationship of expression changes, thereby accurately recognizing the driver's facial expression;

[0055] Head posture estimation: CNN is used to extract key point position information in the head image, such as joint points and feature points, and then RNN is used to model the changes in key point positions over time, thereby estimating the driver's head posture angle;

[0056] Body movement recognition: CNN is used to extract feature information from body movement images, and RNN is used to capture the temporal relationship of movement changes. The extracted features are classified in combination with the classifier model, thereby achieving accurate recognition of the driver's body movements.

[0057] The real-time warning module includes a voice playback unit and an interactive unit. The voice broadcast unit can broadcast the current illegal driving behavior through the speaker. The interactive unit can be used to change the voice broadcast content. It can replace the voice that the driver is more familiar with to avoid the sudden appearance of strange voices during driving and affect the thinking. At the same time, using familiar voices, especially the voices of relatives, can better persuade people against illegal driving behaviors.

[0058] The communication module can efficiently compress video and image data to reduce the amount of data during storage and transmission. At the same time, it uses the RSA encryption algorithm to encrypt and protect the transmitted data to ensure data security and privacy.

[0059] An external cover plate 703 is arranged between the device housing 701 and the monitor 702. The external cover plate 703 is movably connected to the device housing 701. A connecting device 6 is fixedly connected to the side of the device housing 701. The side of the connecting device 6 is rotatably connected to a vertical driving device 5. A motor and a gear set are arranged inside the vertical driving device 5.

[0060] The bottom of the vertical driving device 5 is fixedly connected to the connecting platform 4, and the bottom of the connecting platform 4 is rotatably connected to the horizontal driving device 3, and a motor is arranged inside the horizontal driving device 3.

[0061] The bottom of the horizontal driving device 3 is fixedly connected to a connecting base 1, and a balancing device 2 is rotatably connected between the horizontal driving device 3 and the connecting base 1. A plurality of counterweights are provided on the balancing device 2, which can rotate under the action of inertia when the vehicle moves or stops, thereby dispersing the force and protecting the entire device. The bottom of the connecting base 1 is fixedly connected to a suction cup, which is used to fix it when in use.

[0062] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent driving behavior monitoring device for a car driver, comprising a monitoring device (7), characterized in that: The monitoring device (7) comprises a device housing (701) and a monitor (702), wherein the device housing (701) and the monitor (702) are fixedly connected, a sensor assembly is arranged on the outer surface of the monitor (702), and a data preprocessing module, a data processing module, a real-time warning module and a communication module are arranged inside the monitor (702), wherein the data preprocessing module can preprocess the data collected by the sensor, including image enhancement, noise removal and data normalization, the data processing module is used to classify the data and judge the current driving behavior, the real-time warning module is used to warn the current illegal driving behavior, and the communication module is used to transmit the processed data to a remote server or cloud platform in real time, and when in use, the driver's driving information is obtained through the sensor assembly, the data is processed through the data preprocessing module and the data processing module, and the driver's behavior is judged, and after judging the illegal driving behavior, a warning is issued through the real-time warning module, and the collected information is uploaded to the remote server or cloud platform through the communication module.

2. The intelligent driving behavior monitoring device for a car driver according to claim 1 is characterized by: The sensor assembly comprises a visual sensor (706), an infrared sensor (707), a fill light (705), a photoresistor sensor (704) and a timer. The visual sensor (706) adopts a high-definition camera and is used to capture the driver's facial expression, head posture and body movement video data. The infrared sensor (707) is used to detect the driver's body temperature. The timer is used to record a single driving time. The photoresistor sensor (704) is used to detect the brightness inside the vehicle. The fill light (705) is used to provide fill light.

3. The intelligent driving behavior monitoring device for a car driver according to claim 1 is characterized by: A deep learning model based on convolutional neural network (CNN) and recurrent neural network (RNN) is established inside the data processing module to recognize the driver's facial expressions, head posture and body movements, and compare them with the internal database to determine whether the current behavior constitutes an illegal driving behavior.

4. The intelligent driving behavior monitoring device for a car driver according to claim 1 is characterized by: The real-time warning module includes a voice playback unit and an interaction unit. The voice broadcast unit can broadcast the current illegal driving behavior through audio, and the interaction unit can be used to change the voice broadcast content.

5. The intelligent driving behavior monitoring device for a car driver according to claim 1 is characterized by: The communication module can efficiently compress video and image data, reduce the amount of data during storage and transmission, and use the RSA encryption algorithm to encrypt and protect the transmitted data to ensure data security and privacy.

6. The intelligent driving behavior monitoring device for a car driver according to claim 1 is characterized by: An external cover plate (703) is provided between the device housing (701) and the monitor (702); the external cover plate (703) is movably connected to the device housing (701); a connecting device (6) is fixedly connected to the side of the device housing (701); a vertical driving device (5) is rotatably connected to the side of the connecting device (6); a motor and a gear set are provided inside the vertical driving device (5).

7. The intelligent driving behavior monitoring device for a car driver according to claim 6 is characterized by: The bottom of the vertical driving device (5) is fixedly connected to a connecting platform (4), the bottom of the connecting platform (4) is rotatably connected to a horizontal driving device (3), and a motor is arranged inside the horizontal driving device (3).

8. The intelligent driving behavior monitoring device for a car driver according to claim 7 is characterized by: The bottom of the horizontal driving device (3) is fixedly connected to a connecting base (1), a balancing device (2) is rotatably connected between the horizontal driving device (3) and the connecting base (1), a plurality of counterweights are provided on the balancing device (2), and a suction cup is fixedly connected to the bottom of the connecting base (1).

Images