CN107822623A - A kind of driver fatigue and Expression and Action method based on multi-source physiologic information - Google Patents

A kind of driver fatigue and Expression and Action method based on multi-source physiologic information Download PDF

Info

Publication number
CN107822623A
CN107822623A CN201710942971.1A CN201710942971A CN107822623A CN 107822623 A CN107822623 A CN 107822623A CN 201710942971 A CN201710942971 A CN 201710942971A CN 107822623 A CN107822623 A CN 107822623A
Authority
CN
China
Prior art keywords
driver
fatigue
expression
mrow
action
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.)
Pending
Application number
CN201710942971.1A
Other languages
Chinese (zh)
Inventor
谢平
齐孟松
邹策
张艺滢
孙凯
刘兆军
程生翠
杜义浩
何群
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Yanshan University
Original Assignee
Yanshan University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Yanshan University filed Critical Yanshan University
Priority to CN201710942971.1A priority Critical patent/CN107822623A/en
Publication of CN107822623A publication Critical patent/CN107822623A/en
Pending legal-status Critical Current

Links

Classifications

    • 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
    • A61B5/318—Heart-related electrical modalities, e.g. electrocardiography [ECG]
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
    • A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
    • A61B5/1116—Determining posture transitions
    • 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
    • A61B5/165—Evaluating the state of mind, e.g. depression, anxiety
    • 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
    • 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
    • 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
    • A61B5/389—Electromyography [EMG]
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7235—Details of waveform analysis
    • A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7271—Specific aspects of physiological measurement analysis
    • 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)
  • Physics & Mathematics (AREA)
  • Public Health (AREA)
  • Animal Behavior & Ethology (AREA)
  • Biophysics (AREA)
  • Pathology (AREA)
  • Veterinary Medicine (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Medical Informatics (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • General Health & Medical Sciences (AREA)
  • Psychiatry (AREA)
  • Physiology (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Signal Processing (AREA)
  • Psychology (AREA)
  • Dentistry (AREA)
  • Hospice & Palliative Care (AREA)
  • Educational Technology (AREA)
  • Social Psychology (AREA)
  • Child & Adolescent Psychology (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Developmental Disabilities (AREA)
  • Evolutionary Computation (AREA)
  • Fuzzy Systems (AREA)
  • Mathematical Physics (AREA)
  • Cardiology (AREA)
  • Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)

Abstract

The invention discloses a kind of driver fatigue based on multi-source physiologic information and Expression and Action method, comprise the following steps：1st, synchronous acquisition driver brain electricity, electrocardio, myoelectricity and attitude signal；2nd, physiological signal is pre-processed and feature extraction；3rd, the assessment models of fuzzy neural network are built, realize driver fatigue and mood assessments；4th, based on assessment models and using the evaluation index of the continuous learner driver of genetic algorithm, the rule and method of driver fatigue and Expression and Action is extracted, improves and assesses accuracy.The present invention emphasizes comprehensive and sorting technique the advance of decision information, drastically increases the accuracy rate of driver fatigue and Expression and Action, reduces the probability of happening of traffic accident.

Description

A kind of driver fatigue and Expression and Action method based on multi-source physiologic information
Technical field
The present invention relates to automobile assistant driving field, especially a kind of driver fatigue and feelings based on multi-source physiologic information Thread evaluation method.
Background technology
Recently as increasing sharply for China's automobile guarantee-quantity, traffic safety problem and human pilot comfort level body Test and have become social focus.According to incompletely statistics, there is nearly 100,000 people in China because vehicle traffic accident loses life every year, The first in the world was ranked through continuous 10 years, traffic accident turns into the number of casualties is most in the every accident in the whole nation one, wherein due to The traffic accidents accounting more than 35% that the driving conditions such as driving fatigue, burst disease and road anger disease trigger extremely, serious prestige Coerce the security of the lives and property of vast social groups.Therefore, the physiological status such as fatigue strength, mood to influenceing driving behavior is carried out Monitoring and regulation, reduction dangerous driving behavior ensure traffic safety significant.
In recent years, domestic and foreign scholars to physiological signals such as brain electricity and electrocardios and use neural network recognization driver fatigue Degree has carried out substantial amounts of research.Single fatigue detecting is almost used to refer to currently for the evaluation method of driver fatigue Mark (brain electricity or electrocardio), error rate itself is just very high, and uses traditional sorting technique (such as parameter Estimation and Fuzzy regression) Recognition effect is bad.And fuzzy neural network is particularly suitable for use in and builds nonlinearity model, by the row for imitating neutral net Distributed parallel information processing is carried out for characteristic, the purpose of processing information is reached by the weight of training data adjustment neuron.
The content of the invention
It is described present invention aims at offer a kind of driver fatigue and Expression and Action method based on multi-source physiologic information Method collection driver's brain electricity, electrocardio, myoelectricity and attitude information these four physical signs, utilize fuzzy neural network to carry out tired And Expression and Action, it is based ultimately upon the continuous learner driver's fatigue of genetic algorithm and Expression and Action index and extracting rule, raising is commented The accuracy of valency.
To achieve the above object, following technical scheme is employed：The method of the invention comprises the following steps：
Step 1, using MP150WSW synchronous acquisitions EEG signals, electrocardiosignal and electromyographic signal, inertial sensor is passed through Module gathers head attitude information；
Step 2, software is handled using MP150WSW device datas to believe the EEG signals, electrocardiosignal and myoelectricity that collect Number pre-processed, and feature extraction is carried out to physiological signal；
Step 3, exported by the use of brain electricity, electrocardio, myoelectricity and attitude signal as the input of fuzzy neural network as driver Fatigue and emotional state；
Step 4, based on neutral net assessment models, the evaluation index of the continuous learner driver of genetic algorithm, extraction are utilized The rule and method of driver fatigue and Expression and Action.
Further, in step 1, using MP150WSW synchronous acquisitions EEG signals, electrocardiosignal and electromyographic signal；Its In, electrode for encephalograms, using ears electrode as reference electrode, records corresponding EEG signals using international 10-20 system standards； The arrangement of electrocardioelectrode uses breastbone lead mode；Electromyographic electrode is placed at neck；Adopted using the axle inertial sensors of JY901 nine Collect driver head's positional information.
Further, in step 2, pretreated measured data structure EEG signals x (t), electrocardiosignal y (t), flesh Electric signal z (t) and attitude signal h (t)；Extract power spectrum characteristic data-brain electric fatigue index F of brain electricity(α+θ)/β/Fβ/αWith it is anti- Answer the phase between the average power spectra energy values of β ripples/γ ripples of emotional change, cardiac electrical temporal signatures data-heart rate (HR), RR The root mean square (rMSSD) of phase difference, frequency domain character data-mean power of electromyographic signal between standard deviation (SDNN) and adjacent R R Frequency (MPF) and median frequency (MF).
Further, the specific method of the step 3 is as follows：
Fuzzy neural network model is built, based on input of the physiological characteristic index as neutral net in step 2, output For driver fatigue emotional state, the relation of output-input can be written as：
Y=Wlogsig (VX-B)-C (1)
In formula, logsig (x) represents logarithm sigmoid function：
And introduce LM Algorithm for Training neural network parameters value (be input to hidden layer weight matrix V,It is hidden Containing layer to output weight matrix W,Implicit node deviation matrix B,With output node deviation matrix C,) and hidden layer in the concealed nodes that use (i.e. k)；Using the neutral net trained to EEG, ECG of driver, EMG and attitude signal are classified, and are calculated " clear-headed-fatigue " and " happy-sad " of driver.
Further, the specific method of the step 4 is as follows：Based on neural network classification model extraction driver fatigue and The rule and method of Expression and Action；It can be seen that based on neural network classification unit and driven when driver is in fatigue (mood indignation) When, physiological driver's index (x0,x1,x2…x9) inside some specific threshold values；Threshold based on physiological driver's parameter index Value can directly evaluate the fatigue and emotional state of driver；The rule of generation can represent as follows：
R1:X11＜ x1≤X′11andX12＜ x2≤X′12......X19＜ x9≤X′19Then y=1
andX″1＜ x1≤X″′11andX″12＜ x2≤X″′12......X19＜ x9≤X″′19Then y '=1
R2:X21＜ x1≤X′21andX22＜ x2≤X′22......X29＜ x9≤X′29Then y=1
andX″21＜ x1≤X″′21andX″22＜ x2≤X″′22......X″29＜ x9≤X″′29Then y '=1
Rn:Xn1＜ x1≤X′n1andXn2＜ x2≤X′n2......Xn9＜ x9≤X′n9Then y=1
andX″n1＜ x1≤X″′n1andX″n2＜ x2≤X″′n2......X″n9＜ x9≤X″′n9Then y '=1
Use Genetic algorithm searching parameter (X10,X′10,X″10,X″′10……Xn9,X′n9,X″n9,X″′n9), calculated in heredity In method, n is the regular quantity from neutral net extraction；N is bigger, and data caused by neutral net more may be by rule coverage, more More preferable precision may be realized.
Compared with prior art, the invention has the advantages that：
1st, the fatigue and mood of driver's brain electricity, electrocardio, myoelectricity and attitude signal evaluation driver are merged, emphasizes that decision-making is believed What is ceased is comprehensive, and classifying quality is better than monomeric character；
2nd, based on assessment models and using the continuous learner driver of genetic algorithm evaluation index, extract driver fatigue and The rule and method of Expression and Action, the accuracy rate of driver fatigue and Expression and Action is drastically increased, reduces traffic accident Probability of happening.
Brief description of the drawings
Fig. 1 is the work structuring block diagram of the inventive method.
Fig. 2 is the structural representation of MP150WSW equipment in the present invention.
Fig. 3 is fuzzy neural network assessment models structural representation in the present invention.
Fig. 4 is the structural scheme of mechanism of rule discovery system in the present invention.
Fig. 5 is the FB(flow block) of the inventive method.
Drawing reference numeral：1- Inertia informations Acquisition Instrument, 2- electrode for encephalograms, 3- physiology information detectings instrument, 4- electrocardioelectrodes, 5- fleshes Electrode.
Embodiment
The present invention will be further described below in conjunction with the accompanying drawings：
As shown in Figure 1,5, the inventive method comprises the following steps：
Step 1, using MP150WSW synchronous acquisitions EEG signals, electrocardiosignal and electromyographic signal, inertial sensor is passed through Module gathers head attitude information；
Step 2, software is handled using MP150WSW device datas to believe the EEG signals, electrocardiosignal and myoelectricity that collect Number pre-processed, and feature extraction is carried out to physiological signal；
Step 3, exported by the use of brain electricity, electrocardio, myoelectricity and attitude signal as the input of fuzzy neural network as driver Fatigue and emotional state；
Step 4, based on neutral net assessment models, the evaluation index of the continuous learner driver of genetic algorithm, extraction are utilized The rule and method of driver fatigue and Expression and Action.
Synchronous acquisition driver EEG signals, electrocardiosignal, electromyographic signal and attitude signal as shown in Figure 2：
Eeg signal acquisition and processing：Electrode for encephalograms adopt international standards 10-20 electrodes place standard, pass through electrode cap 1 Realize that electrode contacts with scalp, mastoid process is as reference electrode after M1, M2 lead are connected respectively to left and right ear, from 32 top guide skin brains electricity EEG signals corresponding to FP1, FP2, F3, F4, C3, C4, P3, P4 area are selected in collecting device, remove baseline drift, spilling, eye Dynamic and Hz noise, power spectrum characteristic data-brain electric fatigue index F of extraction brain electricity(α+θ)/β/Fβ/αWith reaction emotional change The average power spectra energy value of β ripples/γ ripples；
Ecg signal acquiring and processing：Electrocardioelectrode arrangement uses breastbone lead mode, and electrocardiosignal is pre-processed And electrocardio time domain charactreristic parameter-heart rate (HR) and HRV (HRV) are extracted, wherein HRV mainly chooses and driver fatigue And between the related RR of mood between the standard deviation (SDNN) of phase, adjacent R R phase difference root mean square (rMSSD).
Electromyographic signal collection and processing：Driver's neck electromyographic signal is gathered, electromyographic signal is pre-processed and extracted Myoelectricity frequency domain character parameter-frequency of average power (MPF) and median frequency (MF).
Attitude signal gathers and processing：Driver head's attitude information is gathered using the axle inertial sensors of JY901 nine.
As shown in figure 3, establishing three layers of fuzzy neural network, input layer is above physiological characteristic parameter：Fatigue index F(α+θ)/β(x0), fatigue index Fβ/α(x1), the equal power spectral energies (x of β popins2), the equal power spectral energies (x of γ popins3), heart rate become Change (x4), between RR the phase standard deviation (x5), between adjacent R R phase difference root mean square (x6), myoelectricity MPF (x7)、MF(x8) and head appearance State information (x9).Every group of training data requires that subject takes self-assessment pattern, and 9 grade amounts are carried out to happy degree and lucidity Table (1-9 points) scoring, scoring is higher, and fatigue strength (sad degree) is higher, as shown in table 1.It can be divided into：
The subject's fatigue mood self-assessment statistical form of table 1：
The input vector of fuzzy neural network is X=(x0,x1,x2,x3……x9)T, output vector Y=(y1,y2)T, input Weights to hidden layer areIn formula, k represents the quantity of concealed nodes, hidden layer to output Weights beConcealed nodes deviation B=(b1,b2,b3……bk)T, output node deviation C =(c1,c2)T, the relation of output-input can be written as：
Y=Wlogsig (VX-B)-C (1)
In formula, logsig (x) represents logarithm sigmoid function：
Introducing LM Algorithm for Training neural network parameters value (hidden layer weight matrix V is input to,Hidden layer To output weight matrix W,Implicit node deviation matrix B,With output node deviation matrix C,) and hidden layer in the concealed nodes that use (i.e. k), the setting of the two parameters greatly affected neutral net and estimate The accuracy of meter.Wherein, hidden layer k takes 5 layers, 8 layers, 11 layers, 14 layers, 17 layers and 20 layers respectively, establishes 6 kinds of different nerve nets Network framework simultaneously tests training error.
Driven as shown in figure 4, can be seen that based on neural network classification unit when driver is in fatigue (mood indignation) When, physiological driver's index (x0,x1,x2…x9) inside some specific threshold values.R1, R2, R3 ... Rn are physiological driver Parameter x0, x1, x2 ... x9 domain, n rule of driver fatigue and angry index can be extracted based on these domains：
R1:X11＜ x1≤X′11andX12＜ x2≤X′12......X19＜ x9≤X′19Then y=1
andX″′11＜ x1≤X″′11andX″12＜ x2≤X″′12......X19＜ x9≤X″′19Then y '=1
R2:X21＜ x1≤X′21andX22＜ x2≤X′22......X29＜ x9≤X′29Then y=1
andX″21＜ x1≤X″′21andX″22＜ x2≤X″′22......X″29＜ x9≤X″′29Then y '=1
Rn:Xn1＜ x1≤X′n1andXn2＜ x2≤X′n2......Xn9＜ x9≤X′n9Then y=1
andX″n1＜ x1≤X″′n1andX″n2＜ x2≤X″′n2......X″n9＜ x9≤X″′n9Then y '=1
Wherein, n is the quantity of rule, and n is bigger, and driver is in fatigue and the data of physiological index of indignation may more be advised Then cover, and more likely realize more preferable precision.Therefore, n value is initially set to 1.If genetic algorithm can not meet end Only condition, then the n increases by 1 before the operation of next genetic algorithm.With n increase, more preferable accuracy rate is more likely realized.
Genetic algorithm creates random population P (t)={ α first1(t),α2(t),α3(t)……αm(t) }, wherein t= 0, Swarm Evolution number is represented, m is the number gone here and there in colony, α1(t) physiological parameter of n rule is represented, can be expressed as：α1 (t)={ x11,x′11,x″11,x″′11,x12,x′12,x″12,x″′12……xn9,x′n9,x′n9,x″′n9}.It is then based on fitness Function assesses each character string, and fitness score distributed into each string, string selection course be score based on each string come Determine the selection of follow-on potential string.Use wheel disc algorithms selection character string α1(t),α2(t),α3(t)……αm(t), enter Next evolution colony P (t), finally by the evolution for intersecting and being mutated progress character string.The fitness function of each character string It is defined as：
Wherein α and β is the sensitivity and specificity of the string representation fatigue state respectively, and α ' and β ' are the character respectively String represents the sensitivity and specificity of emotional state, and λ ∈ [0,1] are the essential constant values for controlling sensitivity and specificity. α, α ', β, β ' are defined as follows
Wherein NTPFor the number of true positives, it is meant that tired driver is correctly judged as fatigue；NFNIt is the number of false negative Amount, it is meant that tired driver is clear-headed by false judgment；NFPIt is the quantity of false positive, it is meant that waking state is by false judgment For fatigue；NTNIt is the number of true negative, it is meant that waking state correctly is judged as regaining consciousness.NTP' be true positives number, meaning Taste mood sadness driver and is correctly judged as sadness；NFN' be false negative quantity, it is meant that mood sadness driver is wrong Erroneous judgement is happiness；NFP' be false positive quantity, it is meant that mood happiness driver by false judgment for sadness；NTN' it is true Negative number, it is meant that mood happiness driver is correctly judged as happiness.α and β is maximized by using genetic algorithm, Character string is improved by evolving, intersecting and be mutated, until meeting end condition.
Intersect and predominantly select two parent strings to carry out intersecting generation new character strings, a new character strings { x11,x′11, x″11,x″′11,x12,x′12,x″12,x″′12……xn9,x′n9,x″n9,x″′n9Generation be shown below：
Wherein, the zoom factor randomly selected in α ∈ [- 0.5,1.5].
One or more of character string parameter value selected by random change is sported to perform operation.Such as parameter xijPerform After mutation, its value is changed into：
x′ij=xij+a×Rij×δ (9)
Wherein, a be equiprobability take at random 1 or -1, RijFor xijThe half of upper and lower bound sum, δ are δ ∈ [0,1].
Newly-generated character string is reinserted into old colony P (t) at random and produces new colony P (t+1) of future generation, was evolved Cheng Chixu is carried out, and until meeting end condition, i.e., only need to find the character string with α, α ' ＞ 0.75 and β, β ' ＞ 0.6.
Embodiment described above is only that the preferred embodiment of the present invention is described, not to the model of the present invention Enclose and be defined, on the premise of design spirit of the present invention is not departed from, technical side of the those of ordinary skill in the art to the present invention The various modifications and improvement that case is made, it all should fall into the protection domain of claims of the present invention determination.

Claims (5)

1. a kind of driver fatigue and Expression and Action method based on multi-source physiologic information, it is characterised in that methods described includes Following steps：
Step 1, using MP150WSW synchronous acquisitions EEG signals, electrocardiosignal and electromyographic signal, inertial sensor module is passed through Gather head attitude information；
Step 2, the EEG signals, electrocardiosignal and electromyographic signal that collect are entered using MP150WSW device datas processing software Row pretreatment, and feature extraction is carried out to physiological signal；
Step 3, exported by the use of brain electricity, electrocardio, myoelectricity and attitude signal as the input of fuzzy neural network as the tired of driver Labor and emotional state；
Step 4, based on neutral net assessment models, using the evaluation index of the continuous learner driver of genetic algorithm, extraction drives The rule and method of member's fatigue and Expression and Action.
2. a kind of driver fatigue and Expression and Action method based on multi-source physiologic information according to claim 1, it is special Sign is：In step 1, using MP150WSW synchronous acquisitions EEG signals, electrocardiosignal and electromyographic signal；Wherein, brain electricity electricity Pole, using ears electrode as reference electrode, records corresponding EEG signals using international 10-20 system standards；Electrocardioelectrode Arrangement uses breastbone lead mode；Electromyographic electrode is placed at neck；Driver's head is gathered using the axle inertial sensors of JY901 nine Portion's positional information.
3. a kind of driver fatigue and Expression and Action method based on multi-source physiologic information according to claim 1, it is special Sign is：In step 2, pretreated measured data structure EEG signals x (t), electrocardiosignal y (t), electromyographic signal z (t) With attitude signal h (t)；Extract power spectrum characteristic data-brain electric fatigue index F of brain electricity(α+θ)/β/Fβ/αWith reaction emotional change The average power spectra energy values of β ripples/γ ripples, cardiac electrical temporal signatures data-heart rate (HR), between RR the phase standard deviation (SDNN) root mean square (rMSSD) of phase difference, frequency domain character data-frequency of average power of electromyographic signal between adjacent R R And median frequency (MF) (MPF).
4. a kind of driver fatigue and Expression and Action method based on multi-source physiologic information according to claim 1, it is special Sign is that the specific method of the step 3 is as follows：
Fuzzy neural network model is built, based on input of the physiological characteristic index as neutral net in step 2, is exported to drive The person's of sailing fatigue emotional state, the relation of output-input can be written as：
Y=Wlogsig (VX-B)-C (1)
In formula, logsig (x) represents logarithm sigmoid function：
<mrow> <mi>log</mi> <mi> </mi> <mi>s</mi> <mi>i</mi> <mi>g</mi> <mrow> <mo>(</mo> <mi>x</mi> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mn>1</mn> <mrow> <mn>1</mn> <mo>+</mo> <msup> <mi>e</mi> <mrow> <mo>-</mo> <mi>x</mi> </mrow> </msup> </mrow> </mfrac> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>2</mn> <mo>)</mo> </mrow> </mrow>
And introduce LM Algorithm for Training neural network parameters value (be input to hidden layer weight matrix V,Hidden layer To output weight matrix W,Implicit node deviation matrix B,With output node deviation matrix C,) and hidden layer in the concealed nodes that use (i.e. k)；Using the neutral net trained to EEG, ECG of driver, EMG and attitude signal are classified, and are calculated " clear-headed-fatigue " and " happy-sad " of driver.
5. a kind of driver fatigue and Expression and Action method based on multi-source physiologic information according to claim 1, it is special Sign is that the specific method of the step 4 is as follows：Based on neural network classification model extraction driver fatigue and Expression and Action Rule and method；It is can be seen that based on neural network classification unit when driver is in fatigue (mood indignation) and driven, driver Physical signs (x0,x1,x2…x9) inside some specific threshold values；Threshold value based on physiological driver's parameter index can be straight Connect the fatigue and emotional state of evaluation driver；The rule of generation can represent as follows：
Use Genetic algorithm searching parameter (X10,X1′0,X″10,X″′10……Xn9,X′n9,X″n9,X″′n9), in genetic algorithm, n It is the regular quantity from neutral net extraction；N is bigger, and data caused by neutral net more may be by rule coverage, more likely in fact Now more preferable precision.
CN201710942971.1A 2017-10-11 2017-10-11 A kind of driver fatigue and Expression and Action method based on multi-source physiologic information Pending CN107822623A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710942971.1A CN107822623A (en) 2017-10-11 2017-10-11 A kind of driver fatigue and Expression and Action method based on multi-source physiologic information

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710942971.1A CN107822623A (en) 2017-10-11 2017-10-11 A kind of driver fatigue and Expression and Action method based on multi-source physiologic information

Publications (1)

Publication Number Publication Date
CN107822623A true CN107822623A (en) 2018-03-23

Family

ID=61647750

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710942971.1A Pending CN107822623A (en) 2017-10-11 2017-10-11 A kind of driver fatigue and Expression and Action method based on multi-source physiologic information

Country Status (1)

Country Link
CN (1) CN107822623A (en)

Cited By (29)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108711204A (en) * 2018-05-18 2018-10-26 长安大学 A kind of the driving abnormality detection system and method for comprehensive people-Che-road multi-source information
CN108871802A (en) * 2018-08-14 2018-11-23 安徽江淮汽车集团股份有限公司 A kind of neck muscle potentiometric detection test method of automotive performance
CN109035960A (en) * 2018-06-15 2018-12-18 吉林大学 Driver's driving mode analysis system and analysis method based on simulation driving platform
CN109508783A (en) * 2018-12-28 2019-03-22 杭州翼兔网络科技有限公司 Mood incorporates roughly model construction and the automatic progress rough acquisition methods of mood into
CN109685149A (en) * 2018-12-28 2019-04-26 杭州翼兔网络科技有限公司 Mood finely incorporates model construction and the automatic method for carrying out mood and finely obtaining into
CN109872183A (en) * 2019-01-16 2019-06-11 深圳壹账通智能科技有限公司 Intelligent Service evaluation method, computer readable storage medium and terminal device
CN109875584A (en) * 2019-03-01 2019-06-14 广东轻工职业技术学院 The detection method and its warning system of physiological driver's fatigue
CN109934171A (en) * 2019-03-14 2019-06-25 合肥工业大学 Driver's passiveness driving condition online awareness method based on layered network model
CN110151166A (en) * 2019-05-07 2019-08-23 中国电子科技集团公司第二十八研究所 A kind of fatigue strength assessment system and appraisal procedure based on bi-mode biology electric signal and physiological data
CN110322016A (en) * 2018-03-28 2019-10-11 中华电信股份有限公司 Analysis server apparatus and its optimal combination analysis method
CN110569968A (en) * 2019-10-18 2019-12-13 上海大学 Method and system for evaluating entrepreneurship failure resilience based on electrophysiological signals
CN111150411A (en) * 2020-01-17 2020-05-15 哈尔滨工业大学 Psychological stress evaluation grading method based on improved genetic algorithm
CN111657923A (en) * 2020-07-02 2020-09-15 哈尔滨工业大学 Method and system for testing danger perception capability of driver
CN112052905A (en) * 2020-09-11 2020-12-08 重庆科技学院 Method for extracting multi-operation fatigue features of driver based on recurrent neural network
CN112154490A (en) * 2018-03-28 2020-12-29 罗伯特·博世有限公司 In-vehicle system for estimating scene inside vehicle cabin
CN112155577A (en) * 2020-10-15 2021-01-01 深圳大学 Social pressure detection method and device, computer equipment and storage medium
CN112188597A (en) * 2018-07-25 2021-01-05 Oppo广东移动通信有限公司 Proximity-aware network creation method and related product
CN112220479A (en) * 2020-09-04 2021-01-15 陈婉婷 Genetic algorithm-based examined individual emotion judgment method, device and equipment
CN112957044A (en) * 2021-02-01 2021-06-15 上海理工大学 Driver emotion recognition system based on double-layer neural network model
CN113158973A (en) * 2021-05-12 2021-07-23 合肥工业大学 Driver emotion intensity measurement method based on fuzzy classification calculation
CN113191212A (en) * 2021-04-12 2021-07-30 合肥综合性国家科学中心人工智能研究院(安徽省人工智能实验室) Driver road rage risk early warning method and system
CN113633288A (en) * 2021-06-25 2021-11-12 杭州电子科技大学 EEMD-based fatigue electroencephalogram feature extraction method
CN114324350A (en) * 2021-12-10 2022-04-12 巢湖学院 Intelligent five-hole socket panel defect detection method and system based on machine vision
CN114343638A (en) * 2022-01-05 2022-04-15 河北体育学院 Fatigue degree evaluation method and system based on multi-modal physiological parameter signals
CN114699080A (en) * 2022-04-28 2022-07-05 电子科技大学 Driver mental stress degree identification method based on fusion characteristics
CN115281676A (en) * 2022-10-08 2022-11-04 齐鲁工业大学 Fatigue detection method based on GRU neural network and ECG signal
CN116671881A (en) * 2023-08-03 2023-09-01 北京九叁有方物联网科技有限公司 Head-wearing brain body operation capability assessment device and method based on graph neural network
CN117236802A (en) * 2023-11-15 2023-12-15 青岛民航凯亚系统集成有限公司 Civil airport operation navigability assessment method and system
CN117643470A (en) * 2024-01-30 2024-03-05 武汉大学 Fatigue driving detection method and device based on electroencephalogram interpretation

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN202183155U (en) * 2011-07-26 2012-04-04 山东大学 Fatigue driving prevention assistant system based on gesture identification
CN103431976A (en) * 2013-07-19 2013-12-11 燕山大学 Lower limb rehabilitation robot system based on myoelectric signal feedback, and control method thereof
CN104127195A (en) * 2014-07-29 2014-11-05 杭州电子科技大学 Electroencephalogram signal and electrocardiogram signal integrated method for detecting fatigue driving
US20160090097A1 (en) * 2014-09-29 2016-03-31 The Boeing Company System for fatigue detection using a suite of physiological measurement devices
CN105496407A (en) * 2016-01-17 2016-04-20 仲佳 Reminding device and method thereof
CN105877766A (en) * 2016-06-21 2016-08-24 东北大学 Mental state detection system and method based on multiple physiological signal fusion
CN105943052A (en) * 2016-05-30 2016-09-21 深圳先进技术研究院 A fatigue driving detection method and device
CN107007291A (en) * 2017-04-05 2017-08-04 天津大学 Intense strain intensity identifying system and information processing method based on multi-physiological-parameter
CN107280694A (en) * 2017-07-18 2017-10-24 燕山大学 A kind of fatigue detection method based on Multi-source Information Fusion

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN202183155U (en) * 2011-07-26 2012-04-04 山东大学 Fatigue driving prevention assistant system based on gesture identification
CN103431976A (en) * 2013-07-19 2013-12-11 燕山大学 Lower limb rehabilitation robot system based on myoelectric signal feedback, and control method thereof
CN104127195A (en) * 2014-07-29 2014-11-05 杭州电子科技大学 Electroencephalogram signal and electrocardiogram signal integrated method for detecting fatigue driving
US20160090097A1 (en) * 2014-09-29 2016-03-31 The Boeing Company System for fatigue detection using a suite of physiological measurement devices
CN105496407A (en) * 2016-01-17 2016-04-20 仲佳 Reminding device and method thereof
CN105943052A (en) * 2016-05-30 2016-09-21 深圳先进技术研究院 A fatigue driving detection method and device
CN105877766A (en) * 2016-06-21 2016-08-24 东北大学 Mental state detection system and method based on multiple physiological signal fusion
CN107007291A (en) * 2017-04-05 2017-08-04 天津大学 Intense strain intensity identifying system and information processing method based on multi-physiological-parameter
CN107280694A (en) * 2017-07-18 2017-10-24 燕山大学 A kind of fatigue detection method based on Multi-source Information Fusion

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
李战芬 等: "基于神经网络遗传算法优化的曲轴疲劳寿命预测", 《中北大学学报（自然科学版）》 *
赵婷 等: "草原公路短时程驾驶疲劳程度多指标划分研究", 《中国安全科学学报》 *
闫楚良 等: "基于遗传算法优化的BP神经网络的材料疲劳寿命预测", 《吉林大学学报（工学版）》 *

Cited By (40)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110322016A (en) * 2018-03-28 2019-10-11 中华电信股份有限公司 Analysis server apparatus and its optimal combination analysis method
CN112154490A (en) * 2018-03-28 2020-12-29 罗伯特·博世有限公司 In-vehicle system for estimating scene inside vehicle cabin
CN108711204A (en) * 2018-05-18 2018-10-26 长安大学 A kind of the driving abnormality detection system and method for comprehensive people-Che-road multi-source information
CN109035960A (en) * 2018-06-15 2018-12-18 吉林大学 Driver's driving mode analysis system and analysis method based on simulation driving platform
CN112188597A (en) * 2018-07-25 2021-01-05 Oppo广东移动通信有限公司 Proximity-aware network creation method and related product
CN112188597B (en) * 2018-07-25 2023-11-03 Oppo广东移动通信有限公司 Method for creating proximity-aware network and related product
CN108871802A (en) * 2018-08-14 2018-11-23 安徽江淮汽车集团股份有限公司 A kind of neck muscle potentiometric detection test method of automotive performance
CN109685149A (en) * 2018-12-28 2019-04-26 杭州翼兔网络科技有限公司 Mood finely incorporates model construction and the automatic method for carrying out mood and finely obtaining into
CN109508783A (en) * 2018-12-28 2019-03-22 杭州翼兔网络科技有限公司 Mood incorporates roughly model construction and the automatic progress rough acquisition methods of mood into
CN109508783B (en) * 2018-12-28 2021-07-20 沛誉(武汉)科技有限公司 Method for constructing rough emotion classifying model and automatically performing rough emotion acquisition
CN109872183A (en) * 2019-01-16 2019-06-11 深圳壹账通智能科技有限公司 Intelligent Service evaluation method, computer readable storage medium and terminal device
CN109875584A (en) * 2019-03-01 2019-06-14 广东轻工职业技术学院 The detection method and its warning system of physiological driver's fatigue
CN109934171A (en) * 2019-03-14 2019-06-25 合肥工业大学 Driver's passiveness driving condition online awareness method based on layered network model
CN110151166A (en) * 2019-05-07 2019-08-23 中国电子科技集团公司第二十八研究所 A kind of fatigue strength assessment system and appraisal procedure based on bi-mode biology electric signal and physiological data
CN110569968A (en) * 2019-10-18 2019-12-13 上海大学 Method and system for evaluating entrepreneurship failure resilience based on electrophysiological signals
CN110569968B (en) * 2019-10-18 2021-09-03 上海大学 Method and system for evaluating entrepreneurship failure resilience based on electrophysiological signals
CN111150411B (en) * 2020-01-17 2022-11-11 哈尔滨工业大学 Psychological stress evaluation grading method based on improved genetic algorithm
CN111150411A (en) * 2020-01-17 2020-05-15 哈尔滨工业大学 Psychological stress evaluation grading method based on improved genetic algorithm
CN111657923A (en) * 2020-07-02 2020-09-15 哈尔滨工业大学 Method and system for testing danger perception capability of driver
CN112220479A (en) * 2020-09-04 2021-01-15 陈婉婷 Genetic algorithm-based examined individual emotion judgment method, device and equipment
CN112052905B (en) * 2020-09-11 2023-02-03 重庆科技学院 Method for extracting multi-operation fatigue features of driver based on recurrent neural network
CN112052905A (en) * 2020-09-11 2020-12-08 重庆科技学院 Method for extracting multi-operation fatigue features of driver based on recurrent neural network
CN112155577A (en) * 2020-10-15 2021-01-01 深圳大学 Social pressure detection method and device, computer equipment and storage medium
CN112155577B (en) * 2020-10-15 2023-05-05 深圳大学 Social pressure detection method and device, computer equipment and storage medium
CN112957044A (en) * 2021-02-01 2021-06-15 上海理工大学 Driver emotion recognition system based on double-layer neural network model
CN113191212A (en) * 2021-04-12 2021-07-30 合肥综合性国家科学中心人工智能研究院(安徽省人工智能实验室) Driver road rage risk early warning method and system
CN113191212B (en) * 2021-04-12 2022-06-07 合肥中聚源智能科技有限公司 Driver road rage risk early warning method and system
CN113158973B (en) * 2021-05-12 2022-08-30 合肥工业大学 Driver emotion intensity measurement method based on fuzzy classification calculation
CN113158973A (en) * 2021-05-12 2021-07-23 合肥工业大学 Driver emotion intensity measurement method based on fuzzy classification calculation
CN113633288A (en) * 2021-06-25 2021-11-12 杭州电子科技大学 EEMD-based fatigue electroencephalogram feature extraction method
CN114324350A (en) * 2021-12-10 2022-04-12 巢湖学院 Intelligent five-hole socket panel defect detection method and system based on machine vision
CN114343638A (en) * 2022-01-05 2022-04-15 河北体育学院 Fatigue degree evaluation method and system based on multi-modal physiological parameter signals
CN114343638B (en) * 2022-01-05 2023-08-22 河北体育学院 Fatigue degree assessment method and system based on multi-mode physiological parameter signals
CN114699080A (en) * 2022-04-28 2022-07-05 电子科技大学 Driver mental stress degree identification method based on fusion characteristics
CN114699080B (en) * 2022-04-28 2023-04-25 电子科技大学 Driver mental stress degree identification method based on fusion characteristics
CN115281676A (en) * 2022-10-08 2022-11-04 齐鲁工业大学 Fatigue detection method based on GRU neural network and ECG signal
CN116671881A (en) * 2023-08-03 2023-09-01 北京九叁有方物联网科技有限公司 Head-wearing brain body operation capability assessment device and method based on graph neural network
CN117236802A (en) * 2023-11-15 2023-12-15 青岛民航凯亚系统集成有限公司 Civil airport operation navigability assessment method and system
CN117643470A (en) * 2024-01-30 2024-03-05 武汉大学 Fatigue driving detection method and device based on electroencephalogram interpretation
CN117643470B (en) * 2024-01-30 2024-04-26 武汉大学 Fatigue driving detection method and device based on electroencephalogram interpretation

Similar Documents

Publication Publication Date Title
CN107822623A (en) A kind of driver fatigue and Expression and Action method based on multi-source physiologic information
US20200367800A1 (en) Method for identifying driving fatigue based on cnn-lstm deep learning model
CN105877766B (en) A kind of state of mind detection system and method based on the fusion of more physiological signals
CN102184415B (en) Electroencephalographic-signal-based fatigue state recognizing method
CN102499676B (en) Effective time sequence and electrode recombination based electroencephalograph signal categorizing system and method
CN102319067B (en) Nerve feedback training instrument used for brain memory function improvement on basis of electroencephalogram
CN108388348A (en) A kind of electromyography signal gesture identification method based on deep learning and attention mechanism
CN104166548B (en) Deep learning method based on Mental imagery eeg data
CN108216254A (en) The road anger Emotion identification method merged based on face-image with pulse information
CN108464839A (en) A kind of vehicle-mounted fatigue monitoring and early warning of driving system based on machine learning
CN111544015B (en) Cognitive power-based control work efficiency analysis method, device and system
CN104586387A (en) Method for extracting and fusing time, frequency and space domain multi-parameter electroencephalogram characters
CN104490390A (en) Electrophysiological signal conjoint analysis-based human exercise ability determination method
CN101828921A (en) Identity identification method based on visual evoked potential (VEP)
CN107095671A (en) Driving fatigue recognition methods and system
CN103472922A (en) Destination selecting system based on P300 and SSVEP (Steady State Visual Evoked Potential) hybrid brain-computer interface
CN104503580A (en) Identification method of steady-state visual evoked potential brain-computer interface target
CN111598453B (en) Control work efficiency analysis method, device and system based on execution force in virtual scene
CN104035563B (en) W-PCA (wavelet transform-principal component analysis) and non-supervision GHSOM (growing hierarchical self-organizing map) based electrocardiographic signal identification method
CN106128032A (en) A kind of fatigue state monitoring and method for early warning and system thereof
CN104548347A (en) Pure idea nerve muscle electrical stimulation control and nerve function evaluation system
CN111523601A (en) Latent emotion recognition method based on knowledge guidance and generation counterstudy
CN104461007A (en) Driver-car interactive system assisting driver based on electroencephalograms
CN109567832A (en) A kind of method and system of the angry driving condition of detection based on Intelligent bracelet
Yang et al. Real-time driver cognitive workload recognition: Attention-enabled learning with multimodal information fusion

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
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20180323

WD01 Invention patent application deemed withdrawn after publication