CN109394248B - Driving fatigue detection method and system - Google Patents
Driving fatigue detection method and system Download PDFInfo
- Publication number
- CN109394248B CN109394248B CN201811576142.7A CN201811576142A CN109394248B CN 109394248 B CN109394248 B CN 109394248B CN 201811576142 A CN201811576142 A CN 201811576142A CN 109394248 B CN109394248 B CN 109394248B
- Authority
- CN
- China
- Prior art keywords
- fatigue
- sample
- calculating
- driver
- samples
- 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
- A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B5/316—Modalities, i.e. specific diagnostic methods
- A61B5/369—Electroencephalography [EEG]
-
- 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
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Pathology (AREA)
- Heart & Thoracic Surgery (AREA)
- Veterinary Medicine (AREA)
- Psychiatry (AREA)
- Psychology (AREA)
- Public Health (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Animal Behavior & Ethology (AREA)
- Biomedical Technology (AREA)
- Surgery (AREA)
- Medical Informatics (AREA)
- Molecular Biology (AREA)
- Educational Technology (AREA)
- Child & Adolescent Psychology (AREA)
- Developmental Disabilities (AREA)
- Social Psychology (AREA)
- Hospice & Palliative Care (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
本发明公开了一种驾驶疲劳检测方法及系统，该方法包括：按照预设的采样频率多次连续采集驾驶员前额两导联的脑电信号，以获得脑电信号样本；对脑电信号样本按照采样周期进行划分，组成以采样周期为长度的样本向量集合；计算每个样本向量集合的两个导联的同步性；对计算得到的同步性结果进行时间段离散化，以筛除奇异样本，获得目标样本；根据目标样本以及驾驶员的驾驶疲劳状态变化特征，确定疲劳阈值；在线采集驾驶员的脑电信号，实时计算同步性结果，将实时计算的同步性结果与疲劳阈值进行对比分析，以确定驾驶员当前是否为疲劳状态。本发明能够解决全脑区导联在实际应用中困难、无法连续性的检测驾驶员的疲劳状态以及计算方法过于复杂的问题。
The invention discloses a driving fatigue detection method and system. The method comprises the following steps: continuously collecting the EEG signals of two leads of the driver's forehead for several times according to a preset sampling frequency to obtain EEG signal samples; Divide according to the sampling period to form a sample vector set whose length is the sampling period; calculate the synchronicity of the two leads of each sample vector set; perform time segment discretization on the calculated synchronicity results to filter out singular samples , obtain the target sample; determine the fatigue threshold according to the target sample and the change characteristics of the driver's driving fatigue state; collect the driver's EEG signal online, calculate the synchronization result in real time, and compare and analyze the synchronization result calculated in real time with the fatigue threshold. , to determine whether the driver is currently fatigued. The invention can solve the problems that the whole brain area lead is difficult in practical application, cannot continuously detect the fatigue state of the driver, and the calculation method is too complicated.
Description
Technical Field
The invention relates to the technical field of driving safety, in particular to a driving fatigue detection method and system.
Background
Driving fatigue refers to a phenomenon in which the driver information acquisition, information processing, and handling ability are reduced due to a long-time continuous driving. In the long-time driving process, a driver is easy to cause driving fatigue, so that the response time of the driver to an emergency is prolonged, and the driving safety is influenced. In order to avoid driving fatigue, the driver is reminded when the driver is in a fatigue state.
The driving fatigue detection method based on the electroencephalogram signals is a main solution for the current driving fatigue detection, but the current driving fatigue detection method based on the electroencephalogram signals also has the following problems:
firstly, the signal acquisition adopts a whole brain area, although the accuracy of fatigue detection can be improved by using the whole brain area in a laboratory, except that two leads in the forehead area are not covered by hair, other leads are difficult to realize in practical application;
secondly, most of the fatigue detection at present is to divide the fatigue state into an extreme fatigue state and a non-fatigue state to realize binarization detection, but the fatigue is a continuous and progressive physiological change, and the binarization fatigue detection is likely to cause that a driver is extremely fatigued to be detected, and at the moment, traffic accidents are likely to happen;
and thirdly, the fatigue detection calculation method is too complex, so that the fatigue detection result is delayed.
Disclosure of Invention
Therefore, an object of the present invention is to provide a driving fatigue detection method, so as to solve the problems that the brain leads are difficult to detect the fatigue state of the driver continuously in practical application, and the calculation method is too complex.
A driving fatigue detection method, comprising:
continuously acquiring two-lead electroencephalogram signals of a forehead of a driver for multiple times according to a preset sampling frequency to obtain an electroencephalogram signal sample in a fatigue state;
dividing the electroencephalogram signal samples according to a sampling period to form a sample vector set with the sampling period as the length;
calculating synchronicity of the two leads of each of the sample vector sets;
carrying out time period discretization on the calculated synchronism result to screen out singular samples and obtain target samples;
determining a fatigue threshold according to the target sample and the driving fatigue state change characteristics of the driver;
and acquiring electroencephalogram signals of the driver on line, calculating a synchronism result in real time, and comparing and analyzing the synchronism result calculated in real time with the fatigue threshold value to determine whether the driver is in a fatigue state at present.
The driving fatigue detection method provided by the invention at least has the following beneficial effects:
1) the invention adopts two forehead leads as a signal source, can avoid the problem of difficult acquisition caused by covering of an electroencephalogram signal during acquisition, and can realize the detection effect with high accuracy by a two-lead synchronism calculation method;
2) the invention uses the two-lead synchronism value as the output result, and outputs the continuous value as the detection basis through each sampling period, thereby realizing the real-time data output and continuously detecting the fatigue state of the driver in real time;
3) in the prior art, the real value output by using the forehead region leads is mainly output by using an entropy method, but the entropy output method needs to reconstruct a time domain periodic signal and has high time complexity.
In addition, the driving fatigue detection method according to the present invention may further include the following additional features:
further, the step of calculating the synchronicity of the two leads of each sample vector set adopts a phase synchronization calculation method, specifically:
the calculation method of the phase-locked value PLV between the two leads is as follows:
therein
Is a time series xi(t) and xj(t) calculating the phase change of the time domain signal using a hilbert transform, the hilbert transform being obtained for a continuous time sequence x (t) by the following equation:
where PV represents the cauchy principal value and the phase change is calculated by the following equation:
further, in the step of calculating the synchronicity of the two leads of each sample vector set, a time domain vector synchronization calculation method is adopted, specifically:
where t is the time component in the brain electrical signal sample, N is the length of one sampling period, xiAnd xjRespectively representing a sample time series of two electrodes.
Further, the step of dividing the electroencephalogram signal samples according to the sampling period to form a sample vector set with the sampling period as the length includes:
the data was truncated according to the sampling frequency, and for a signal of M minutes for one experiment, the samples were organized into a matrix of 2 x 60M x H, where 60M is the number of seconds sampled, H is the number of H data points collected in one second, and 2 is two leads.
Further, the time-segment discretization of the calculated synchronicity result is performed to screen out singular samples, and the step of obtaining the target sample includes:
selecting each row in the calculated matrix of the synchronism result, dividing the vectors in the matrix into M vectors according to the step length 60, and then reserving two bits behind a decimal point to obtain an L-60 matrix, wherein L is the number of samples;
calculating the distance between each row of the L-60 matrix to eliminate singular samples, wherein the calculation method comprises the following steps of firstly calculating the Fisher distance between row vectors, and the calculation formula is as follows:
where μ represents the mean of the vector, σ represents the standard deviation of the vector, Fi,jComputingThe result is a matrix of L and defines Fi,jA value of (d);
statistical samples, if there is a vector m, where FimOr FjmExceeding a defined Fi,jAnd if so, determining m as a singular sample, and removing the singular sample m.
Further, the step of determining a fatigue threshold according to the target sample and the driving fatigue state change characteristics of the driver comprises:
calculating a matrix of the target samples;
taking M/2 as a step length, averaging the matrix of the target sample according to rows, then averaging according to columns, and finally obtaining two numbers, namely a first half-segment numerical value and a second half-segment numerical value;
and determining the second half segment value as a fatigue threshold value.
Further, the step of collecting electroencephalogram signals of the driver on line, calculating a synchronicity result in real time, and comparing and analyzing the synchronicity result calculated in real time and the fatigue threshold value to determine whether the driver is in a fatigue state at present comprises:
acquiring electroencephalogram signals of a driver on line, calculating synchronism once every 1 second to obtain a numerical value, marking the fatigue state of the second as 1 if the numerical value exceeds the fatigue threshold value, otherwise marking the fatigue state of the second as 0, continuously calculating for 1 minute to obtain a marking vector, summing the marking vector, judging that the driver is in the fatigue state at present if the summation result exceeds a preset value, and outputting the fatigue value of the last second.
Another objective of the present invention is to provide a driving fatigue detection system to solve the problems that the whole brain region leads are difficult to detect the fatigue state of the driver continuously in practical application and the calculation method is too complex.
A driving fatigue detection system, the system comprising:
the acquisition module is used for continuously acquiring the two-lead electroencephalogram signals of the forehead of the driver for multiple times according to a preset sampling frequency so as to obtain an electroencephalogram signal sample in a fatigue state;
the dividing module is used for dividing the electroencephalogram signal samples according to the sampling period to form a sample vector set taking the sampling period as the length;
a calculation module for calculating the synchronicity of the two leads of each sample vector set;
the dispersion module is used for carrying out time period discretization on the calculated synchronism result so as to screen out singular samples and obtain target samples;
the threshold value determining module is used for determining a fatigue threshold value according to the target sample and the driving fatigue state change characteristics of the driver;
and the fatigue determining module is used for collecting electroencephalogram signals of the driver on line, calculating a synchronism result in real time, and comparing and analyzing the synchronism result calculated in real time and the fatigue threshold value to determine whether the driver is in a fatigue state at present.
The driving fatigue detection system provided by the invention at least has the following beneficial effects:
1) the invention adopts two forehead leads as a signal source, can avoid the problem of difficult acquisition caused by covering of an electroencephalogram signal during acquisition, and can realize the detection effect with high accuracy by a two-lead synchronism calculation method;
2) the invention uses the two-lead synchronism value as the output result, and outputs the continuous value as the detection basis through each sampling period, thereby realizing the real-time data output and continuously detecting the fatigue state of the driver in real time;
3) in the prior art, the real value output by using the forehead region leads is mainly output by using an entropy method, but the entropy output method needs to reconstruct a time domain periodic signal and has high time complexity.
In addition, the driving fatigue detection system according to the present invention may further have the following additional features:
further, the calculation module may calculate the synchronicity of the two leads of each sample vector set by using a phase synchronization calculation method, specifically:
the calculation method of the phase-locked value PLV between the two leads is as follows:
therein
Is a time series xi(t) and xj(t) calculating the phase change of the time domain signal using a hilbert transform, the hilbert transform being obtained for a continuous time sequence x (t) by the following equation:
where PV represents the cauchy principal value and the phase change is calculated by the following equation:
further, the calculating module may calculate the synchronicity of the two leads of each sample vector set by using a time domain vector synchronization calculating method, specifically:
where t is the time component in the brain electrical signal sample, N is the length of one sampling period, xiAnd xjRespectively representing a sample time series of two electrodes.
Further, the dividing module is specifically configured to:
the data was truncated according to the sampling frequency, and for a signal of M minutes for one experiment, the samples were organized into a matrix of 2 x 60M x H, where 60M is the number of seconds sampled, H is the number of H data points collected in one second, and 2 is two leads.
Further, the discrete module is specifically configured to:
selecting each row in the calculated matrix of the synchronism result, dividing the vectors in the matrix into M vectors according to the step length 60, and then reserving two bits behind a decimal point to obtain an L-60 matrix, wherein L is the number of samples;
calculating the distance between each row of the L-60 matrix to eliminate singular samples, wherein the calculation method comprises the following steps of firstly calculating the Fisher distance between row vectors, and the calculation formula is as follows:
where μ represents the mean of the vector, σ represents the standard deviation of the vector, Fi,jCalculating the result as a matrix of L x L and defining Fi,jA value of (d);
statistical samples, if there is a vector m, where FimOr FjmExceeding a defined Fi,jAnd if so, determining m as a singular sample, and removing the singular sample m.
Further, the threshold determination module is specifically configured to:
calculating a matrix of the target samples;
taking M/2 as a step length, averaging the matrix of the target sample according to rows, then averaging according to columns, and finally obtaining two numbers, namely a first half-segment numerical value and a second half-segment numerical value;
and determining the second half segment value as a fatigue threshold value.
Further, the fatigue determination module is specifically configured to:
acquiring electroencephalogram signals of a driver on line, calculating synchronism once every 1 second to obtain a numerical value, marking the fatigue state of the second as 1 if the numerical value exceeds the fatigue threshold value, otherwise marking the fatigue state of the second as 0, continuously calculating for 1 minute to obtain a marking vector, summing the marking vector, judging that the driver is in the fatigue state at present if the summation result exceeds a preset value, and outputting the fatigue value of the last second.
Drawings
The above and/or additional aspects and advantages of embodiments of the present invention will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
fig. 1 is a flowchart of a driving fatigue detection method according to a first embodiment of the present invention;
fig. 2 is a schematic configuration diagram of a driving fatigue detection system according to a second embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, a driving fatigue detection method according to a first embodiment of the present invention includes steps S101 to S103:
s101, continuously collecting two-lead electroencephalogram signals of a forehead of a driver for multiple times according to a preset sampling frequency to obtain electroencephalogram signal samples in a fatigue state;
the method includes the steps that the existing portable electroencephalogram signal acquisition equipment (for example, neurosky forehead two-lead electroencephalogram signal acquisition equipment) can be used for acquiring forehead two-lead electroencephalogram signals of a driver, multiple continuous acquisition is conducted on the driver, the sampling frequency can be set to be 256Hz or 512Hz or 1000Hz or 1024Hz, and the like, the adopted frequency can be selected according to actual conditions, and the method is not limited.
In this embodiment, the sampling frequency is described by taking 1000Hz as an example, driver electroencephalograms are continuously collected for 40 minutes, each subject (i.e., driver) is collected for multiple times, after each collection is completed, a questionnaire is given to the subject to determine whether the subject is tired, when the questionnaire result of the subject is tired, the sample is stored, otherwise, the collection is performed, in this embodiment, the number of samples in a sample set of the subject is not less than 30, and then the collection experiment can be stopped.
S102, dividing the electroencephalogram signal samples according to a sampling period to form a sample vector set with the sampling period as the length;
in one embodiment, the data may be truncated according to the sampling frequency, and for a signal with M minutes in one experiment, the samples are grouped into a matrix of 2 × 60M × H, where 60M is the number of seconds of sampling, H is the number of data points acquired in one second, and 2 is two leads.
In this embodiment, the data is truncated according to the sampling frequency, and for a 40-minute signal from one experiment, the samples form a matrix of 2 x 2400 x 1000, where 2400 is the sampling time (seconds) of 40 minutes, 1000 is the data point collected in one second (corresponding to the sampling frequency of 1000Hz), and 2 represents two leads.
S103, calculating the synchronism of the two leads of each sample vector set;
the method for calculating the synchronicity of the two leads of each sample vector set can adopt various synchronous calculation methods such as phase synchronization or time domain vector synchronization.
The specific process of calculating the synchronism of the two leads by adopting a phase synchronization calculation method comprises the following steps:
the calculation method of the phase-locked value PLV between the two leads is as follows:
therein
Is a time series xi(t) and xj(t) calculating the phase change of the time domain signal using a hilbert transform, the hilbert transform being obtained for a continuous time sequence x (t) by the following equation:
where PV represents the cauchy principal value and the phase change is calculated by the following equation:
the following formula is specifically adopted for calculating the synchronism of the two leads by adopting a time domain vector synchronous calculation method:
where t is the time component in the brain electrical signal sample, N is the length of one sampling period, xiAnd xjRespectively representing a sample time series of two electrodes.
Specifically, in this embodiment, a time domain vector synchronization calculation method is adopted to calculate the synchronicity of the two leads. Where N is 1000, one sample period length, xiAnd xjThe time series of samples representing the two electrodes respectively, a 2400 vector can be calculated for each sample, each value of the vector represents the synchronicity value of the two-lead electroencephalogram signal at the second moment, and 30 samples can be constructed into a 30 × 2400 matrix, wherein 30 is the number of samples.
S104, performing time-period discretization on the calculated synchronism result to screen out singular samples and obtain target samples;
the method specifically comprises the following steps of screening out singular samples:
selecting each row in the calculated matrix of the synchronism result, dividing the vectors in the matrix into M vectors according to the step length 60, and then reserving two bits behind a decimal point to obtain an L-60 matrix, wherein L is the number of samples;
calculating the distance between each row of the L-60 matrix to eliminate singular samples, wherein the calculation method comprises the following steps of firstly calculating the Fisher distance between row vectors, and the calculation formula is as follows:
where μ represents the mean of the vector, σ represents the standard deviation of the vector, Fi,jCalculating the result as a matrix of L x L and defining Fi,jA value of (d);
statistical samples, if there is a vector m, where FimOr FjmExceeding a defined Fi,jAnd if so, determining m as a singular sample, and removing the singular sample m.
Specifically, in this embodiment, each row is selected, the vector of 2400 is divided into 40 vectors according to the step size 60 (i.e., 2400 seconds is divided into 40 minutes), and then two digits after the decimal point are reserved, so that a matrix of 30 × 60 is obtained, where L is 30;
calculating the distance between each row of the matrix of 30-60 to eliminate singular samples, wherein the calculation method comprises the following steps of firstly, calculating the Fisher distance between row vectors, and the calculation formula is as follows:
where μ represents the mean of the vector, σ represents the standard deviation of the vector, and F results in a 30 x 30 matrix. Then, the value of F is defined, in this embodiment, F is defined as 1.8, when F isi,jIf the distance between the i line and the j line is not more than 1.8, the tentative i line and the tentative j line are not singular vectors, and if the distance between the i line and the j line is not more than 1.8, the other vector m exists, wherein F isimOr FjmBeyond 1.8, then m can be determined to be a singular sample. In this example, 25 samples were finally screened out of 30 samples.
S105, determining a fatigue threshold according to the target sample and the driving fatigue state change characteristics of the driver;
since the occurrence of fatigue is a gradual process, in order to simplify the threshold determination process, the fatigue threshold may be determined by the following method:
calculating a matrix of the target samples, in this embodiment, (1) calculating a matrix of 25 × 40 obtained in step S104 (25 is the number of samples after deletion, and 40 is the vector calculated in step S104);
taking M/2 as a step length, averaging the matrix of the target sample according to rows, then averaging according to columns, and finally obtaining two numbers, namely a first half value and a second half value, in the embodiment, taking 20 as a step length, averaging the matrix of the target sample according to rows, and finally obtaining a 25 x 2 matrix, then averaging according to columns, and finally obtaining two numbers, in the embodiment, finally obtaining two numbers, namely a first half value and a second half value, specifically (0.64, 0.35);
since the occurrence of fatigue is a gradual process and usually from a non-fatigue state to a fatigue state, the second half value is determined as a fatigue threshold, i.e., the fatigue threshold is 0.35.
In addition, as a specific example, since each person may know whether or not they are in a fatigue state differently, when calculating the fatigue threshold, the driver may adjust the fatigue threshold within a certain range based on the calculated fatigue threshold, for example, the fatigue threshold may be adjusted to 0.4 when the driver has a strong fatigue resistance, whereas the fatigue threshold may be adjusted to 0.3 when the driver has a weak fatigue resistance, so as to more conform to the fatigue state of the person, and the adjusted fatigue threshold is the criterion for subsequent determination.
S106, collecting electroencephalogram signals of the driver on line, calculating a synchronism result in real time, and comparing and analyzing the synchronism result calculated in real time and the fatigue threshold value to determine whether the driver is in a fatigue state at present.
The method comprises the steps of collecting electroencephalogram signals of a driver on line, calculating synchronism every 1 second to obtain a numerical value, marking the fatigue state of the second as 1 if the numerical value exceeds a fatigue threshold value, otherwise marking the fatigue state of the second as 0, continuously calculating for 1 minute (namely 60 seconds) to obtain a marking vector, summing the marking vector, and judging that the driver is in the fatigue state currently (in the minute) if the summation result exceeds a preset value (the preset value can be adjusted, for example, 40 is adopted, and if the summation result exceeds 40 is adopted. In addition, in general, when it is determined that the vehicle is in a fatigue state, the fatigue value of the last second is usually the largest within 1 minute, and therefore the fatigue value of the last second can be output as the final detection result.
According to the driving fatigue detection method provided by the embodiment, at least the following beneficial effects are achieved:
1) the invention adopts two forehead leads as a signal source, can avoid the problem of difficult acquisition caused by covering of an electroencephalogram signal during acquisition, and can realize the detection effect with high accuracy by a two-lead synchronism calculation method;
2) the invention uses the two-lead synchronism value as the output result, and outputs the continuous value as the detection basis through each sampling period, thereby realizing the real-time data output and continuously detecting the fatigue state of the driver in real time;
3) in the prior art, the real value output by using the forehead region leads is mainly output by using an entropy method, but the entropy output method needs to reconstruct a time domain periodic signal and has high time complexity.
Referring to fig. 2, a driving fatigue detecting system according to a second embodiment of the present invention based on the same inventive concept includes:
the acquisition module 10 is used for continuously acquiring the two-lead electroencephalogram signals of the forehead of the driver for multiple times according to a preset sampling frequency so as to obtain an electroencephalogram signal sample in a fatigue state;
the dividing module 20 is configured to divide the electroencephalogram signal samples according to a sampling period to form a sample vector set with the sampling period as a length;
a calculation module 30 for calculating the synchronicity of the two leads of each of the sample vector sets;
the discretization module 40 is used for performing time-segment discretization on the calculated synchronicity result to screen out singular samples and obtain target samples;
a threshold determination module 50, configured to determine a fatigue threshold according to the target sample and the driving fatigue state change characteristic of the driver;
and the fatigue determining module 60 is configured to acquire electroencephalogram signals of the driver on line, calculate a synchronicity result in real time, and compare and analyze the synchronicity result calculated in real time with the fatigue threshold value to determine whether the driver is in a fatigue state at present.
The calculating module 30 may calculate the synchronicity of the two leads of each sample vector set by using a phase synchronization calculating method, specifically:
the calculation method of the phase-locked value PLV between the two leads is as follows:
therein
Is a time series xi(t) and xj(t) calculating the phase change of the time domain signal using a hilbert transform, the hilbert transform being obtained for a continuous time sequence x (t) by the following equation:
where PV represents the cauchy principal value and the phase change is calculated by the following equation:
the calculating module 30 may calculate the synchronicity of the two leads of each sample vector set by using a time domain vector synchronization calculating method, specifically:
where t is the time component in the brain electrical signal sample, N is the length of one sampling period, xiAnd xjRespectively representing a sample time series of two electrodes.
Wherein, the dividing module 20 is specifically configured to:
the data was truncated according to the sampling frequency, and for a signal of M minutes for one experiment, the samples were organized into a matrix of 2 x 60M x H, where 60M is the number of seconds sampled, H is the number of H data points collected in one second, and 2 is two leads.
Wherein the discrete module 40 is specifically configured to:
selecting each row in the calculated matrix of the synchronism result, dividing the vectors in the matrix into M vectors according to the step length 60, and then reserving two bits behind a decimal point to obtain an L-60 matrix, wherein L is the number of samples;
calculating the distance between each row of the L-60 matrix to eliminate singular samples, wherein the calculation method comprises the following steps of firstly calculating the Fisher distance between row vectors, and the calculation formula is as follows:
where μ represents the mean of the vector, σ represents the standard deviation of the vector, Fi,jCalculating the result as a matrix of L x L and defining Fi,jA value of (d);
statistical samples, if there is a vector m, where FimOr FjmExceeding a defined Fi,jAnd if so, determining m as a singular sample, and removing the singular sample m.
Wherein the threshold determination module 50 is specifically configured to:
calculating a matrix of the target samples;
taking M/2 as a step length, averaging the matrix of the target sample according to rows, then averaging according to columns, and finally obtaining two numbers, namely a first half-segment numerical value and a second half-segment numerical value;
and determining the second half segment value as a fatigue threshold value.
Wherein the fatigue determination module 60 is specifically configured to:
acquiring electroencephalogram signals of a driver on line, calculating synchronism once every 1 second to obtain a numerical value, marking the fatigue state of the second as 1 if the numerical value exceeds the fatigue threshold value, otherwise marking the fatigue state of the second as 0, continuously calculating for 1 minute to obtain a marking vector, summing the marking vector, judging that the driver is in the fatigue state at present if the summation result exceeds a preset value, and outputting the fatigue value of the last second.
The driving fatigue detection system provided by the embodiment has at least the following beneficial effects:
1) the invention adopts two forehead leads as a signal source, can avoid the problem of difficult acquisition caused by covering of an electroencephalogram signal during acquisition, and can realize the detection effect with high accuracy by a two-lead synchronism calculation method;
2) the invention uses the two-lead synchronism value as the output result, and outputs the continuous value as the detection basis through each sampling period, thereby realizing the real-time data output and continuously detecting the fatigue state of the driver in real time;
3) in the prior art, the real value output by using the forehead region leads is mainly output by using an entropy method, but the entropy output method needs to reconstruct a time domain periodic signal and has high time complexity.
The logic and/or steps represented in the flowcharts or otherwise described herein, e.g., an ordered listing of executable instructions that can be considered to implement logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device.
More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium could even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via for instance optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner if necessary, and then stored in a computer memory.
It should be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit of a logic gate circuit specifically used for realizing a logic function for a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
While embodiments of the invention have been shown and described, it will be understood by those of ordinary skill in the art that: various changes, modifications, substitutions and alterations can be made to the embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.
Claims (7)
1. A driving fatigue detection method, characterized in that the method comprises:
continuously acquiring two-lead electroencephalogram signals of a forehead of a driver for multiple times according to a preset sampling frequency to obtain an electroencephalogram signal sample in a fatigue state;
dividing the electroencephalogram signal samples according to a sampling period to form a sample vector set with the sampling period as the length;
calculating the synchronism of the two leads of each sample vector set by adopting a phase synchronization calculation method or a time domain vector synchronization calculation method;
carrying out time period discretization on the calculated synchronism result to screen out singular samples and obtain target samples;
determining a fatigue threshold according to the target sample and the driving fatigue state change characteristics of the driver;
acquiring electroencephalogram signals of a driver on line, calculating a synchronism result in real time, and comparing and analyzing the synchronism result calculated in real time with the fatigue threshold value to determine whether the driver is in a fatigue state at present;
the step of dividing the electroencephalogram signal samples according to the sampling period to form a sample vector set with the sampling period as the length comprises the following steps:
truncating the data according to the sampling frequency, forming the samples into a matrix of 2 x 60M x H for a signal of M minutes in one experiment, wherein 60M is the number of seconds of sampling, H is the number of data points acquired in one second, and 2 is two leads;
the time period discretization is carried out on the calculated synchronism result to screen out singular samples, and the step of obtaining the target sample comprises the following steps:
selecting each row in the calculated matrix of the synchronism result, dividing the vectors in the matrix into M vectors according to the step length 60, and then reserving two bits behind a decimal point to obtain an L-60 matrix, wherein L is the number of samples;
calculating the distance between each row of the L-60 matrix to eliminate singular samples, wherein the calculation method comprises the following steps of firstly calculating the Fisher distance between row vectors, and the calculation formula is as follows:
where μ represents the mean of the vector, σ represents the standard deviation of the vector, Fi,jCalculating the result as a matrix of L x L and defining Fi,jA value of (d);
statistical samples, if there is a vector m, where FimOr FjmExceeding a defined Fi,jIf so, determining m as a singular sample, and removing the singular sample m;
the step of determining the fatigue threshold according to the target sample and the driving fatigue state change characteristics of the driver comprises the following steps:
calculating a matrix of the target samples;
taking M/2 as a step length, averaging the matrix of the target sample according to rows, then averaging according to columns, and finally obtaining two numbers, namely a first half-segment numerical value and a second half-segment numerical value;
and determining the second half segment value as a fatigue threshold value.
2. The driving fatigue detection method according to claim 1, wherein the step of calculating the synchronicity of the two leads of each sample vector set employs a phase synchronization calculation method, specifically:
the calculation method of the phase-locked value PLV between the two leads is as follows:
therein
Is a time series xi(t) and xj(t) calculating the phase change of the time domain signal using a hilbert transform, the hilbert transform being obtained for a continuous time sequence x (t) by the following equation:
where PV represents the cauchy principal value and the phase change is calculated by the following equation:
3. the driving fatigue detection method according to claim 1, wherein the step of calculating the synchronicity of the two leads of each sample vector set employs a time domain vector synchronization calculation method, specifically:
where t is the time component in the brain electrical signal sample, N is the length of one sampling period, xiAnd xjRespectively representing a sample time series of two electrodes.
4. The driving fatigue detection method of claim 1, wherein the step of collecting electroencephalogram signals of the driver on line, calculating a synchronicity result in real time, and comparing and analyzing the synchronicity result calculated in real time with the fatigue threshold to determine whether the driver is currently in a fatigue state comprises:
acquiring electroencephalogram signals of a driver on line, calculating synchronism once every 1 second to obtain a numerical value, marking the fatigue state of the second as 1 if the numerical value exceeds the fatigue threshold value, otherwise marking the fatigue state of the second as 0, continuously calculating for 1 minute to obtain a marking vector, summing the marking vector, judging that the driver is in the fatigue state at present if the summation result exceeds a preset value, and outputting the fatigue value of the last second.
5. A driving fatigue detection system, the system comprising:
the acquisition module is used for continuously acquiring the two-lead electroencephalogram signals of the forehead of the driver for multiple times according to a preset sampling frequency so as to obtain an electroencephalogram signal sample in a fatigue state;
the dividing module is used for dividing the electroencephalogram signal samples according to the sampling period to form a sample vector set taking the sampling period as the length;
the calculation module is used for calculating the synchronism of the two leads of each sample vector set by adopting a phase synchronization calculation method or a time domain vector synchronization calculation method;
the dispersion module is used for carrying out time period discretization on the calculated synchronism result so as to screen out singular samples and obtain target samples;
the threshold value determining module is used for determining a fatigue threshold value according to the target sample and the driving fatigue state change characteristics of the driver;
the fatigue determining module is used for collecting electroencephalogram signals of a driver on line, calculating a synchronism result in real time, and comparing and analyzing the synchronism result calculated in real time and the fatigue threshold value to determine whether the driver is in a fatigue state at present;
the dividing module is specifically configured to:
truncating the data according to the sampling frequency, forming the samples into a matrix of 2 x 60M x H for a signal of M minutes in one experiment, wherein 60M is the number of seconds of sampling, H is the number of data points acquired in one second, and 2 is two leads;
the discrete module is specifically configured to:
selecting each row in the calculated matrix of the synchronism result, dividing the vectors in the matrix into M vectors according to the step length 60, and then reserving two bits behind a decimal point to obtain an L-60 matrix, wherein L is the number of samples;
calculating the distance between each row of the L-60 matrix to eliminate singular samples, wherein the calculation method comprises the following steps of firstly calculating the Fisher distance between row vectors, and the calculation formula is as follows:
where μ represents the mean of the vector, σ represents the standard deviation of the vector, Fi,jCalculating the result as a matrix of L x L and defining Fi,jA value of (d);
statistical samples, if there is a vector m, where FimOr FjmExceeding a defined Fi,jIf so, determining m as a singular sample, and removing the singular sample m;
the threshold determination module is specifically configured to:
calculating a matrix of the target samples;
taking M/2 as a step length, averaging the matrix of the target sample according to rows, then averaging according to columns, and finally obtaining two numbers, namely a first half-segment numerical value and a second half-segment numerical value;
and determining the second half segment value as a fatigue threshold value.
6. The driving fatigue detection system of claim 5, wherein the calculation module calculates synchronicity of two leads of each sample vector set by using a phase synchronization calculation method, specifically:
the calculation method of the phase-locked value PLV between the two leads is as follows:
therein
Is a time series xi(t) and xj(t) calculating the phase change of the time domain signal using a hilbert transform, the hilbert transform being obtained for a continuous time sequence x (t) by the following equation:
where PV represents the cauchy principal value and the phase change is calculated by the following equation:
7. the driving fatigue detection system of claim 5, wherein the calculation module calculates synchronicity of two leads of each sample vector set by using a time domain vector synchronization calculation method, specifically:
where t is the time component in the brain electrical signal sample, N is the length of one sampling period, xiAnd xjRespectively representing a sample time series of two electrodes.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811576142.7A CN109394248B (en) | 2018-12-22 | 2018-12-22 | Driving fatigue detection method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811576142.7A CN109394248B (en) | 2018-12-22 | 2018-12-22 | Driving fatigue detection method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109394248A CN109394248A (en) | 2019-03-01 |
| CN109394248B true CN109394248B (en) | 2022-02-22 |
Family
ID=65460650
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811576142.7A Active CN109394248B (en) | 2018-12-22 | 2018-12-22 | Driving fatigue detection method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109394248B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111450380A (en) * | 2020-03-30 | 2020-07-28 | 重庆大学 | Driving fatigue intelligent detection warning wake-up system based on near-infrared brain functional imaging |
| CN114435373B (en) * | 2022-03-16 | 2023-12-22 | 一汽解放汽车有限公司 | Fatigue driving detection method, device, computer equipment and storage medium |
| CN116616794B (en) * | 2023-05-18 | 2024-02-20 | 中国人民解放军海军特色医学中心 | Underwater operator fatigue adjustment method and system based on electroencephalogram signals |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105286890A (en) * | 2015-09-22 | 2016-02-03 | 江西科技学院 | Driver sleepy state monitoring method based on electroencephalogram signal |
| CN205582206U (en) * | 2016-04-14 | 2016-09-14 | 江西科技学院 | A brain wave fatigue driving detector |
| CN206007246U (en) * | 2016-06-01 | 2017-03-15 | 南京工程学院 | A kind of wear-type brain concentration degree monitoring and alarm equipment |
| US20170071525A1 (en) * | 2015-09-16 | 2017-03-16 | National Chiao Tung University | Device of drowsiness detection and alarm and method of the same |
| US20180186234A1 (en) * | 2017-01-03 | 2018-07-05 | General Electric Company | Control system and method |
| CN108903958A (en) * | 2018-05-11 | 2018-11-30 | 南京邮电大学 | Driver attention evaluates early warning system and its implementation method |
| CN108926352A (en) * | 2017-05-22 | 2018-12-04 | 北京大学 | A kind of driving fatigue detection method and system |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8872640B2 (en) * | 2011-07-05 | 2014-10-28 | Saudi Arabian Oil Company | Systems, computer medium and computer-implemented methods for monitoring health and ergonomic status of drivers of vehicles |
| KR102287315B1 (en) * | 2017-04-14 | 2021-08-09 | 현대자동차주식회사 | Apparatus and method for controlling vehicle based on degree of fatigue |
-
2018
- 2018-12-22 CN CN201811576142.7A patent/CN109394248B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170071525A1 (en) * | 2015-09-16 | 2017-03-16 | National Chiao Tung University | Device of drowsiness detection and alarm and method of the same |
| CN105286890A (en) * | 2015-09-22 | 2016-02-03 | 江西科技学院 | Driver sleepy state monitoring method based on electroencephalogram signal |
| CN205582206U (en) * | 2016-04-14 | 2016-09-14 | 江西科技学院 | A brain wave fatigue driving detector |
| CN206007246U (en) * | 2016-06-01 | 2017-03-15 | 南京工程学院 | A kind of wear-type brain concentration degree monitoring and alarm equipment |
| US20180186234A1 (en) * | 2017-01-03 | 2018-07-05 | General Electric Company | Control system and method |
| CN108926352A (en) * | 2017-05-22 | 2018-12-04 | 北京大学 | A kind of driving fatigue detection method and system |
| CN108903958A (en) * | 2018-05-11 | 2018-11-30 | 南京邮电大学 | Driver attention evaluates early warning system and its implementation method |
Non-Patent Citations (3)
| Title |
|---|
| 基于二阶盲辨识的运动想象脑电信号分类;胡剑锋等;《中国组织工程研究与临床康复》;20080325;第12卷(第13期);第2481-2484页 * |
| 基于生物电信号的疲劳驾驶检测;周展鹏;《中国优秀硕士学位论文全文数据库信息科技辑》;20140915(第9期);第21-22页，第29-38页 * |
| 基于相位同步的脑电信号分类算法研究;胡剑锋等;《微电子学与计算机》;20080930;第25卷(第9期);第138-140,143页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109394248A (en) | 2019-03-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Azami et al. | Amplitude-aware permutation entropy: Illustration in spike detection and signal segmentation | |
| CN109394248B (en) | Driving fatigue detection method and system | |
| CN107095669B (en) | A method and system for processing EEG signals of patients with epilepsy | |
| Uthayakumar et al. | Epileptic seizure detection in EEG signals using multifractal analysis and wavelet transform | |
| WO2018010107A1 (en) | Method and device for determining whether pulse wave is valid | |
| CN114366103B (en) | Attention assessment method, device and electronic equipment | |
| Turnip et al. | Adaptive principal component analysis based recursive least squares for artifact removal of EEG signals | |
| CN108937920B (en) | Ventricular fibrillation signal detection method, system and ventricular fibrillation detection device | |
| Hadiyoso et al. | Signal Dynamics Analysis for Epileptic Seizure Classification on EEG Signals. | |
| Moser et al. | Classification and detection of single evoked brain potentials using time-frequency amplitude features | |
| CN115399789A (en) | Visual vertigo electroencephalogram signal classification method, device, equipment and medium | |
| CN109953757A (en) | Real-time monitoring method of mental load for continuous tracking control and shooting tasks | |
| de l'Aulnoit et al. | Baseline fetal heart rate analysis: eleven automatic methods versus expert consensus | |
| Fraga et al. | Comparison of higuchi, katz and multiresolution box-counting fractal dimension algorithms for eeg waveform signals based on event-related potentials | |
| CN105825225A (en) | Method for making electroencephalogram target judgment with assistance of machine vision | |
| CN113662564A (en) | An event-evoked potential extraction device based on eigenmode function coherence analysis | |
| EP3182892B1 (en) | Method and system for eeg signal processing | |
| Misiūnas et al. | On the implementation and improvement of automatic EEG spike detection algorithm | |
| WO2018035299A1 (en) | Reliable seizure detection with a parallelizable, multi-trajectory estimate of lyapunov exponents | |
| CN113827256B (en) | Fatigue detection method, device, terminal and storage medium based on brain electrical micro state | |
| WO2024084391A1 (en) | Method for classifying electroencephalography data with the use of artificial neural networks | |
| CN115969398A (en) | Blink detection method and device | |
| EP3713475A1 (en) | System for real-time measurement of the activity of a cognitive function and method for calibrating such a system | |
| Nancy et al. | A brain EEG classification system for the mild cognitive impairment analysis | |
| CN112515686B (en) | Electroencephalogram data processing method and device and computer readable storage medium |
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 |