Real-time drunk driving early warning method and system, electronic equipment and storage medium

Abstract

The invention belongs to the technical field of drunk driving monitoring, and discloses a real-time drunk driving early warning method and system, electronic equipment and a storage medium, and the method comprises the steps: collecting physiological feature data, behavior pattern data and environment data of a driver through a vehicle-mounted camera and a sensor; preprocessing the collected data to obtain data features; constructing a neural network prediction model, and inputting the preprocessed data features to obtain a drunk driving risk score; and judging whether drunk driving occurs based on the drunk driving risk score and taking corresponding early warning and measures. According to the invention, potential drunk driving behaviors of the driver can be identified and warned in advance, and traffic safety is guaranteed.

Description

Technical Field

[0001] The present invention belongs to the technical field of drunk driving monitoring, and specifically relates to a real-time drunk driving warning method and system, an electronic device, and a storage medium. Background Art

[0002] Traditional drunk driving detection methods mainly rely on breath alcohol testers or blood tests to determine whether a driver is driving under the influence. Although these methods are effective, they have a lag. The breath alcohol detection method may, due to the different alcohol metabolism speeds of different people, result in some drivers having dangerous driving behaviors even when the alcohol concentration does not reach the limit standard. The blood test method is complex in operation, low in efficiency, and not easy to popularize. In addition, existing in-vehicle drunk driving warnings can only be carried out when the driver gets in the vehicle and starts the vehicle. Some drivers may also take evasive measures, such as using mouthwash to interfere with the test results, and it is impossible to take measures immediately after the driver drinks alcohol and prevent them from driving. Summary of the Invention

[0003] In view of the deficiencies of the prior art, the present invention discloses a real-time drunk driving warning method and system, an electronic device, and a storage medium.

[0004] The present invention is realized through the following technical solutions:

[0005] In a first aspect, a real-time drunk driving warning method includes:

[0006] Collecting driver physiological characteristic data, behavior pattern data, and environmental data through an in-vehicle camera and sensors;

[0007] Preprocessing the collected data to obtain data characteristics;

[0008] Constructing a neural network prediction model, and inputting the preprocessed data characteristics to obtain a drunk driving risk score;

[0009] Judging whether it is drunk driving based on the drunk driving risk score and taking corresponding warnings and measures.

[0010] In some embodiments, the in-vehicle camera and sensors include a high-precision in-vehicle camera, an alcohol sensor, a heart rate sensor, and a body temperature sensor.

[0011] In some embodiments, the physiological characteristic data includes respiratory rate, alcohol concentration, heart rate, and body temperature, the behavior characteristic data includes facial images and driver voices, and the environmental data includes vehicle speed and driving route.

[0012] In some embodiments, the preprocessing includes data cleaning, denoising, and extracting key features.

[0013] In some embodiments, face monitoring and feature extraction are performed on the facial image, and the features include eyes, mouth, and facial contour;

[0014] By recognizing the features, the expression features after drinking are obtained, including red face and smaller eyes.

[0015] In some embodiments, the driver's voice is recognized, and the speech feature vectors extracted include pitch, speech rate, and volume.

[0016] In some embodiments, the neural network prediction model includes a convolutional neural network model and a recurrent neural network model.

[0017] In a second aspect, a real-time drunk driving warning system includes:

[0018] A data acquisition module, configured to collect driver physiological characteristic data, behavior pattern data, and environmental data through an in-vehicle camera and sensors;

[0019] A data processing module, configured to preprocess the collected data to obtain data features;

[0020] An AI prediction module, configured to construct a neural network prediction model and input the preprocessed data features to obtain a drunk driving risk score;

[0021] A drunk driving warning module, configured to determine whether it is drunk driving according to the drunk driving risk score and make corresponding warnings and measures.

[0022] In a third aspect, a computer-readable storage medium stores one or more programs, and when the one or more programs are executed, the above method can be implemented.

[0023] In a fourth aspect, a device includes a processor, a communication interface, a memory, and a communication bus; the processor, the communication interface, and the memory communicate with each other through the communication bus; characterized in that the processor is configured to execute the program stored in the above computer-readable storage medium.

[0024] Compared with the prior art, the present invention has the following advantages:

[0025] 1. Drunk driving monitoring is performed by non-invasive means;

[0026] 2. Monitor drunk driving features from multiple aspects to ensure the accuracy of drunk driving test results;

[0027] 3. Monitor the driver's drinking situation in real time, achieve early warning, and avoid the driver's drunk driving behavior.

[0028] Other features and advantages of the present invention will be set forth in the following description, and in part will be obvious from the description, or may be learned by practice of the present invention. The objectives and other advantages of the present invention may be realized and attained by the structure particularly pointed out in the written description, claims, as well as the appended drawings. Brief Description of the Drawings

[0029] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or in the prior art, the following briefly introduces the accompanying drawings required for the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0030] Figure 1 is a schematic diagram of a real-time drunk driving warning method according to this embodiment.

[0031] Figure 2 is a schematic diagram of a real-time drunk driving warning system according to this embodiment.

[0032] Figure 3 is a schematic diagram of an electronic device according to this embodiment. Detailed Embodiments

[0033] The accompanying drawings are only for illustrative purposes and should not be construed as limiting the present patent; for better illustrating this embodiment, some components in the accompanying drawings are omitted, enlarged or reduced, and do not represent the size of the actual product; for those skilled in the art, it is understandable that some well-known structures and their descriptions in the accompanying drawings may be omitted. The positional relationships described in the accompanying drawings are only for illustrative purposes and should not be construed as limiting the present patent.

[0034] In some embodiments, as Figure 1 shown in a real-time drunk driving warning method, includes the following steps:

[0035] S1. Collect driver physiological characteristic data, behavior pattern data, and environmental data through in-vehicle cameras and sensors;

[0036] Specifically, the in-vehicle cameras and sensors include high-precision in-vehicle cameras, alcohol sensors, heart rate sensors, and body temperature sensors. The collected physiological characteristic data includes respiratory rate, alcohol concentration, heart rate, and body temperature. The behavior characteristic data includes facial images and driver voices. The environmental data includes vehicle speed and driving route.

[0037] S2. Preprocess the collected data to obtain data features;

[0038] Specifically, the preprocessing includes data cleaning, denoising, and extracting key features. Noise and outliers are removed through cleaning, and then the data is standardized to ensure that data collected by different sensors has the same dimension and range.

[0039] In some embodiments, image processing algorithms are used to perform face monitoring and feature extraction on the captured face images, and the features include eyes, mouth, and facial contours.

[0040] Furthermore, by recognizing the facial expressions and eye features of the driver, the expression features after drinking are obtained, including red face and smaller eyes.

[0041] Furthermore, by recognizing the driver's speech, speech feature vectors including pitch, speech rate, and volume are extracted, so as to analyze whether there are drunk-driving features such as slurred speech.

[0042] Furthermore, the changing trends of the driver's heart rate and body temperature data are analyzed to determine whether the driver is in an abnormal physiological state.

[0043] Furthermore, the driving speed and driving route status of the vehicle are analyzed to determine whether the driving state of the driver is abnormal.

[0044] S3. Build a neural network prediction model, and input the data features after the preprocessing to obtain a drunk-driving risk score;

[0045] Specifically, the neural network prediction model includes a convolutional neural network (CNN) model and a recurrent neural network (RNN) model. Design the corresponding model architecture, determine the number and type of the input layer, hidden layer, and output layer, and then use a large amount of historical drunk-driving case data and non-drunk-driving data as the training set to train the neural network prediction model. Evaluate the performance of the model through cross-validation, and adjust the model parameters accordingly to improve the accuracy and reliability of the prediction.

[0046] Furthermore, design the structure of the convolutional neural network (CNN), including:

[0047] An input layer that processes the image data captured by the on-vehicle camera, including the facial expressions and eye features of the driver, represented as corresponding three-dimensional matrices.

[0048] A convolutional layer that extracts local features from the image data through multiple convolutional kernels, and performs downsampling on the local features using max pooling or average pooling. The max pooling selects the maximum value in each pooling window as the output, and the average pooling calculates the average value of each pooling window as the output.

[0049] A pooling layer that performs downsampling on the feature map output by the convolutional layer through max pooling or average pooling, outputs a feature vector, and reduces the feature dimension.

[0050] Fully connected layer, which converts the vector features output by the pooling layer into a drunk driving risk score. Each neuron in the fully connected layer is connected to all neurons in the previous layer, and the output value is obtained through weighted summation and bias terms.

[0051] Furthermore, design the Recurrent Neural Network (RNN) structure, including:

[0052] Input layer, which normalizes or standardizes the collected physiological feature data including respiratory rate, blood alcohol concentration, heart rate and body temperature, and environmental feature data including vehicle speed and driving route.

[0053] Hidden layer, which memorizes the feature data processed by the input layer through recurrent units, including LSTM units (Long Short-Term Memory network) or GRU units (Gated Recurrent Unit), and then updates the memory state at the current moment according to the current memory state and the memory state at the previous moment. The LSTM unit includes three gating mechanisms: forget gate, input gate and output gate, which control the flow and forgetting of information through the gating mechanism. The GRU unit is a simplified version of LSTM, which combines the forget gate and the input gate into an update gate, reducing the number of parameters and computational complexity.

[0054] Output layer, which converts the output of the hidden layer into a drunk driving risk score, using activation functions including Sigmoid or Tanh to map the output value to a specific range.

[0055] S4. Judge whether it is drunk driving based on the drunk driving risk score and take corresponding warnings and measures.

[0056] Specifically, set a threshold for drunk driving determination. If the drunk driving risk score exceeds the threshold, the warning mechanism will be automatically triggered, and the driver's attention will be attracted by the alarm sound, lights, voice or warning text displayed on the large screen in the vehicle. If the driver ignores the warning and continues to drive, further safety measures will be triggered, including restricting vehicle startup, automatically decelerating or stopping. In addition, the warning information and the driver's physiological and behavioral feature data will be sent to the cloud server for further analysis and processing by relevant departments.

[0057] In some embodiments, as Figure 2 shown in a real-time drunk driving warning system, including:

[0058] Data acquisition module, which is used to collect the driver's physiological feature data, behavior pattern data and environmental data through on-vehicle cameras and sensors;

[0059] Data processing module, which is used to preprocess the collected data to obtain data features;

[0060] An AI prediction module, configured to build a neural network prediction model and obtain a drunk driving risk score by inputting the preprocessed data features;

[0061] A drunk driving warning module, configured to determine whether it is drunk driving based on the drunk driving risk score and issue a warning and take measures.

[0062] As Figure 3 shown, an embodiment of the present disclosure further provides an electronic device, including a processor, a communication interface, a memory, and a communication bus. Among them, the processor, the communication interface, and the memory communicate with each other through the communication bus.

[0063] The memory is a computer-readable storage medium for storing one or more programs.

[0064] The processor is configured to execute the programs stored in the computer-readable storage medium.

[0065] This computer-readable storage medium may be included in the device / apparatus described in the above embodiments; or it may exist alone without being assembled into the device / apparatus.

[0066] Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A real-time drunk driving warning method, characterized in that: include: Collect driver's physiological characteristics data, behavior pattern data and environmental data through on-board cameras and sensors; Preprocess the collected data to obtain data features; Constructing a neural network prediction model, inputting the preprocessed data features to obtain a drunk driving risk score; Based on the drunk driving risk score, determine whether the person is driving under the influence of alcohol and make corresponding warnings and measures.

2. A real-time drunk driving warning method according to claim 1, characterized in that: The vehicle-mounted camera and sensors include a high-precision vehicle-mounted camera, an alcohol sensor, a heart rate sensor and a body temperature sensor.

3. A real-time drunk driving warning method according to claim 1, characterized in that: The physiological characteristic data include breathing rate, alcohol concentration, heart rate and body temperature, the behavioral characteristic data include facial images and driver voice, and the environmental data include vehicle speed and driving route.

4. A real-time drunk driving warning method according to claim 1, characterized in that: The preprocessing includes data cleaning, denoising and extracting key features.

5. A real-time drunk driving warning method according to claim 3, characterized in that: Performing face detection and feature extraction on the facial image, wherein the features include eyes, mouth and facial contour; By identifying the features, the facial expression features of drunkenness are obtained, including reddening of the face and shrinking of the eyes.

6. A real-time drunk driving warning method according to claim 3, characterized in that: The driver's voice is recognized and a voice feature vector including pitch, speech speed and volume is extracted.

7. A real-time drunk driving warning method according to claim 1, characterized in that: The neural network prediction model includes a convolutional neural network model and a recurrent neural network model.

8. A real-time drunk driving warning system, characterized in that: include: A data acquisition module, used to collect driver's physiological characteristics data, behavior pattern data and environmental data through vehicle-mounted cameras and sensors; A data processing module is used to pre-process the collected data to obtain data features; An AI prediction module is used to construct a neural network prediction model and input the preprocessed data features to obtain a drunk driving risk score; The drunk driving warning module is used to determine whether the person is driving under the influence of alcohol based on the drunk driving risk score and to make corresponding warnings and measures.

9. A computer-readable storage medium, characterized in that: One or more programs are stored, and when the one or more programs are executed, the methods described in claims 1 to 8 can be implemented.

10. A device comprising a processor, a communication interface, a memory and a communication bus; the processor, the communication interface and the memory communicate with each other via the communication bus; characterized in that: The processor is used to execute the program stored in the computer-readable storage medium according to claim 9.

Images