CN114209325B - Driver fatigue behavior monitoring method, device, equipment and storage medium - Google Patents
Driver fatigue behavior monitoring method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN114209325B CN114209325B CN202111607058.9A CN202111607058A CN114209325B CN 114209325 B CN114209325 B CN 114209325B CN 202111607058 A CN202111607058 A CN 202111607058A CN 114209325 B CN114209325 B CN 114209325B
- Authority
- CN
- China
- Prior art keywords
- driver
- fatigue
- wavelet packet
- fatigue behavior
- target
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/16—Devices for psychotechnics; Testing reaction times ; Devices for evaluating the psychological state
- A61B5/18—Devices for psychotechnics; Testing reaction times ; Devices for evaluating the psychological state for vehicle drivers or machine operators
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/16—Devices for psychotechnics; Testing reaction times ; Devices for evaluating the psychological state
- A61B5/168—Evaluating attention deficit, hyperactivity
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/08—Testing, supervising or monitoring using real traffic
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B2503/00—Evaluating a particular growth phase or type of persons or animals
- A61B2503/20—Workers
- A61B2503/22—Motor vehicles operators, e.g. drivers, pilots, captains
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Developmental Disabilities (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Educational Technology (AREA)
- Psychology (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Biology (AREA)
- Veterinary Medicine (AREA)
- Public Health (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Child & Adolescent Psychology (AREA)
- Artificial Intelligence (AREA)
- Hospice & Palliative Care (AREA)
- Psychiatry (AREA)
- Evolutionary Computation (AREA)
- Social Psychology (AREA)
- Biophysics (AREA)
- Pathology (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Signal Processing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Traffic Control Systems (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention relates to the technical field of intelligent driving, and discloses a method, a device, equipment and a storage medium for monitoring fatigue behavior of a driver. The method comprises the following steps: monitoring interference of current actions of a driver on wireless signals to obtain channel state information; extracting characteristics of the channel state information to obtain signal characteristics; performing dimension reduction on the signal characteristics to obtain target characteristics; inputting the target characteristics into a pre-trained target fatigue behavior recognition model to obtain a classification result output by the target fatigue behavior recognition model; and judging whether the current action of the driver is the fatigue driving action according to the classification result, and obtaining the fatigue behavior monitoring result of the driver. By the method, the actions of the driver are monitored by combining wireless sensing, whether the actions of the driver are fatigue driving actions or not is identified by using the trained fatigue behavior identification model, so that the monitoring of fatigue driving is realized, the occurrence of fatigue driving is reduced, and the occurrence rate of road traffic safety accidents is reduced.
Description
Technical Field
The present invention relates to the field of intelligent driving technologies, and in particular, to a method, an apparatus, a device, and a storage medium for monitoring fatigue behavior of a driver.
Background
At present, fatigue driving of a driver is one of the most important factors causing road traffic safety accidents. When a driver drives in a fatigue state, the phenomena of gear shifting, untimely braking and the like can occur, the phenomena of slow action, forgetting to operate and the like can occur if the driver is heavy, the phenomenon of short sleep can occur even more seriously, the control capability of the vehicle is lost, and a series of safety accidents are caused. Therefore, the occurrence rate of road traffic safety accidents can be effectively reduced by reducing the occurrence rate of fatigue driving.
The prior method for monitoring the fatigue of the driver comprises two methods, wherein the first method is to utilize image processing analysis to monitor the fatigue of the driver, and judge whether the driver is in a fatigue state or not by analyzing the actions of the driver during driving, in particular according to the blink frequency, the nodding frequency, the yawning and the like of the driver, wherein the fatigue driving generally occurs at night, and the illumination intensity is insufficient during night driving, so that the recognition rate of the fatigue driving monitoring mode based on images is greatly reduced, and the method has the characteristic of insufficient universality. The second way is by analyzing the driver physiological signal such as: the changes of signals such as blood pressure, heart rate and pulse are used for judging whether the driver is in a fatigue state or not, and the driver is required to carry with him/her equipment comprising various sensors for acquiring physiological data of the driver in real time, so that the driving feeling of the driver or the driving operation of the driver can be influenced by the way, and unnecessary accidents are caused.
The foregoing is provided merely for the purpose of facilitating understanding of the technical solutions of the present invention and is not intended to represent an admission that the foregoing is prior art.
Disclosure of Invention
The invention mainly aims to provide a method, a device, equipment and a storage medium for monitoring fatigue behavior of a driver, which aim to solve the technical problem of how to reduce the occurrence of fatigue driving so as to reduce the occurrence rate of road traffic safety accidents.
To achieve the above object, the present invention provides a driver fatigue behavior monitoring method comprising the steps of:
monitoring interference of current actions of a driver on wireless signals to obtain channel state information;
extracting the characteristics of the channel state information to obtain signal characteristics;
performing dimension reduction on the signal characteristics to obtain target characteristics;
inputting the target characteristics into a pre-trained target fatigue behavior recognition model to obtain a classification result output by the target fatigue behavior recognition model;
and judging whether the current action of the driver is the fatigue driving action according to the classification result, and obtaining a fatigue behavior monitoring result of the driver.
Optionally, the extracting the characteristics of the channel state information to obtain signal characteristics includes:
carrying out wavelet packet decomposition on the channel state information, and extracting wavelet packet energy characteristics and wavelet packet coefficient statistical characteristics;
and splicing the wavelet packet energy characteristics and the wavelet packet coefficient statistical characteristics to obtain signal characteristics.
Optionally, the performing wavelet packet decomposition on the channel state information, extracting wavelet packet energy features and wavelet packet coefficient statistical features, includes:
performing multi-layer wavelet packet decomposition on the channel state information to obtain wavelet packet coefficients;
forming a wavelet packet coefficient statistical feature according to the mean value, standard deviation, variance and maximum value corresponding to the wavelet packet coefficient;
determining the signal energy duty ratio corresponding to each sub-band;
determining wavelet packet energy vectors of each layer according to the signal energy duty ratio;
and constructing wavelet packet energy characteristics according to the wavelet packet energy vectors of each layer.
Optionally, the step of reducing the dimension of the signal feature to obtain a target feature includes:
carrying out standardization processing on the signal characteristics to obtain a standardized sample matrix;
calculating a covariance matrix corresponding to the standardized sample matrix;
determining corresponding eigenvalues and eigenvectors according to the covariance matrix;
sorting the feature vectors according to the feature values to obtain a feature matrix;
and extracting a plurality of feature vectors from the feature matrix according to a preset accumulated contribution rate to obtain target features.
Optionally, before the monitoring the interference of the current action of the driver on the wireless signal and obtaining the channel state information, the method further includes:
monitoring interference of various fatigue driving actions of a driver on a wireless signal to obtain a plurality of sample channel state information;
respectively extracting the characteristics of the sample channel state information to obtain corresponding sample signal characteristics;
performing dimension reduction on each sample signal characteristic to obtain a corresponding target sample characteristic;
constructing a training set and a verification set according to a plurality of target sample characteristics and corresponding fatigue action labels;
training the initial fatigue behavior recognition model according to the training set to obtain a trained fatigue behavior recognition model;
verifying the trained fatigue behavior recognition model according to the verification set;
and when the verification is passed, obtaining a trained target fatigue behavior recognition model.
Optionally, the verifying the trained fatigue behavior recognition model according to the verification set includes:
identifying each target sample characteristic in the verification set according to the trained fatigue behavior identification model to obtain an identification result corresponding to each target sample characteristic in the verification set;
determining corresponding recognition rates according to recognition results corresponding to the features of each target sample in the verification set;
and determining whether the trained fatigue behavior recognition model passes verification according to the recognition rate, wherein when the recognition rate is larger than a preset threshold value, the trained fatigue behavior recognition model is judged to pass verification.
Optionally, the method further includes, after determining whether the current action of the driver is a fatigue driving action according to the classification result and obtaining the fatigue behavior monitoring result of the driver:
when the current action of the driver is the fatigue driving action, determining the occurrence times of the fatigue driving action in a preset time period;
and reminding the driver when the occurrence times are greater than the preset times.
In addition, in order to achieve the above object, the present invention also proposes a driver fatigue behavior monitoring device including:
the acquisition module is used for monitoring the interference of the current action of the driver on the wireless signal and obtaining channel state information;
the feature extraction module is used for extracting the features of the channel state information to obtain signal features;
the dimension reduction module is used for reducing the dimension of the signal characteristics to obtain target characteristics;
the classification module is used for inputting the target characteristics into a target fatigue behavior recognition model trained in advance to obtain a classification result output by the target fatigue behavior recognition model;
and the judging module is used for judging whether the current action of the driver is the fatigue driving action according to the classification result to obtain the fatigue behavior monitoring result of the driver.
In addition, in order to achieve the above object, the present invention also proposes a driver fatigue behavior monitoring apparatus including: the system comprises a memory, a processor and a driver fatigue behavior monitoring program stored on the memory and capable of running on the processor, wherein the driver fatigue behavior monitoring program is configured to realize the driver fatigue behavior monitoring method.
In addition, in order to achieve the above object, the present invention also proposes a storage medium having stored thereon a driver fatigue behavior monitoring program which, when executed by a processor, implements the driver fatigue behavior monitoring method as described above.
The method and the device acquire channel state information by monitoring the interference of the current action of the driver on the wireless signal; extracting characteristics of the channel state information to obtain signal characteristics; performing dimension reduction on the signal characteristics to obtain target characteristics; inputting the target characteristics into a pre-trained target fatigue behavior recognition model to obtain a classification result output by the target fatigue behavior recognition model; and judging whether the current action of the driver is the fatigue driving action according to the classification result, and obtaining the fatigue behavior monitoring result of the driver. By the method, the actions of the driver are monitored by combining wireless sensing, whether the actions of the driver are fatigue driving actions or not is identified by using the trained fatigue behavior identification model, so that the fatigue driving is monitored, the occurrence of fatigue driving is reduced, the occurrence rate of road traffic safety accidents is reduced, and the fatigue driving is monitored relative to the acquired images, so that the fatigue driving monitoring method is not influenced by illumination conditions, and even if the fatigue monitoring identification rate is not reduced at night; compared with a mode of analyzing the physiological signals of the driver, the method does not need to be provided with signal acquisition equipment, and the driving experience feeling and operation of the driver are not affected.
Drawings
FIG. 1 is a schematic diagram of a driver fatigue behavior monitoring device for a hardware operating environment according to an embodiment of the present invention;
FIG. 2 is a flowchart of a first embodiment of a method for monitoring fatigue behavior of a driver according to the present invention;
FIG. 3 is a flowchart of a second embodiment of a method for monitoring fatigue behavior of a driver according to the present invention;
FIG. 4 is a flowchart of a third embodiment of a method for monitoring fatigue behavior of a driver according to the present invention;
fig. 5 is a block diagram showing the structure of a first embodiment of the driver fatigue behavior monitoring device of the present invention.
The achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
Detailed Description
It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
Referring to fig. 1, fig. 1 is a schematic structural diagram of a driver fatigue behavior monitoring device in a hardware running environment according to an embodiment of the present invention.
As shown in fig. 1, the driver fatigue behavior monitoring device may include: a processor 1001, such as a central processing unit (Central Processing Unit, CPU), a communication bus 1002, a user interface 1003, a network interface 1004, a memory 1005. Wherein the communication bus 1002 is used to enable connected communication between these components. The user interface 1003 may include a Display, an input unit such as a Keyboard (Keyboard), and the optional user interface 1003 may further include a standard wired interface, a wireless interface. The network interface 1004 may optionally include a standard wired interface, a Wireless interface (e.g., a Wireless-Fidelity (Wi-Fi) interface). The Memory 1005 may be a high-speed random access Memory (Random Access Memory, RAM) or a stable nonvolatile Memory (NVM), such as a disk Memory. The memory 1005 may also optionally be a storage device separate from the processor 1001 described above.
It will be appreciated by those skilled in the art that the structure shown in fig. 1 does not constitute a limitation of the driver fatigue behavior monitoring device, and may include more or fewer components than shown, or may combine certain components, or a different arrangement of components.
As shown in fig. 1, an operating system, a network communication module, a user interface module, and a driver fatigue behavior monitoring program may be included in the memory 1005 as one storage medium.
In the driver fatigue behavior monitoring device shown in fig. 1, the network interface 1004 is mainly used for data communication with a network server; the user interface 1003 is mainly used for data interaction with a user; the processor 1001 and the memory 1005 in the driver fatigue behavior monitoring device of the present invention may be provided in the driver fatigue behavior monitoring device, where the driver fatigue behavior monitoring device invokes the driver fatigue behavior monitoring program stored in the memory 1005 through the processor 1001, and executes the driver fatigue behavior monitoring method provided by the embodiment of the present invention.
The embodiment of the invention provides a method for monitoring fatigue behavior of a driver, and referring to fig. 2, fig. 2 is a schematic flow chart of a first embodiment of the method for monitoring fatigue behavior of a driver.
In this embodiment, the method for monitoring fatigue behavior of the driver includes the following steps:
step S10: and monitoring interference of the current action of the driver on the wireless signal to obtain channel state information.
It can be understood that the execution body of the embodiment is a driver fatigue behavior monitoring device, and the driver fatigue behavior monitoring device may be a vehicle machine or a controller disposed on a vehicle.
It should be noted that, in this embodiment, a data acquisition environment of channel state information (Channel State Information, CSI) is disposed on a vehicle machine configured with the Ubuntu system, and meanwhile, an Intel5300 network card is externally disposed as a receiving and transmitting end of a wireless signal, where a transmitting end of the wireless signal adopts 1 antenna, a receiving end of the wireless signal adopts 3 antennas, and a working frequency band selection signal is less interfered in a 2.4GHz frequency band. The Channel Frequency Response (CFR) is read as a CSI signal by a preset tool installed on the vehicle, which mainly reflects the variation of the signal amplitude and phase of the subcarrier, specifically, the preset tool may be a Linux 802.11n CSI tool, and the CFR is read as the CSI signal by a device conforming to an orthogonal frequency division multiplexing (Orthogonal Frequency Division Multiplexing, OFDM) technology in the preset tool. In a specific implementation, the channel model establishment is represented by formula (1):
wherein the method comprises the steps of
Representing the received vector>
Representing the transmit vector,/->
Representing noise, H represents frequency response.
The channel frequency response can be expressed specifically by the formula (2):
wherein N is T For the number of transmitting antennas, N R For the number of receiving antennas i e 1, N T ]，j∈[1，N R ]，k∈[1，30]，h ijk Representing CSI values for the kth subcarrier in the data stream between the ith transmit antenna and the jth receive antenna.
Wherein h is ijk In particular, it can also be expressed by the formula (3):
wherein, |h ijk I represents amplitude response, and then h ijk Representing the phase response, the amplitude response and the phase response of the signal can be resolved from the CSI information by equation (3).
It should be understood that when a wireless signal propagates in a channel as it is transmitted from a transmitting end, the channel is unstable, causing changes in diffraction, refraction, scattering, etc. during transmission, which are reflected in fluctuations in the signal at the receiving end. The changes of the actions performed by the driver on the environment are reflected in the waveforms, the channel state information is monitored through the channel model expressed by the formula (1), and the channel state information is analyzed to determine whether the driver performs fatigue actions, so that whether the driver is in a fatigue driving state is determined.
Step S20: and extracting the characteristics of the channel state information to obtain signal characteristics.
In this embodiment, the feature matrix is generated by extracting feature information from channel state information. In the specific implementation, the acquired channel state information is subjected to wavelet packet decomposition, the wavelet packet energy characteristics and the wavelet packet coefficient statistical characteristics are extracted, and the interference of noise on an original signal can be effectively avoided while the signal characteristics are extracted.
Step S30: and reducing the dimension of the signal characteristics to obtain target characteristics.
It should be understood that, in this embodiment, the signal feature may be reduced in dimension by using a linear dimension reduction method, and the signal feature may also be reduced in dimension by using a nonlinear dimension reduction method. By reducing the dimension of the signal characteristics, the complexity of the characteristics is reduced, noise mixed in the data is removed, and effective characteristic data is provided for model identification.
Specifically, the step S30 includes: carrying out standardization processing on the signal characteristics to obtain a standardized sample matrix; calculating a covariance matrix corresponding to the standardized sample matrix; determining corresponding eigenvalues and eigenvectors according to the covariance matrix; sorting the feature vectors according to the feature values to obtain a feature matrix; and extracting a plurality of feature vectors from the feature matrix according to a preset accumulated contribution rate to obtain target features.
It should be noted that, assuming that the signal features are represented as feature matrices P of i rows and j columns, the signal features P are normalized to obtain a normalized sample matrix R, specifically, the normalization is performed by the formula (4):
wherein i=1, 2, m; j=1, 2,..n.
Calculating a covariance matrix corresponding to the normalized sample matrix by the formula (5):
and determining a characteristic equation corresponding to the covariance matrix, and determining m characteristic values by the characteristic equation. The characteristic equation is expressed by the formula (6):
|C-λI m |＝0(6)
according to the calculated characteristic value lambda i (i=1, 2,., m), a unit feature vector P can be obtained 1 ，P 2 ，...，P k . The main component determines the dimension reduction degree according to the accumulated contribution rate, and extracts a plurality of feature vectors from the feature matrix to realize the dimension reduction of the signal features.
Step S40: and inputting the target characteristics into a pre-trained target fatigue behavior recognition model to obtain a classification result output by the target fatigue behavior recognition model.
It should be understood that a pre-trained target fatigue behavior recognition model is deployed on the vehicle machine in this embodiment, and classification recognition is performed based on target features corresponding to the current action according to the target fatigue behavior recognition model, so as to determine a classification result. Specifically, the target fatigue behavior recognition model is a model trained according to a large number of fatigue behavior sample features.
Step S50: and judging whether the current action of the driver is the fatigue driving action according to the classification result, and obtaining a fatigue behavior monitoring result of the driver.
The classification result is a label corresponding to the fatigue driving action or a label corresponding to other actions, and when the classification result is a label corresponding to the fatigue driving action, the current action of the driver is judged to be the fatigue driving action, and in specific implementation, the method can be set to monitor the fatigue driving action of the driver, namely, prompt the driver, and can also be set to prompt the driver when the driver is monitored to continuously appear the action reflecting the fatigue for a plurality of times within a certain time.
Further, after the step S50, the method further includes: when the current action of the driver is the fatigue driving action, determining the occurrence times of the fatigue driving action in a preset time period; and reminding the driver when the occurrence times are greater than the preset times.
It should be understood that in this embodiment, the target fatigue behavior recognition model is trained in advance, deployed on a vehicle, placed in a data acquisition environment on the vehicle, and two Intel5300 network cards are assembled in front of a driver as transmitting and receiving ends in data acquisition, so as to monitor amplitude variation of CSI signals caused when the current driver acts, perform feature extraction and dimension reduction on CSI, input the CSI signals into the target fatigue behavior recognition model for classification, and when it is determined that fatigue-reflecting actions of a preset number of times and above continuously occur in a preset time period, the control system reminds the driver to rest.
In the embodiment, the channel state information is obtained by monitoring the interference of the current action of the driver on the wireless signal; extracting characteristics of the channel state information to obtain signal characteristics; performing dimension reduction on the signal characteristics to obtain target characteristics; inputting the target characteristics into a pre-trained target fatigue behavior recognition model to obtain a classification result output by the target fatigue behavior recognition model; and judging whether the current action of the driver is the fatigue driving action according to the classification result, and obtaining the fatigue behavior monitoring result of the driver. By the method, the actions of the driver are monitored by combining wireless sensing, whether the actions of the driver are fatigue driving actions or not is identified by using the trained fatigue behavior identification model, so that the fatigue driving is monitored, the occurrence of fatigue driving is reduced, the occurrence rate of road traffic safety accidents is reduced, and the fatigue driving is monitored relative to the acquired images, and the embodiment is not influenced by illumination conditions, even if the identification rate of the fatigue monitoring at night is not reduced; compared with a mode of analyzing the physiological signals of the driver, the method does not need to be provided with signal acquisition equipment, and driving experience feeling and operation of the driver are not affected.
Referring to fig. 3, fig. 3 is a flowchart illustrating a second embodiment of the method for monitoring fatigue behavior of a driver according to the present invention.
Based on the above-described first embodiment, the step S20 of the driver fatigue behavior monitoring method of the present embodiment includes:
step S201: and carrying out wavelet packet decomposition on the channel state information, and extracting wavelet packet energy characteristics and wavelet packet coefficient statistical characteristics.
Specifically, the step S201 includes: performing multi-layer wavelet packet decomposition on the channel state information to obtain wavelet packet coefficients; forming a wavelet packet coefficient statistical feature according to the mean value, standard deviation, variance and maximum value corresponding to the wavelet packet coefficient; determining the signal energy duty ratio corresponding to each sub-band; determining wavelet packet energy vectors of each layer according to the signal energy duty ratio; and constructing wavelet packet energy characteristics according to the wavelet packet energy vectors of each layer.
It should be understood that, preferably, the present embodiment performs three-layer wavelet packet decomposition on the channel state information, adopts a wavelet basis function to perform three-layer wavelet packet decomposition on the original channel state information, obtains eight wavelet packet coefficients of different action signals, and extracts the Mean (Mean), standard deviation (Std), variance (Var) and maximum (Max) of the coefficients as the wavelet packet coefficient statistical features.
It should be noted that the signal x is assumed k，m (i) The corresponding length is N, and the signal wavelet packet energy ratio is extracted through a formula (7) to be used as wavelet packet energy characteristics:
where k is the number of decomposition layers and m is the signal subband.
Determining formula (8) according to the principle of conservation of energy:
the signal energy duty ratio of each sub-band is determined by normalization processing, expressed as formula (9):
selecting a relative wavelet packet energy vector of an n-th layer, which is expressed as a formula (10):
W n ＝(E n (0)，E n (1)，...，E n (2 n -1)) (10)
and determining three layers of wavelet packet energy vectors according to a formula (1) to form wavelet packet energy characteristics.
Step S202: and splicing the wavelet packet energy characteristics and the wavelet packet coefficient statistical characteristics to obtain signal characteristics.
In this embodiment, the channel state information caused by the current action is extracted by utilizing wavelet packet decomposition, the wavelet packet energy characteristic and the wavelet packet coefficient statistical characteristic are spliced to obtain the signal characteristic, and when the sample data is extracted by adopting the same splicing mode, the interference of noise on the original signal is effectively avoided.
In the embodiment, the channel state information is obtained by monitoring the interference of the current action of the driver on the wireless signal; carrying out wavelet packet decomposition on the channel state information, and extracting wavelet packet energy characteristics and wavelet packet coefficient statistical characteristics; splicing the wavelet packet energy characteristics and the wavelet packet coefficient statistical characteristics to obtain signal characteristics; performing dimension reduction on the signal characteristics to obtain target characteristics; inputting the target characteristics into a pre-trained target fatigue behavior recognition model to obtain a classification result output by the target fatigue behavior recognition model; and judging whether the current action of the driver is the fatigue driving action according to the classification result, and obtaining the fatigue behavior monitoring result of the driver. By the method, the acquired channel state information is subjected to wavelet packet decomposition, the wavelet packet energy characteristics and the wavelet packet coefficient statistical characteristics are extracted and spliced to obtain the signal characteristics, the interference of noise on the original signal is effectively avoided while the signal characteristics are extracted, the accuracy of model classification is improved, the monitoring of fatigue driving is realized, the occurrence of fatigue driving is reduced, and the occurrence rate of road traffic safety accidents is reduced.
Referring to fig. 4, fig. 4 is a flowchart illustrating a third embodiment of a method for monitoring fatigue behavior of a driver according to the present invention.
Based on the first embodiment, the method for monitoring fatigue behavior of a driver according to the present embodiment further includes, before the step S10:
step S101: and monitoring interference of various fatigue driving actions of a driver on the wireless signal to obtain a plurality of sample channel state information.
It can be understood that the vehicle machine of this embodiment is provided with a sample data acquisition guiding flow, and guides the driver to make various different fatigue driving actions, specifically may be actions such as yawning, frequent nodding, head shaking, bending, and the like, and when the driver makes the fatigue driving action, the wireless signal transceiver determines sample channel state information caused by the fatigue driving action, and records the sample channel state information.
Step S102: and respectively extracting the characteristics of the sample channel state information to obtain corresponding sample signal characteristics.
The method comprises the steps of carrying out wavelet packet decomposition on each sample channel state information, extracting corresponding wavelet packet energy characteristics and wavelet packet coefficient statistical characteristics, and splicing the wavelet packet energy characteristics and the wavelet packet coefficient statistical characteristics corresponding to each sample channel state information to obtain sample signal characteristics.
Step S103: and performing dimension reduction on each sample signal characteristic to obtain a corresponding target sample characteristic.
It can be appreciated that feature dimension reduction processing is performed on each sample signal feature by using Principal Component Analysis (PCA) to obtain a plurality of target sample features.
Step S104: and constructing a training set and a verification set according to the target sample characteristics and the corresponding fatigue action labels.
It should be noted that, in this embodiment, a large amount of collected sample data is divided according to a preset proportion, and a training set and a verification set are constructed, where each sample data includes a target sample feature and a corresponding fatigue action tag. The preset ratio may be 7:3.
Step S105: and training the initial fatigue behavior recognition model according to the training set to obtain a trained fatigue behavior recognition model.
It may be understood that the initial fatigue behavior recognition model may be a neural network model, preferably, the initial fatigue behavior recognition model in this embodiment is a Support Vector Machine (SVM), and the Support Vector Machine (SVM) is used to classify the target sample features of the training set, adjust the model parameters, and perform training learning to obtain the trained fatigue behavior recognition model.
Step S106: and verifying the trained fatigue behavior recognition model according to the verification set.
Specifically, the step S106 includes: identifying each target sample characteristic in the verification set according to the trained fatigue behavior identification model to obtain an identification result corresponding to each target sample characteristic in the verification set; determining corresponding recognition rates according to recognition results corresponding to the features of each target sample in the verification set; and determining whether the trained fatigue behavior recognition model passes verification according to the recognition rate, wherein when the recognition rate is larger than a preset threshold value, the trained fatigue behavior recognition model is judged to pass verification.
It should be noted that, each target sample feature in the verification set is identified through the fatigue behavior identification model after training, the identification result is compared with the fatigue action label corresponding to each target sample feature, when the identification result is matched with the fatigue action label, the identification result is recorded as identification success, when the identification result is not matched with the fatigue action label, the identification result is recorded as identification failure, the corresponding identification rate is determined according to the number of identification successes and the number of identification failures, the preset threshold value can be a fixed value set in advance, in specific implementation, the preset threshold value is set to 90%, and when the identification rate of the support vector machine is above 90%, the trained target fatigue behavior identification model is obtained.
Step S107: and when the verification is passed, obtaining a trained target fatigue behavior recognition model.
According to the embodiment, the interference of various fatigue driving actions of a driver on a wireless signal is monitored to obtain a plurality of sample channel state information; extracting the characteristics of the channel state information of each sample to obtain the corresponding sample signal characteristics; performing dimension reduction on each sample signal characteristic to obtain a corresponding target sample characteristic; constructing a training set and a verification set according to the characteristics of a plurality of target samples and the corresponding fatigue action labels; training the initial fatigue behavior recognition model according to the training set to obtain a trained fatigue behavior recognition model; verifying the trained fatigue behavior recognition model according to the verification set; and when the verification is passed, obtaining a trained target fatigue behavior recognition model. By the method, the sample channel state information caused by various fatigue driving actions of the driver is monitored, the sample channel state information is subjected to feature extraction and dimension reduction to obtain target sample features, a training set and a verification set are constructed, an initial fatigue behavior recognition model is trained, and the features corresponding to the current actions of the driver are recognized by utilizing the trained target fatigue behavior recognition model, so that whether the actions of the driver are fatigue driving actions or not is determined, the monitoring of fatigue driving is realized, the occurrence of fatigue driving is reduced, and the occurrence rate of road traffic safety accidents is reduced.
In addition, the embodiment of the invention also provides a storage medium, wherein the storage medium is stored with a driver fatigue behavior monitoring program, and the driver fatigue behavior monitoring program realizes the driver fatigue behavior monitoring method when being executed by a processor.
Because the storage medium adopts all the technical schemes of all the embodiments, the storage medium has at least all the beneficial effects brought by the technical schemes of the embodiments, and the description is omitted here.
Referring to fig. 5, fig. 5 is a block diagram showing the structure of a first embodiment of the driver fatigue behavior monitoring device of the present invention.
As shown in fig. 5, the driver fatigue behavior monitoring device according to the embodiment of the present invention includes:
the acquisition module 10 is configured to monitor interference of current actions of the driver on the wireless signal, and obtain channel state information.
And the feature extraction module 20 is configured to perform feature extraction on the channel state information to obtain a signal feature.
And the dimension reduction module 30 is used for reducing the dimension of the signal characteristic to obtain a target characteristic.
And the classification module 40 is used for inputting the target characteristics into a pre-trained target fatigue behavior recognition model to obtain a classification result output by the target fatigue behavior recognition model.
And the judging module 50 is used for judging whether the current action of the driver is the fatigue driving action according to the classification result to obtain the fatigue behavior monitoring result of the driver.
It should be understood that the foregoing is illustrative only and is not limiting, and that in specific applications, those skilled in the art may set the invention as desired, and the invention is not limited thereto.
In the embodiment, the channel state information is obtained by monitoring the interference of the current action of the driver on the wireless signal; extracting characteristics of the channel state information to obtain signal characteristics; performing dimension reduction on the signal characteristics to obtain target characteristics; inputting the target characteristics into a pre-trained target fatigue behavior recognition model to obtain a classification result output by the target fatigue behavior recognition model; and judging whether the current action of the driver is the fatigue driving action according to the classification result, and obtaining the fatigue behavior monitoring result of the driver. By the method, the actions of the driver are monitored by combining wireless sensing, whether the actions of the driver are fatigue driving actions or not is identified by using the trained fatigue behavior identification model, so that the fatigue driving is monitored, the occurrence of fatigue driving is reduced, the occurrence rate of road traffic safety accidents is reduced, and the fatigue driving is monitored relative to the acquired images, and the embodiment is not influenced by illumination conditions, even if the identification rate of the fatigue monitoring at night is not reduced; compared with a mode of analyzing the physiological signals of the driver, the method does not need to be provided with signal acquisition equipment, and driving experience feeling and operation of the driver are not affected.
It should be noted that the above-described working procedure is merely illustrative, and does not limit the scope of the present invention, and in practical application, a person skilled in the art may select part or all of them according to actual needs to achieve the purpose of the embodiment, which is not limited herein.
In addition, technical details not described in detail in the present embodiment may refer to the method for monitoring fatigue behavior of a driver provided in any embodiment of the present invention, which is not described herein.
In an embodiment, the feature extraction module 20 is further configured to perform wavelet packet decomposition on the channel state information, extract a wavelet packet energy feature and a wavelet packet coefficient statistical feature, and splice the wavelet packet energy feature and the wavelet packet coefficient statistical feature to obtain a signal feature.
In an embodiment, the feature extraction module 20 is further configured to perform multi-layer wavelet packet decomposition on the channel state information to obtain wavelet packet coefficients, form a wavelet packet coefficient statistical feature according to a mean value, a standard deviation, a variance and a maximum value corresponding to the wavelet packet coefficients, determine a signal energy ratio corresponding to each subband, determine each layer of wavelet packet energy vectors according to the signal energy ratio, and form a wavelet packet energy feature according to each layer of wavelet packet energy vectors.
In an embodiment, the dimension reduction module 30 is further configured to perform normalization processing on the signal features to obtain a normalized sample matrix, calculate a covariance matrix corresponding to the normalized sample matrix, determine a corresponding feature value and a feature vector according to the covariance matrix, sort the feature vectors according to the feature value to obtain a feature matrix, and extract a plurality of feature vectors from the feature matrix according to a preset cumulative contribution rate to obtain a target feature.
In one embodiment, the driver fatigue behavior monitoring device further comprises a training module;
the training module is used for monitoring interference of various fatigue driving actions of a driver on wireless signals to obtain a plurality of sample channel state information, respectively extracting features of the sample channel state information to obtain corresponding sample signal features, performing dimension reduction on the sample signal features to obtain corresponding target sample features, constructing a training set and a verification set according to the target sample features and corresponding fatigue action labels, training an initial fatigue behavior recognition model according to the training set to obtain a trained fatigue behavior recognition model, verifying the trained fatigue behavior recognition model according to the verification set, and obtaining a trained target fatigue behavior recognition model when verification passes.
In an embodiment, the training module is further configured to identify each target sample feature in the verification set according to the trained fatigue behavior identification model, obtain an identification result corresponding to each target sample feature in the verification set, determine a corresponding identification rate according to the identification result corresponding to each target sample feature in the verification set, and determine whether the trained fatigue behavior identification model passes verification according to the identification rate, where when the identification rate is greater than a preset threshold, it is determined that the trained fatigue behavior identification model passes verification.
In one embodiment, the driver fatigue behavior monitoring device further comprises a reminding module;
the reminding module is used for determining the occurrence times of the fatigue driving action in a preset time period when the current action of the driver is the fatigue driving action, and reminding the driver when the occurrence times are greater than the preset times.
Furthermore, it should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or system. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or system that comprises the element.
The foregoing embodiment numbers of the present invention are merely for the purpose of description, and do not represent the advantages or disadvantages of the embodiments.
From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. Read Only Memory)/RAM, magnetic disk, optical disk) and including several instructions for causing a terminal device (which may be a mobile phone, a computer, a server, or a network device, etc.) to perform the method according to the embodiments of the present invention.
The foregoing description is only of the preferred embodiments of the present invention, and is not intended to limit the scope of the invention, but rather is intended to cover any equivalents of the structures or equivalent processes disclosed herein or in the alternative, which may be employed directly or indirectly in other related arts.
Claims (8)
1. A driver fatigue behavior monitoring method, characterized in that the driver fatigue behavior monitoring method comprises:
monitoring interference of current actions of a driver on wireless signals to obtain channel state information;
extracting the characteristics of the channel state information to obtain signal characteristics;
performing dimension reduction on the signal characteristics to obtain target characteristics;
inputting the target characteristics into a pre-trained target fatigue behavior recognition model to obtain a classification result output by the target fatigue behavior recognition model;
judging whether the current action of the driver is a fatigue driving action according to the classification result, and obtaining a driver fatigue behavior monitoring result;
the extracting the characteristics of the channel state information to obtain signal characteristics includes:
carrying out wavelet packet decomposition on the channel state information, and extracting wavelet packet energy characteristics and wavelet packet coefficient statistical characteristics;
splicing the wavelet packet energy characteristics and the wavelet packet coefficient statistical characteristics to obtain signal characteristics;
the method for extracting the channel state information comprises the steps of carrying out wavelet packet decomposition on the channel state information, extracting wavelet packet energy characteristics and wavelet packet coefficient statistical characteristics, and comprises the following steps:
performing multi-layer wavelet packet decomposition on the channel state information to obtain wavelet packet coefficients;
forming a wavelet packet coefficient statistical feature according to the mean value, standard deviation, variance and maximum value corresponding to the wavelet packet coefficient;
determining the signal energy duty ratio corresponding to each sub-band;
determining wavelet packet energy vectors of each layer according to the signal energy duty ratio;
and constructing wavelet packet energy characteristics according to the wavelet packet energy vectors of each layer.
2. The method for monitoring fatigue behavior of a driver according to claim 1, wherein the step of reducing the dimension of the signal feature to obtain a target feature comprises:
carrying out standardization processing on the signal characteristics to obtain a standardized sample matrix;
calculating a covariance matrix corresponding to the standardized sample matrix;
determining corresponding eigenvalues and eigenvectors according to the covariance matrix;
sorting the feature vectors according to the feature values to obtain a feature matrix;
and extracting a plurality of feature vectors from the feature matrix according to a preset accumulated contribution rate to obtain target features.
3. The method for monitoring fatigue behavior of a driver according to claim 1, wherein the monitoring the interference of the current motion of the driver on the wireless signal, before obtaining the channel state information, further comprises:
monitoring interference of various fatigue driving actions of a driver on a wireless signal to obtain a plurality of sample channel state information;
respectively extracting the characteristics of the sample channel state information to obtain corresponding sample signal characteristics;
performing dimension reduction on each sample signal characteristic to obtain a corresponding target sample characteristic;
constructing a training set and a verification set according to a plurality of target sample characteristics and corresponding fatigue action labels;
training the initial fatigue behavior recognition model according to the training set to obtain a trained fatigue behavior recognition model;
verifying the trained fatigue behavior recognition model according to the verification set;
and when the verification is passed, obtaining a trained target fatigue behavior recognition model.
4. A driver fatigue behavior monitoring method according to claim 3, wherein the validating the trained fatigue behavior recognition model according to the validation set comprises:
identifying each target sample characteristic in the verification set according to the trained fatigue behavior identification model to obtain an identification result corresponding to each target sample characteristic in the verification set;
determining corresponding recognition rates according to recognition results corresponding to the features of each target sample in the verification set;
and determining whether the trained fatigue behavior recognition model passes verification according to the recognition rate, wherein when the recognition rate is larger than a preset threshold value, the trained fatigue behavior recognition model is judged to pass verification.
5. The method for monitoring fatigue behavior of a driver according to any one of claims 1-4, wherein the step of determining whether the current driver motion is a fatigue driving motion according to the classification result, after obtaining the driver fatigue behavior monitoring result, further comprises:
when the current action of the driver is the fatigue driving action, determining the occurrence times of the fatigue driving action in a preset time period;
and reminding the driver when the occurrence times are greater than the preset times.
6. A driver fatigue behavior monitoring device, characterized by comprising:
the acquisition module is used for monitoring the interference of the current action of the driver on the wireless signal and obtaining channel state information;
the feature extraction module is used for extracting the features of the channel state information to obtain signal features;
the dimension reduction module is used for reducing the dimension of the signal characteristics to obtain target characteristics;
the classification module is used for inputting the target characteristics into a target fatigue behavior recognition model trained in advance to obtain a classification result output by the target fatigue behavior recognition model;
the judging module is used for judging whether the current action of the driver is the fatigue driving action according to the classification result to obtain a fatigue behavior monitoring result of the driver;
the characteristic extraction module is also used for carrying out wavelet packet decomposition on the channel state information and extracting wavelet packet energy characteristics and wavelet packet coefficient statistical characteristics;
splicing the wavelet packet energy characteristics and the wavelet packet coefficient statistical characteristics to obtain signal characteristics;
the characteristic extraction module is further used for carrying out multi-layer wavelet packet decomposition on the channel state information to obtain wavelet packet coefficients;
forming a wavelet packet coefficient statistical feature according to the mean value, standard deviation, variance and maximum value corresponding to the wavelet packet coefficient;
determining the signal energy duty ratio corresponding to each sub-band;
determining wavelet packet energy vectors of each layer according to the signal energy duty ratio;
and constructing wavelet packet energy characteristics according to the wavelet packet energy vectors of each layer.
7. A driver fatigue behavior monitoring device, characterized in that the device comprises: a memory, a processor, and a driver fatigue behavior monitoring program stored on the memory and executable on the processor, the driver fatigue behavior monitoring program configured to implement the driver fatigue behavior monitoring method of any one of claims 1 to 5.
8. A storage medium having stored thereon a driver fatigue behavior monitoring program which when executed by a processor implements the driver fatigue behavior monitoring method according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111607058.9A CN114209325B (en) | 2021-12-23 | 2021-12-23 | Driver fatigue behavior monitoring method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111607058.9A CN114209325B (en) | 2021-12-23 | 2021-12-23 | Driver fatigue behavior monitoring method, device, equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114209325A CN114209325A (en) | 2022-03-22 |
| CN114209325B true CN114209325B (en) | 2023-06-23 |
Family
ID=80705818
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111607058.9A Active CN114209325B (en) | 2021-12-23 | 2021-12-23 | Driver fatigue behavior monitoring method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114209325B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101256768A (en) * | 2008-04-03 | 2008-09-03 | 清华大学 | Time frequency two-dimension converse spectrum characteristic extracting method for recognizing language species |
| JP2011125040A (en) * | 2002-01-23 | 2011-06-23 | Qualcomm Inc | Reallocation of excess power for full channel-state information (csi) multiple-input, multiple-output (mimo) systems |
| MX2014006442A (en) * | 2014-05-29 | 2015-11-30 | Víctor Hugo Ortiz Flores | Electronic system for keeping a person awake. |
| CN106384129A (en) * | 2016-09-13 | 2017-02-08 | 西安科技大学 | Driver driving state recognition method based on closed loop driving model |
| CN107280663A (en) * | 2017-07-07 | 2017-10-24 | 南京邮电大学 | A kind of method of the tired brain electrical feature research based on different experiments difficulty |
| CN107944418A (en) * | 2017-12-07 | 2018-04-20 | 上海交通大学 | A kind of method using Wi FiCSI infomation detection fatigue drivings |
| CN109697831A (en) * | 2019-02-25 | 2019-04-30 | 湖北亿咖通科技有限公司 | Fatigue driving monitoring method, device and computer readable storage medium |
| CN111832432A (en) * | 2020-06-23 | 2020-10-27 | 华中科技大学 | Cutter wear real-time prediction method based on wavelet packet decomposition and deep learning |
| CN113071512A (en) * | 2021-04-25 | 2021-07-06 | 东风柳州汽车有限公司 | Safe driving reminding method, device, equipment and storage medium |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10735298B2 (en) * | 2012-12-05 | 2020-08-04 | Origin Wireless, Inc. | Method, apparatus, server and system for vital sign detection and monitoring |
| US20200064444A1 (en) * | 2015-07-17 | 2020-02-27 | Origin Wireless, Inc. | Method, apparatus, and system for human identification based on human radio biometric information |
| WO2017200522A1 (en) * | 2016-05-16 | 2017-11-23 | Ford Global Technologies, Llc | Methods and apparatus for on-demand fuel delivery |
| US10623640B2 (en) * | 2017-03-17 | 2020-04-14 | Uurmi Systems Pvt Ltd | Method and system for detection of contaminants present on a lens of an imaging device |
| US20190077409A1 (en) * | 2017-07-31 | 2019-03-14 | Alcohol Countermeasure Systems (International) Inc. | Non-intrusive assessment of fatigue in drivers using eye tracking |
-
2021
- 2021-12-23 CN CN202111607058.9A patent/CN114209325B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2011125040A (en) * | 2002-01-23 | 2011-06-23 | Qualcomm Inc | Reallocation of excess power for full channel-state information (csi) multiple-input, multiple-output (mimo) systems |
| CN101256768A (en) * | 2008-04-03 | 2008-09-03 | 清华大学 | Time frequency two-dimension converse spectrum characteristic extracting method for recognizing language species |
| MX2014006442A (en) * | 2014-05-29 | 2015-11-30 | Víctor Hugo Ortiz Flores | Electronic system for keeping a person awake. |
| CN106384129A (en) * | 2016-09-13 | 2017-02-08 | 西安科技大学 | Driver driving state recognition method based on closed loop driving model |
| CN107280663A (en) * | 2017-07-07 | 2017-10-24 | 南京邮电大学 | A kind of method of the tired brain electrical feature research based on different experiments difficulty |
| CN107944418A (en) * | 2017-12-07 | 2018-04-20 | 上海交通大学 | A kind of method using Wi FiCSI infomation detection fatigue drivings |
| CN109697831A (en) * | 2019-02-25 | 2019-04-30 | 湖北亿咖通科技有限公司 | Fatigue driving monitoring method, device and computer readable storage medium |
| CN111832432A (en) * | 2020-06-23 | 2020-10-27 | 华中科技大学 | Cutter wear real-time prediction method based on wavelet packet decomposition and deep learning |
| CN113071512A (en) * | 2021-04-25 | 2021-07-06 | 东风柳州汽车有限公司 | Safe driving reminding method, device, equipment and storage medium |
Non-Patent Citations (3)
| Title |
|---|
| Fuzzy wavelet packet based feature extraction method and its application to biomedical signal classification;Li DQ,等;IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING;第1132-1139页 * |
| 基于EEG的驾驶疲劳水平分级方法研究;李佳栗;《中国优秀硕士学位论文全文数据库工程科技Ⅱ辑》;全文 * |
| 基于信道状态信息的驾驶员行为研究;潘兴达;《中国优秀硕士学位论文全文数据库工程科技II辑》;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114209325A (en) | 2022-03-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111126258B (en) | Image recognition method and related device | |
| Zhang et al. | Robust sound event recognition using convolutional neural networks | |
| Malta et al. | A study of driver behavior under potential threats in vehicle traffic | |
| CN107463874A (en) | The intelligent safeguard system of Emotion identification method and system and application this method | |
| DE102015111969A1 (en) | Identification of the driver of a vehicle | |
| CN103413050A (en) | Motor imagery electroencephalogram voting strategy sorting method based on extreme learning machines | |
| CN109658272A (en) | Driving behavior evaluation system and Insurance Pricing system based on driving behavior | |
| US11529960B2 (en) | Method and apparatus for detecting driver's abnormalities based on machine learning using vehicle can bus signal | |
| CN102880875A (en) | Semi-supervised learning face recognition method based on low-rank representation (LRR) graph | |
| CN112699793A (en) | Fatigue driving detection optimization identification method based on random forest | |
| US8106756B2 (en) | Adaptive interface providing apparatus and method | |
| Öztürk et al. | Driver status identification from driving behavior signals | |
| CN102592593A (en) | Emotional-characteristic extraction method implemented through considering sparsity of multilinear group in speech | |
| Suresh et al. | Driver drowsiness detection using deep learning | |
| DE102019105651A1 (en) | Classification of signal objects within ultrasonic reception signals and compressed transmission of symbols as representatives of these signal objects to a computer unit for object recognition | |
| CN114209325B (en) | Driver fatigue behavior monitoring method, device, equipment and storage medium | |
| CN110097120B (en) | Network flow data classification method, equipment and computer storage medium | |
| Ramakrishnan et al. | Epileptic eeg signal classification using multi-class convolutional neural network | |
| CN114782933A (en) | Driver fatigue detection system based on multi-mode Transformer network | |
| Kumbhar et al. | Anytime prediction as a model of human reaction time | |
| CN114860922B (en) | Method for obtaining classification model of psychological assessment scale, screening method and system | |
| CN115844402A (en) | Attention prediction method, attention prediction device, electronic equipment and computer readable storage medium | |
| CN111611848B (en) | Cadaver iris recognition method and device | |
| CN114358065A (en) | Method, device and equipment for monitoring smoking behavior of driver and storage medium | |
| KR20220005413A (en) | Earthquake Event Classification Method and Apparatus Using Multi-task Deep Learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |