CN109670479A - A kind of intelligent control method and system - Google Patents
A kind of intelligent control method and system Download PDFInfo
- Publication number
- CN109670479A CN109670479A CN201811635209.XA CN201811635209A CN109670479A CN 109670479 A CN109670479 A CN 109670479A CN 201811635209 A CN201811635209 A CN 201811635209A CN 109670479 A CN109670479 A CN 109670479A
- Authority
- CN
- China
- Prior art keywords
- signal
- eeg signals
- intelligent
- temporo
- video monitoring
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B13/00—Burglar, theft or intruder alarms
- G08B13/18—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength
- G08B13/189—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems
- G08B13/194—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems
- G08B13/196—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems using television cameras
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/02—Preprocessing
- G06F2218/04—Denoising
- G06F2218/06—Denoising by applying a scale-space analysis, e.g. using wavelet analysis
Abstract
The invention proposes a kind of intelligent control method and systems.The method comprise the steps that step 101, acquisition EEG signals and video monitoring signal；Step 103 carries out the first judgement to the EEG signals based on EEG signals model, carries out the second judgement to the video monitoring signal based on intelligent algorithm；Step 105, in conjunction with the first judging result and the second judging result, using corresponding measure.What is provided through the invention is a kind of based on man-machine co-melting smart security guard and alarm method and system, or the dependent events identification function such as operating area automatic alarm and non-standard dressing dangerous to illegal disengaging may be implemented.
Description
Technical field
The present invention relates to monitoring field more particularly to a kind of intelligent control method and systems.
Background
Security product is the important component of social public security product, and security product is directly related to the stabilization of society
With the safety of people's lives and properties.Security industry is the important component of social public security industry.By more than 20 years
Fast development, China's security industry have formd collection research and development, producing and selling, engineering and the system integration, alarm operation and intermediary
The emerging national economy rising industry that service etc. is integrated.Security system on the market is generally basede on image recognition side at present
Method, using the intelligent algorithm of pattern-recognition, prevention larceny case that cannot be comprehensive occurs, and effect is poor, especially has
Smog, block or rapid scene switch when, not as good as human eye it is accurate.Sometimes, human eye sees the suspect image quickly flashed,
It does not react for the moment, and the image monitored in real time is pass by, and is often possible to have ignored at this time again due to various reasons
A little feel uncertain.
Summary of the invention
To solve the above-mentioned problems, the invention proposes a kind of method and system of intelligent monitoring.
Wherein, a kind of intelligent control method comprising:
Step 101, acquisition EEG signals and video monitoring signal；
Step 103 carries out the first judgement to the EEG signals based on EEG signals model, is based on intelligent algorithm pair
The video monitoring signal carries out the second judgement；
Step 105, in conjunction with the first judging result and the second judging result, using corresponding measure.
Wherein, the EEG signals model is established to include the following steps:
The EEG signals of step 201, all leads of acquisition, the EEG signals include: Fp1 (left front volume), Fp2 (right
Forehead), F3 (left volume), F4 (right volume), C3 (left centre), C4 (right median), P3 (left top), P4 (right top), O1 (left pillow), O2
(right pillow), F7 (left front temporo), F8 (temporo before right), T3 (left temporo), T4 (temporo in the right side), T5 (left back temporo), T6 (temporo after right).To adopting
The EEG signals of collection are pre-processed, and the first signal is obtained；
Step 203 carries out feature extraction to first signal；
Step 205 carries out selection sort to the signal after feature extraction.
Wherein, the step 203 the following steps are included:
Step 2011 carries out total average reference (Common Average to the EEG signals of all lead acquisitions
Reference, CAR), obtain a qualified reference electrode；
Step 2013 makes the difference C3, C4 lead signals and reference electrode current potential in P3, P4 of top and middle area, calculates institute
Obtaining current potential is respectively P3 ', P4 ', C3 ', C4 ' potential data.
Step 2015 takes the bandpass filter that frequency range is 0.5-30HZ to carry out band to P3 ', P4 ', C3 ', C4 ' potential data
Pass filter processing；
Step 2017 handles the P3 ' after above-mentioned bandpass filtering, P4 ', C3 ', C4 ' signal using independent component analysis method
Data, by signal eye movement EOG and myoelectricity EMG artefact remove；
Step 2019, albefaction de-noising carry out modeling analysis to spontaneous brain electricity signal using AR model, spontaneous brain electricity are converted
For white noise, removed it using wavelet noise.
Wherein, it other than three electrodes Fz, Cz, Pz representing central part, is placed on above brain
The position of remaining 16 electrode is all symmetrical.
Wherein, the intelligent algorithm includes image frame differential method, image segmentation algorithm, image alignment algorithm, mesh
Mark tracing algorithm, bayes classification method etc..
The invention also provides a kind of intelligent monitor systems based on the above method comprising:
Eeg signal acquisition device is set to security personnel head, for acquiring EEG signals；
Video monitoring apparatus is used to acquire video monitoring data；
Wherein, the intelligent monitor system further include:
Video display devices are used to show monitor video；
Warning device is used to generate alarm signal to user.
Wherein, the intelligent monitor system further include: serial communication modular, wired camera or wireless network camera.
Wherein, the video display devices include large screen monitor screen, display, mobile phone, tablet computer.
Wherein, the intelligent monitor system: signal acquisition module is used to acquire EEG signals and video monitoring signal；
Judgment module is used to judge the EEG signals based on EEG signals model, be calculated based on artificial intelligence
Method judges the video monitoring signal
What is provided through the invention is a kind of based on man-machine co-melting smart security guard and alarm method and system, may be implemented to non-
Method disengaging danger or the dependent events identification function such as operating area automatic alarm and non-standard dressing.
Detailed description of the invention
Fig. 1 schemes for the characteristics of P300 potential response in the prior art；
Fig. 2 be and a kind of consistent flow chart of intelligent control method of one embodiment of the invention；
Fig. 3 be and a kind of consistent flow chart of intelligent control method of another embodiment of the present invention；
Fig. 4 is and a kind of a certain consistent system block diagram of intelligent monitor system of embodiment of the present invention.
Specific embodiment
Brain-computer interface (BCI) is that a kind of handle is converted into control command from the collected electric signal of human brain and is directly passed to calculating
The device of machine or other machines, effect be establish a brain independently of human body and electronic equipment exchange control channel.
The research of brain-computer interface is related to Neuscience, signal detection, signal processing, the multiple fields such as control realization.Evoked ptential for
It is also highly useful for the operation of BCI system.For example, in the past more than 40 years a large number of studies show that, accidental stimulation meeting
It induces EEG and generates a positive response, this response is referred to as P300 current potential.P300 current potential appears in after stimulation presentation
The quader cortex of 300ms.As shown in Figure 1, for the feature of P300 response, the amplitude of P300 current potential reaches in parietal bone electrode points
Maximum value, and when record point arrival center and front end, P300 current potential is decaying.The appearance of P300 current potential needs while meeting
Four conditions.First, there is a random stimulation time sequence；Second, it is necessary to which finding suitable sorting algorithm can be by this
A little events are divided into two classes；Third must use the classifying rules in the task of subject；4th, must have one in two class events
Class event seldom occurs.P300 is a kind of Evoked ptential, is caused by visual stimulus that nervous system receives under AD HOC
Specific electrical activity betides the specific time, obvious in the distribution of specific brain scalp region signal energy, relatively holds
Easily detection, is suitable for brain computer interface application.
Factor due to influencing P300 is mainly frequency of stimulation and the event degree of correlation, therefore P300 has good time resolution
Rate and stationary phase and brain-computer interface based on P300, which can produce stabilization signal and be repeated several times without user, trains.
As a result, as shown in Figure 2,3, the invention proposes a kind of methods of intelligent monitoring comprising following steps:
Step 101, acquisition EEG signals and video monitoring signal.
In a certain embodiment, the present invention carries out eeg signal acquisition using brain wave acquisition equipment.
Wherein, the brain wave acquisition equipment includes: acquisition Fp1 (left front volume), Fp2 (right forehead), F3 (left volume), F4
(right volume), C3 (left centre), C4 (right median), P3 (left top), P4 (right top), O1 (left pillow), O2 (right pillow), F7 (left front temporo),
F8 (right before temporo), T3 (left temporo), T4 (temporo in the right side), T5 (left back temporo), T6 (temporo after right) 16 positions EEG signals electricity
Pole sensor, reference electrode of the electrode (A1, A2) as corresponding ipsilateral ear-lobe；And electroencephalogramsignal signal acquisition module.In addition to generation
Outside three electrodes Fz, Cz, Pz of table central part, the position for remaining 16 electrode being placed on above brain be all it is symmetrical,
Reference electrode of the electrode (A1, A2) generally as corresponding ipsilateral ear-lobe
The EEG signals of acquisition are sent to signal handling equipment by the brain wave acquisition equipment.
Step 103 carries out the first judgement to the EEG signals based on EEG signals model, is based on intelligent algorithm pair
The video monitoring signal carries out the second judgement.
In a certain embodiment, the EEG signals model is established using following steps:
Step 201 pre-processes the EEG signals of acquisition, obtains the first signal；
In the present invention, the pre-treatment step specifically includes:
Step 2011 carries out total average reference (Common Average Reference, CAR) to the EEG signals,
Obtain a qualified reference electrode.
In a certain embodiment, for ease of calculation, while it is available meet expected reference electrode, be total to
Average reference analyzes data, and all leads are carried out sum-average arithmetic, subtract all leads with the data of each lead
The sum average value of original signal.Formula are as follows:
WhereinFor the potential difference between electrode i and reference electrode, lead number is n.
Step 2013, Conduction choice do C3, C4 lead signals and reference electrode current potential in P3, P4 of top and middle area
Difference, calculating gained current potential is respectively P3 ', P4 ', C3 ', C4 ' potential data.
Step 2015, data filtering, take frequency range be 0.5-30HZ bandpass filter to P3 ', P4 ', C3 ', C4 ' data into
Row processing, inhibits noise, improves signal-to-noise ratio.
Step 2017, independent component analysis (ICA), by independent component analysis source signals, by EOG
The removal of (Electrooculography, EOG) and EMG (Electromyography, EMG) artefact.
In a certain embodiment, source separation is carried out to signal with maximization non-Gaussian system fast independent component analysis method,
EOG and EMG artefact is removed；The model of its matrix form is as follows:
It is abbreviated as X=AS
Wherein X is N-dimensional observation signal matrix, and S is the mutually independent source signal matrix for generating observation signal, observation signal
X (t) is that unknown matrix A linear hybrid of the source signal S (t) by N × M dimension obtains.
Using observation signal with source signal statistical iteration it is assumed that passing through a linear transformation separation matrix W, it is contemplated that obtain
EEG signals can be led to it using ICA analytic approach closer to actual signal S (t) by output signal U (t)=WX (t)=WAS (t)
He interferes separation.
Step 2019, albefaction de-noising carry out modeling analysis to spontaneous brain electricity using AR model, convert spontaneous brain electricity to white
Noise is removed it using wavelet noise.
In a certain embodiment, the AR model is based on this linear system and is formed,
Wherein w (k) is white-noise excitation, and x (k) is any random signal with reasonable power spectral density,
When H (z) only has pole, become autoregression (AR) model, the difference equation of model may be expressed as:
System function are as follows:
aiFor the parameter of model, p is the order of model, we select p=30 here.Spontaneous brain electricity is defeated by system A (z)
Random white noise w (k) out,
Step 203 carries out feature extraction to first signal.
The characteristics of in a certain embodiment according to vision induced P300 current potential, selects few coherence average combination Wigner-
Villa distribution extracts characteristic potential, realizes the extract real-time of P300 feature.
In a certain embodiment, the phase average algorithm is extracted evoked brain potential and is assumed based on following three:
(1) Evoked ptential and noise are additive relations, and the two is mutually indepedent；
(2) the evoked brain potential waveform that stimulation generates every time is constant；
(3) noise is steady, zero-mean random signal, irrelevant between each noise recorded.
If the signal being recorded indicates are as follows:
xi(n)=s (n)+ni(n)
Wherein xiIt (n) is measuring signal, niIt (n) is noise signal, i=1,2,3 ..., N, variable n are n-th in record
Sampled value n=1,2,3 ..., M.
According to assuming initially that obtainThen
If the power of signal is P, noise mean value is zero, and the variance of noise is σ2, at this time signal-to-noise ratio beFrom knot
Fruit can be seen that after n times coherence average, and the power of signal remains as P, and noise mean value is zero, and variance becomes σ2/ N, because
Signal-to-noise ratio after this coherence average are as follows: SNR=PN/ σ2, PSNR power signal-to-noise ratio improves N times.
The Wigner-ville distribution of signal s (t) is defined as follows:
Wherein z (t) is the analytic signal of s (t), it may be assumed that
Wherein t
It is real variable with τ, P.V. indicates to take the main value of integral.
Step 205 carries out selection sort to the signal after feature extraction.
In a certain embodiment, the classification of brain electrical feature is carried out according to Fisher linear discriminant, establishes the brain electricity based on P300
Signal model, can detecte the P300 signal of certain threshold value when vision induced, and recognition correct rate meets desired effect.
Classification, the brain electrical feature classification based on Fisher linear discriminant, carries out r to each lead2Value calculates, and remembers each lead
Join r2The maximum value of value isOnly choose r2Value is greater thanThose of dimension, Codimension omit, according to per one-dimensional each lead
FLDA classifier is established respectively；Assuming that sample training collection includes the sample x of N number of n dimension1,x2,...,xN, wherein N1It is a to belong to ω1
Class, N2It is a to belong to ω2Class, it is clear that have N=N1+N2.If the component to sample does linear combination, scalar can be obtained:
yn=ωTxn, n=1,2 ..., Ni
The set y of N number of one-dimensional sample available in this way1,y2,…,yn, make sample ynWith maximum classification degree, then
Need to find a good transformation vector w*, solution formula are as follows:
In the space N-dimensional X
<1>Different categories of samples mean vector m is calculatedi,
<2>the within class scatter matrix S of Different categories of samples is calculatediWith total within class scatter matrix Sw,
Sw=S1+S2
<3>matrix between samples S is calculatedb,
Sb=(m1-m2)(m1-m2)T
In the one-dimensional space Y
<1>Different categories of samples mean vector is calculated
<2>within class scatter matrix of Different categories of samples is calculatedWith total within class scatter matrix
Fisher criterion function is chosen, it is often desirable that and in the one-dimensional space, Different categories of samples can separate as far as possible, that is,
Say the difference of the mean value of two class samplesIt is the bigger the better, at the same time it is wished that the inside of Different categories of samples function is intensive as far as possible, i.e., it is uncommon
Hope within-cluster variance the smaller the better.Therefore, Fisher criterion function are as follows:
Obviously, it should which searching makes JF(W) molecule is as big as possible, and denominator is as small as possible, i.e. JF(W) w as big as possible makees
For projecting direction.But w is free of in above formula, therefore need to be tried JF(W) become the explicit function containing w.Finally above formula is converted into showing
Formula containing w, and acquire so that JF(W) maximum w optimal solution w*For
Obtain w*After, it can be shown below by original each vector projection to one-dimensional:
Wherein, x is vector to be projected, and y is after projecting as a result, for a constant.A suitable threshold values is determined for y
y0, according to y and y0Size, that is, deducibility vector x to be projected which kind of belongs to.Threshold value y0:
Wherein, y0For the two respective sample averages of class sample, P (ω1)、P(ω2) be respectively two class samples prior probability,
Sample size containing P300 feature in experiment is free from the 1/5 of P300 feature samples quantity.
In a certain embodiment, classification further includes classification results conversion, and the classifier output result of all leads is carried out
Superposition.
Step 105, in conjunction with the first judging result and the second judging result, using corresponding measure.
When the discovery of one of EEG signals and video monitoring signal has specific condition, then alarm signal is generated.The alarm
Signal include sound, light, electricity mode alarm, or by the signal of a certain period to user show.Can by user again into
Whether row verifies alarm verification correct.
In a certain embodiment, as shown in figure 4, the invention also provides a kind of intelligent monitoring system based on the above method
System comprising:
Eeg signal acquisition device is set to security personnel head, for acquiring EEG signals；
Video monitoring apparatus is used to acquire video monitoring data；
Wherein, the system also includes:
Video display devices are used to show monitor video；
Warning device is used to generate alarm signal to user.
Wherein the system also includes serial communication modular, wired camera or wireless network cameras.
Wherein, the video display devices include large screen monitor screen, display, mobile phone, tablet computer.
The invention also provides a kind of intelligent monitor systems comprising:
Signal acquisition module is used to acquire EEG signals and video monitoring signal；
Judgment module is used to judge the EEG signals based on EEG signals model, be calculated based on artificial intelligence
Method judges the video monitoring signal.
The invention also provides a kind of computer-readable mediums, are stored thereon with computer program and make processing execution following
Method:
Step 101, acquisition EEG signals and video monitoring signal；
Step 103 carries out the first judgement to the EEG signals based on EEG signals model, is based on intelligent algorithm pair
The video monitoring signal carries out the second judgement；
Step 105, in conjunction with the first judging result and the second judging result, using corresponding measure.
Wherein, the EEG signals model is established to include the following steps:
The EEG signals of step 201, all leads of acquisition, the EEG signals include: Fp1 (left front volume), Fp2 (right
Forehead), F3 (left volume), F4 (right volume), C3 (left centre), C4 (right median), P3 (left top), P4 (right top), O1 (left pillow), O2
(right pillow), F7 (left front temporo), F8 (temporo before right), T3 (left temporo), T4 (temporo in the right side), T5 (left back temporo), T6 (temporo after right).To adopting
The EEG signals of collection are pre-processed, and the first signal is obtained；
Step 203 carries out feature extraction to first signal；
Step 205 carries out selection sort to the signal after feature extraction.
Wherein, the step 203 the following steps are included:
Step 2011 carries out total average reference (Common Average to the EEG signals of all lead acquisitions
Reference, CAR), obtain a qualified reference electrode；
Step 2013 makes the difference C3, C4 lead signals and reference electrode current potential in P3, P4 of top and middle area, calculates institute
Obtaining current potential is respectively P3 ', P4 ', C3 ', C4 ' potential data.
Step 2015 takes the bandpass filter that frequency range is 0.5-30HZ to carry out band to P3 ', P4 ', C3 ', C4 ' potential data
Pass filter processing；
Step 2017 handles the P3 ' after above-mentioned bandpass filtering, P4 ', C3 ', C4 ' signal using independent component analysis method
Data, by signal eye movement EOG and myoelectricity EMG artefact remove；
Step 2019, albefaction de-noising carry out modeling analysis to spontaneous brain electricity signal using AR model, spontaneous brain electricity are converted
For white noise, removed it using wavelet noise.
Wherein, it other than three electrodes Fz, Cz, Pz representing central part, is placed on above brain
The position of remaining 16 electrode is all symmetrical.
Wherein, the intelligent algorithm includes image frame differential method, image segmentation algorithm, image alignment algorithm, mesh
Mark tracing algorithm, bayes classification method etc..
The embodiment of the present invention can take whole hardware embodiments, whole software implementations or comprising hardware and software
The form of the embodiment of unit.In a preferred embodiment, the present invention is by software realization comprising but be not limited to firmware, be resident it is soft
Part, microcode etc..In addition, can take can or computer-readable media available from the computer that provides program code by the present invention
The form of the computer program product of middle access, the program code be used to being used by computer or any instruction execution system or
It is combined with computer or any instruction execution system.
For explanation, computer is available or computer-readable media can be may include, store, communicating, propagating,
Or transport program any device, described program be used for by instruction execution system, device or equipment using or with instruction execution system
System, device or equipment combine.The media can be electronics, magnetic, optics, electromagnetism, infrared or semiconductor system and (or device or set
It is standby) or communications media.The example of computer-readable media includes semiconductor or solid-state memory, tape, computer can be removed
Floppy disk, random access memory (RAM), read-only memory (ROM), hard disc and CD.The example of current CD includes causing
Close disk-read-only memory (CD-ROM), compact disk-read/write (CD-R/W) and DVD.
Data processing system suitable for storing and/or performing program code will include that at least one is direct by system bus
Or the processor coupled indirectly with storage unit.Storage unit may include the sheet disposed during the practical execution of program code
Ground memory, high-capacity storage and cache memory, the cache memory provide at least certain program code
Temporary storing device to reduce the number that code must be retrieved from high-capacity storage during execution.Input/output
Or I/O equipment (including but not limited to keyboard, display, indicating equipment etc.) can be by directly or through intermediate I/O controller
It is coupled to system.Network adapter is also coupled to system so that data processing system can become through intermediate individual or public affairs
Many networks are coupled to other data processing systems or remote printer or storage equipment.Modem, cable modem
It is only the network adapter of a part of currently available type with Ethernet card.
Although an embodiment of the present invention has been shown and described, it will be understood by those skilled in the art that: not
A variety of change, modification, replacement and modification can be carried out to these embodiments in the case where being detached from the principle of the present invention and objective, this
The range of invention is defined by the claims and their equivalents.
Claims (10)
1. a kind of intelligent control method comprising:
Step 101, acquisition EEG signals and video monitoring signal；
Step 103 carries out the first judgement to the EEG signals based on EEG signals model, based on intelligent algorithm to described
Video monitoring signal carries out the second judgement；
Step 105, in conjunction with the first judging result and the second judging result, using corresponding measure.
2. intelligent control method as described in claim 1, wherein establish the EEG signals model and include the following steps:
The EEG signals of step 201, all leads of acquisition, the EEG signals include: Fp1 (left front volume), Fp2 (before right
Volume), F3 (left volume), F4 (right volume), C3 (left centre), C4 (right median), P3 (left top), P4 (right top), O1 (left pillow), O2 it is (right
Pillow), F7 (left front temporo), F8 (right before temporo), T3 (left temporo), T4 (temporo in the right side), T5 (left back temporo), T6 (temporo after right).To acquisition
EEG signals pre-processed, obtain the first signal；
Step 203 carries out feature extraction to first signal；
Step 205 carries out selection sort to the signal after feature extraction.
3. intelligent control method as described in claim 1, wherein the step 203 the following steps are included:
Step 2011 carries out total average reference (Common Average to the EEG signals of all lead acquisitions
Reference, CAR), obtain a qualified reference electrode；
Step 2013 makes the difference C3, C4 lead signals and reference electrode current potential in P3, P4 of top and middle area, calculates gained electricity
Position is respectively P3 ', P4 ', C3 ', C4 ' potential data.
Step 2015 takes the bandpass filter that frequency range is 0.5-30HZ to carry out band logical filter to P3 ', P4 ', C3 ', C4 ' potential data
Wave processing；
Step 2017 handles the P3 ' after above-mentioned bandpass filtering, P4 ', C3 ', C4 ' signal data using independent component analysis method,
By in signal eye movement EOG and myoelectricity EMG artefact removal；
Step 2019, albefaction de-noising carry out modeling analysis to spontaneous brain electricity signal using AR model, convert spontaneous brain electricity to white
Noise is removed it using wavelet noise.
4. intelligent control method as claimed in claim 2, wherein other than three electrodes Fz, Cz, Pz representing central part,
The position for remaining 16 electrode being placed on above brain is all symmetrical.
5. intelligent control method as described in claim 1, wherein the intelligent algorithm includes image frame differential method, figure
As partitioning algorithm, image alignment algorithm, target tracking algorithm and bayes classification method.
6. such as the intelligent monitor system of claim 1-5 the method comprising:
Eeg signal acquisition device is set to security personnel head, for acquiring EEG signals；
Video monitoring apparatus is used to acquire video monitoring data.
7. intelligent monitor system as claimed in claim 6, further include:
Video display devices are used to show monitor video；
Warning device is used to generate alarm signal to user.
8. intelligent monitor system as claimed in claim 6, further include: serial communication modular, wired camera or wireless network
Video camera.
9. intelligent monitor system as claimed in claim 7, wherein the video display devices include large screen monitor screen, display
Device, mobile phone, tablet computer.
10. such as the intelligent monitor system of claim 1-6 the method comprising:
Signal acquisition module is used to acquire EEG signals and video monitoring signal；
Judgment module is used to judge the EEG signals based on EEG signals model, is based on intelligent algorithm pair
The video monitoring signal is judged.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811635209.XA CN109670479B (en) | 2018-12-29 | 2018-12-29 | Intelligent monitoring method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811635209.XA CN109670479B (en) | 2018-12-29 | 2018-12-29 | Intelligent monitoring method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109670479A true CN109670479A (en) | 2019-04-23 |
| CN109670479B CN109670479B (en) | 2020-07-28 |
Family
ID=66146931
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811635209.XA Active CN109670479B (en) | 2018-12-29 | 2018-12-29 | Intelligent monitoring method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109670479B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112069963A (en) * | 2020-08-28 | 2020-12-11 | 北京理工大学 | Low-quality dangerous target detection method and system based on non-invasive neural signals |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102541261B (en) * | 2012-01-19 | 2015-01-07 | 北京工业大学 | Film editing and selecting auxiliary instrument and realization method based on characteristics of electroencephalogram signal |
| CN105054927A (en) * | 2015-07-16 | 2015-11-18 | 西安交通大学 | Biological quantitative assessment method for active participation degree in lower limb rehabilitation system |
| KR20160107885A (en) * | 2015-03-06 | 2016-09-19 | 이효재 | Door lock system that can open with thoughts |
| US20170258390A1 (en) * | 2016-02-12 | 2017-09-14 | Newton Howard | Early Detection Of Neurodegenerative Disease |
| CN108647645A (en) * | 2018-05-11 | 2018-10-12 | 广州飞宇智能科技有限公司 | A kind of multi-modal Diagnosis of Epilepsy system and method based on video analysis |
| CN108939248A (en) * | 2018-06-06 | 2018-12-07 | 中国人民解放军第四军医大学 | Soldier based on artificial intelligence is wearable to adapt to auxiliary treatment system again |
-
2018
- 2018-12-29 CN CN201811635209.XA patent/CN109670479B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102541261B (en) * | 2012-01-19 | 2015-01-07 | 北京工业大学 | Film editing and selecting auxiliary instrument and realization method based on characteristics of electroencephalogram signal |
| KR20160107885A (en) * | 2015-03-06 | 2016-09-19 | 이효재 | Door lock system that can open with thoughts |
| CN105054927A (en) * | 2015-07-16 | 2015-11-18 | 西安交通大学 | Biological quantitative assessment method for active participation degree in lower limb rehabilitation system |
| US20170258390A1 (en) * | 2016-02-12 | 2017-09-14 | Newton Howard | Early Detection Of Neurodegenerative Disease |
| CN108647645A (en) * | 2018-05-11 | 2018-10-12 | 广州飞宇智能科技有限公司 | A kind of multi-modal Diagnosis of Epilepsy system and method based on video analysis |
| CN108939248A (en) * | 2018-06-06 | 2018-12-07 | 中国人民解放军第四军医大学 | Soldier based on artificial intelligence is wearable to adapt to auxiliary treatment system again |
Non-Patent Citations (2)
| Title |
|---|
| ASHUTOSH SINGANDHUPE: "Reliable Security Algorithm for Drones Using Individual Characteristics From an EEG Signal", 《IEEE ACCESS》 * |
| 程万里: "视觉诱发脑-机接口中脑电信号处理算法研究", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112069963A (en) * | 2020-08-28 | 2020-12-11 | 北京理工大学 | Low-quality dangerous target detection method and system based on non-invasive neural signals |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109670479B (en) | 2020-07-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wang et al. | Channel selection method for EEG emotion recognition using normalized mutual information | |
| Maiorana et al. | On the permanence of EEG signals for biometric recognition | |
| EP3037036B1 (en) | Biometric authentication method and apparatus | |
| CN111329474B (en) | Electroencephalogram identity recognition method and system based on deep learning and information updating method | |
| Gao et al. | A recurrence network-based convolutional neural network for fatigue driving detection from EEG | |
| Gao et al. | Single-trial EEG emotion recognition using Granger Causality/Transfer Entropy analysis | |
| Lee et al. | Real-time physiological and vision monitoring of vehicle driver for non-intrusive drowsiness detection | |
| CN110151203B (en) | Fatigue driving identification method based on multistage avalanche convolution recursive network EEG analysis | |
| US20070060830A1 (en) | Method and system for detecting and classifying facial muscle movements | |
| Pan et al. | Emotion recognition based on EEG using generative adversarial nets and convolutional neural network | |
| CN109993068A (en) | A kind of contactless human emotion's recognition methods based on heart rate and facial characteristics | |
| Jiajie et al. | Clinical decision support system for alcoholism detection using the analysis of EEG signals | |
| CN106991409A (en) | A kind of Mental imagery EEG feature extraction and categorizing system and method | |
| CN110141258A (en) | A kind of emotional state detection method, equipment and terminal | |
| Poorna et al. | EEG based control-a study using wavelet features | |
| Geng et al. | A fusion algorithm for EEG signal processing based on motor imagery brain-computer interface | |
| CN109670479A (en) | A kind of intelligent control method and system | |
| Xie et al. | A spatio-temporal method for extracting gamma-band features to enhance classification in a rapid serial visual presentation task | |
| Rosenthal et al. | Evoked neural responses to events in video | |
| Vahid et al. | Human identification with EEG signals in different emotional states | |
| Imran et al. | Motor imagery EEG signal classification scheme based on wavelet domain statistical features | |
| CN110613446A (en) | Signal processing method and device | |
| Hu et al. | EEG authentication system based on auto-regression coefficients | |
| Farizal et al. | Classifying known/unknown information in the brain using electroencephalography (EEG) signal analysis | |
| Wang | Influence of roi selection for remote photoplethysmography with singular spectrum analysis |
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 | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20230911 Address after: 300192 Tianjin Science and Technology Plaza 5-1-403 (Tiankai Park), West Side of Science and Technology East Road, Nankai District, Tianjin Patentee after: Suishi (Tianjin) Intelligent Technology Co.,Ltd. Address before: 300384 No. 391 Binshui West Road, Xiqing District, Tianjin Patentee before: TIANJIN University OF TECHNOLOGY |