CN103164026B - Based on brain-machine interface method and the device of the fractal intercept feature of box peacekeeping - Google Patents
Based on brain-machine interface method and the device of the fractal intercept feature of box peacekeeping Download PDFInfo
- Publication number
- CN103164026B CN103164026B CN201310095778.0A CN201310095778A CN103164026B CN 103164026 B CN103164026 B CN 103164026B CN 201310095778 A CN201310095778 A CN 201310095778A CN 103164026 B CN103164026 B CN 103164026B
- Authority
- CN
- China
- Prior art keywords
- eeg signals
- box
- intercept
- brain
- fractal
- 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.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 37
- 210000004556 brain Anatomy 0.000 claims abstract description 41
- 230000005611 electricity Effects 0.000 claims abstract description 25
- 238000000605 extraction Methods 0.000 claims abstract description 11
- 238000001514 detection method Methods 0.000 claims abstract description 8
- 238000001914 filtration Methods 0.000 claims description 18
- 238000002474 experimental method Methods 0.000 claims description 15
- 238000012549 training Methods 0.000 claims description 13
- 210000004934 left little finger Anatomy 0.000 claims description 10
- 230000033001 locomotion Effects 0.000 claims description 9
- 238000012360 testing method Methods 0.000 claims description 9
- 239000000284 extract Substances 0.000 claims description 6
- 230000008859 change Effects 0.000 claims description 4
- 230000000694 effects Effects 0.000 abstract description 4
- 230000003340 mental effect Effects 0.000 abstract description 3
- 230000008569 process Effects 0.000 abstract description 3
- 230000006870 function Effects 0.000 description 4
- 210000003710 cerebral cortex Anatomy 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000013461 design Methods 0.000 description 2
- 210000003414 extremity Anatomy 0.000 description 2
- 230000033764 rhythmic process Effects 0.000 description 2
- 208000012661 Dyskinesia Diseases 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 201000010901 lateral sclerosis Diseases 0.000 description 1
- 208000005264 motor neuron disease Diseases 0.000 description 1
- 201000006938 muscular dystrophy Diseases 0.000 description 1
- 230000002232 neuromuscular Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 210000000278 spinal cord Anatomy 0.000 description 1
- 208000020431 spinal cord injury Diseases 0.000 description 1
Abstract
The present invention relates to the brain-machine interface method based on the fractal intercept feature of box peacekeeping and device, based on the detection of hardware platform realization to brain electricity condition that eeg amplifier and computing machine are formed.First gather EEG signals by eeg amplifier and data collecting card, then the EEG signals collected is delivered to computing machine and process, realize the feature extraction of the fractal intercept of box peacekeeping, and complete the classification to EEG signals by Boosting sorter.The present invention utilizes the fractal intercept of characteristic effect good box peacekeeping to carry out feature extraction to EEG signals, and by Boosting sorter, obtains the key words sorting to brain-computer interface Mental imagery EEG signals.
Description
Technical field
The present invention relates to the brain-machine interface method based on the fractal intercept feature of box peacekeeping and device, belong to the technical field of brain-computer interface.
Technical background
There is a lot of patient in actual life because suffering from serious dyskinesia, such as spinal cord injury or muscular dystrophy lateral sclerosis of spinal cord (ALS) etc., and losing the basic ability of carrying out language or limbs with the external world and linking up.This has had a strong impact on the quality of life of patient, causes great burden also to its family and society.Brain-computer interface (BCI) is by setting up a kind of man-machine interactive system not relying on conventional brain information output channel between human brain and the external world.Brain-computer interface technology all has a wide range of applications at the numerous areas such as rehabilitation medical, military affairs.
The different body part cerebral cortex region activated of moving is also different; Monolateral limb motion or imagination motion can activate main sensorimotor cortex, brain offside produces Event-related desynchronization current potential ERD (Event Related Desynchronization), and brain homonymy produces event-related design current potential ERS (Event Related Synchronization); ERD refers to that, when a certain cortical region is active, the periodic activity of characteristic frequency shows the reduction of amplitude, and ERS refers to when a certain activity does not make relevant cortical region active significantly in certain moment, and characteristic frequency just shows amplitude and raises.Electrophysiologic studies shows, Mental imagery can cause the change of brain wave rhythm.Mental imagery can cause frequency to be the u rhythm and pace of moving things of 8-12Hz and frequency being amplitude compacting and the Event-related desynchronization ERD of the beta response of 13-28Hz, or amplitude increase and event-related design ERS.
BCI technology, by extracting the brain electric information of user, then utilizes some machine algorithms that the different conditions of brain is converted into controlling order, and then realizes the control to external unit.The object of BCI sets up a system that user can be helped directly to carry out exchanging with the external world, and need not by means of traditional neuromuscular approach, and wherein, seeking effective feature extracting method is one of gordian technique improving discrimination.Identical feature uses different sorters to classify, and the result of gained also can be different.Therefore, while selection feature, the selection of sorter is also most important.
The method of existing various features extraction at present, as adaptive spatial domain pattern, band power, AR model etc. altogether.2007, the paper " Preprocessing and meta-classification for brain-computer interfaces " that the people such as Hammon PS deliver on IEEE Transactions on Biomedical Engineering proposes a kind of method of pre-service and multi-categorizer, achieves good result.But the pre-service of the method and aftertreatment all more complicated, add the difficulty that the method realizes, and also reduce the speed that method performs on the other hand to a great extent.
Summary of the invention
For the deficiencies in the prior art, the present invention proposes a kind of brain-machine interface method based on the fractal intercept feature of box peacekeeping.The method is using the fractal intercept feature of EEG signals box peacekeeping extracted as input parameter, sends in Boosting sorter and classifies, and then obtains brain electricity condition testing result.
The present invention also provides a kind of device performing the above-mentioned brain-machine interface method based on the fractal intercept feature of box peacekeeping.
Summary of the invention:
A kind of brain-machine interface method based on the fractal intercept feature of box peacekeeping is the detection that the hardware platform formed based on eeg amplifier and computing machine realizes to brain electricity condition; First EEG signals is gathered by eeg amplifier and data collecting card, then the EEG signals collected is delivered to computing machine to process, realize the feature extraction of the fractal intercept of box peacekeeping, and complete the classification to EEG signals by Boosting sorter, send control command.
Detailed Description Of The Invention:
Technical scheme of the present invention is as follows:
Based on a brain-machine interface method for the fractal intercept feature of box peacekeeping, comprise the following steps:
1) when collection experimenter imagines left little finger of toe, tongue movements, the EEG signals that brain produces, sample frequency is 1000Hz; The classification logotype that when experimenter imagines that left little finger of toe moves, its EEG signals is corresponding is 0 class, and the classification logotype that during imagination tongue movements, its EEG signals is corresponding is 1 class;
2) carry out down-sampled to the EEG signals collected, sample frequency is 100Hz;
3) to through step 2) down-sampled after EEG signals carry out the bandpass filtering of 8-30Hz;
4) extract the fractal intercept feature of box peacekeeping of each passage in EEG signals, to the step method of the described fractal intercept of extraction EEG signals each passage box peacekeeping be wherein:
A) by length after down-sampled and filtering be the EEG signals S of L point, average mark cuts G the subsegment of growing into H point, then calculates the fractal intercept of box peacekeeping of each subsegment;
B) to each subsegment of EEG signals S, continued to split T the segment (r=2 growing into r
h, (0 < h < log
2h), T equals the integral part of H/r), the length of side of getting the box covering signal equals r.To i-th segment (wherein i=1 ..., T), if the minimum value of its amplitude and maximal value drop in kth and l box respectively, then box number n (i) of covering needed for i-th segment is:
n(i)=l-k+1;
C) box sum Num (r) covered needed for this subsegment is:
D) the box counting dimension D of this subsegment EEG signals is:
E) when box the length of side r change time, steps d) described in formula meet straight-line equation:
logNum(r)=D·log(1r)+C
Wherein, the slope of straight line is D, and intercept is C; Get different r values, calculate some groups (r, Num (r)); Application least square curve fitting algorithm, tries to achieve slope D and the intercept C of this straight-line equation; Here, slope D is the box counting dimension of this subsegment EEG signals, and intercept C is then the fractal intercept of this subsegment EEG signals;
5) the fractal intercept feature of the box peacekeeping of step 4) being extracted is input to Boosting sorter and classifies, and obtains output probability value;
6) output probability value and predetermined threshold value are compared, wherein said predetermined threshold value is 0.5, obtains brain electricity condition testing result and is converted to control command:
When output probability value is greater than predetermined threshold value, then the EEG signals when brain electricity condition judging now is imagination tongue, and be converted to control command 1;
When output probability value is less than or equal to predetermined threshold value, then the EEG signals when brain electricity condition judging now is the imagination left little finger of toe, and be converted to control command 2;
The method of EEG signals being carried out to filtering described in step 3), comprises step as follows:
The Butterworth filter on J rank is utilized to carry out the bandpass filtering of 8-30Hz, preferred J=5 to EEG signals;
The performing step of the Boosting sorter described in step 5) is:
A) the training data feature set X={X used trained by sorter
j∈ R
k, j=1 ..., N}, the mark Y={y of its correspondence
j∈ 0,1}, j=1 ..., N}, wherein, K=Ch × s is the number of feature, and wherein Ch is port number, and s to be single experiment single lead on the number of proper vector that extracts, the number of single experiment of N for comprising in training data; F
mrepresent the sorter after m step; Setting iterations is M; The EEG signals feature vector, X of setting jth time single experiment
jfor thinking languet probability P
0(y
j=1|X
j)=0.5, j=1 ..., N, the EEG signals feature vector, X of setting jth time single experiment
jpreliminary classification device be F
0(X
j)=0, j=1 ..., N;
B) m represents iterative steps, from m=1, carry out following loop iteration:
I. sorter F is asked
mthe gradient of likelihood function
:
Wherein,
be after m-1 walks iteration, feature vector, X
jbelong to the probable value of imagination tongue brain electricity;
Ii., in least square meaning, the Weak Classifier f matched most with gradient is selected
m:
Wherein, regression coefficient vector w is tried to achieve by least-squares algorithm.
Iii. F is obtained according to given training data
mbernoulli Jacob's log-likelihood function:
Iv. Weak Classifier f is calculated
mweights γ
mfor:
γ
m=argmaxL(F
m-1+γf
m;X,Y);
V. sorter is upgraded:
F
m=F
m-1+εγ
mf
m;
Wherein, ε is a minimal value, is set to 0.05;
Vi. by sorter F
mcalculate feature vector, X
jbelong to the probable value of imagination tongue brain electricity:
Wherein, F
m(X
j) represent the rear corresponding training data X of m step
jsorter.
Vii. make m=m+1, repeat above-mentioned circulation, if m=M, then loop iteration terminates, the sorter F=F obtained
m;
Described in step 5) by the method for classifier calculated output probability value be: fractal for the box peacekeeping in step 4) intercept feature vector, X is sent into sorter F, utilizes formula:
Obtain EEG signals for thinking languet probability P;
A kind of device utilizing said method to carry out brain-computer interface, comprise the eeg amplifier, data collecting card and the computing machine that connect with circuit, the brain electro-detection module detecting brain electricity condition is provided with in described computing machine, be transferred in computing machine after utilizing eeg amplifier and data collecting card to gather EEG signals, utilize brain electro-detection module to EEG signals carry out filtering and box dimension, fractal intercept feature extraction, and extracted proper vector is sent in Boosting sorter, obtain output probability value; Output probability value is compared with predetermined threshold value, obtains brain electricity condition testing result and be converted into the control command controlling external unit.
Useful effect of the present invention is:
Utilize the fractal intercept of characteristic effect good box peacekeeping to collection and eeg data after pretreatment carries out feature extraction, the proper vector extracted is sent in Boosting sorter, thus obtains the mark of the EEG signals to the motion of the difference imagination; In brain-computer interface technical field, the present invention further increases eeg signal classification accuracy.
Accompanying drawing explanation
Fig. 1 is structured flowchart of the present invention;
Fig. 2 is FB(flow block) of the present invention;
Fig. 3 is EEG signals when imagining left little finger of toe after filtering;
Fig. 4 is EEG signals when imagining tongue after filtering;
Fig. 5 is the variation diagram of EEG signals box dimensional feature;
Fig. 6 is the variation diagram of the fractal intercept feature of EEG signals.
Embodiment
Below in conjunction with accompanying drawing and example, the present invention will be further described, and the present invention is not limited to this;
Embodiment 1,
As shown in figures 1 to 6;
The present invention gathers EEG signals by electrode, and EEG signals is amplified and data collecting card through eeg amplifier, and input computing machine realizes eeg signal classification, and produces control command control external unit;
Based on a brain-machine interface method for the fractal intercept feature of box peacekeeping, its process flow diagram as shown in Figure 2, comprises the following steps:
1) when collection experimenter imagines left little finger of toe, tongue movements, the EEG signals that brain produces, sample frequency is 1000Hz; The classification logotype that when experimenter imagines that left little finger of toe moves, its EEG signals is corresponding is 0 class, and the classification logotype that during imagination tongue movements, its EEG signals is corresponding is 1 class, and single experiment EEG signals duration is 3 seconds;
The original EEG signals gathered as shown in Figure 3; Choose front 278 experiments of experimenter as training sample, remaining 100 times experiments are as test sample book;
2) carry out down-sampled to the EEG signals collected, sample frequency is 100Hz;
3) to through step 2) down-sampled after EEG signals carry out the bandpass filtering of 8-30Hz; The described method of EEG signals being carried out to filtering, comprises step as follows:
The Butterworth filter on J rank is utilized to carry out bandpass filtering, preferred J=5 to EEG signals; EEG signals after filtering as shown in Figure 4;
4) extract the fractal intercept feature of box peacekeeping of each passage in EEG signals, to the step method of the described fractal intercept of extraction EEG signals each passage box peacekeeping be wherein:
A) by the EEG signals S for L=300 point long after down-sampled and filtering, average mark cuts G=3 the subsegment of growing into H=100 point, then calculates the fractal intercept of box peacekeeping of each subsegment;
B) to each subsegment of EEG signals S, continued to split T the segment (r=2 growing into r
h, (0 < h < log
2h), T equals the integral part of H/r), the length of side of getting the box covering signal equals r.To i-th segment (wherein i=1 ..., T), if the minimum value of its amplitude and maximal value drop in kth and l box respectively, then box number n (i) of covering needed for i-th segment is:
n(i)=l-k+1;
C) box sum Num (r) covered needed for this subsegment is
D) the box counting dimension D of this subsegment EEG signals is:
E) when box the length of side r change time, steps d) described in formula meet straight-line equation:
logNum(r)=D·log(1r)+C
Wherein, the slope of straight line is D, and intercept is C; Get different r values, calculate some groups (r, Num (r)); Application least square curve fitting algorithm, tries to achieve slope D and the intercept C of this straight-line equation; Here, slope D is the box counting dimension of this subsegment EEG signals, and intercept C is then the fractal intercept of this subsegment EEG signals; Fig. 5 is the box dimensional feature of EEG signals, and Fig. 6 is the fractal intercept feature of EEG signals;
5) the fractal intercept feature of the box peacekeeping of step 4) being extracted is input to Boosting sorter and classifies, and obtains output probability value;
The specific implementation step of the Boosting sorter described in step 5) is:
A) the training dataset X={X used trained by sorter
j∈ R
k, j=1 ..., N}, the mark Y={y of its correspondence
j∈ 0,1}, j=1 ... N}, wherein, K=Ch × s is the number of feature, wherein Ch is that port number equals 64, and s to be single experiment single lead on the number of proper vector that extracts equal 6, N and be the single experiment comprised in training data number equals 278; F
mrepresent the sorter after m step; Setting iterations is M=200; The feature vector, X of the EEG signals of setting jth time single thought experiment
jfor thinking languet probability P
0(y
j=1|X
j)=0.5, j=1 ..., N, the feature vector, X of the EEG signals of setting jth time single thought experiment
jpreliminary classification device be F
0(X
j)=0, j=1 ..., N;
B) m represents iterative steps, from m=1, carry out following loop iteration:
I. sorter F is asked
mthe gradient of likelihood function
:
Wherein,
be after m-1 walks iteration, feature vector, X
jbelong to the probable value of imagination tongue brain electricity;
Ii., in least square meaning, the Weak Classifier f matched most with gradient is selected
m:
Wherein, regression coefficient vector w is tried to achieve by least-squares algorithm.
Iii. F is obtained according to given training data
mbernoulli Jacob's log-likelihood function:
Iv. Weak Classifier f is calculated
mweights γ
mfor:
γ
m=argmaxL(F
m-1+γf
m;X,Y)
V. sorter is upgraded:
F
m=F
m-1+εγ
mf
m;
Wherein, ε is a minimal value, is set to 0.05;
Vi. by sorter F
mcalculate feature vector, X
jbelong to the probable value of imagination tongue brain electricity:
Wherein, F
m(X
j) represent the rear corresponding training data X of m step
jsorter.
Vii. make m=m+1, repeat above-mentioned circulation, if m=M, then loop iteration terminates, the sorter F=F obtained
m;
Described in step 5) by the method for classifier calculated output probability value be: fractal for the box peacekeeping in step 4) intercept feature vector, X is sent into sorter F, utilizes formula:
Obtain EEG signals for thinking languet probability P;
6) output probability value and predetermined threshold value are compared, wherein said predetermined threshold value is 0.5, obtains brain electricity condition testing result and is converted to control command:
When output probability value is greater than predetermined threshold value, then the EEG signals when brain electricity condition judging now is imagination tongue, and be converted to control command 1;
When output probability value is less than or equal to predetermined threshold value, then the EEG signals when brain electricity condition judging now is the imagination left little finger of toe, and be converted to control command 2.
Embodiment 2,
A kind of method as described in Example 1 that utilizes carries out the device of brain-computer interface, as shown in Figure 2, comprise the eeg amplifier, data collecting card and the computing machine that connect with circuit, the brain electro-detection module detecting brain electricity condition is provided with in described computing machine, be transferred in computing machine after utilizing eeg amplifier and data collecting card to gather EEG signals, utilize brain electro-detection module to EEG signals carry out filtering and box dimension, fractal intercept feature extraction, and extracted proper vector is sent in Boosting sorter, obtain output probability value; Output probability value is compared with predetermined threshold value, obtains brain electricity condition testing result and be converted into the control command controlling wheelchair.
Utilize the present invention to detect test brain electricity sample, the accuracy of identification reaches 92%.
Claims (6)
1., based on a brain-machine interface method for the fractal intercept feature of box peacekeeping, it is characterized in that, the method comprises the following steps:
1) when collection experimenter imagines left little finger of toe, tongue movements, the EEG signals that brain produces, sample frequency is 1000Hz; The classification logotype that when experimenter imagines that left little finger of toe moves, its EEG signals is corresponding is 0 class, and the classification logotype that during imagination tongue movements, its EEG signals is corresponding is 1 class;
2) carry out down-sampled to the EEG signals collected, sample frequency is 100Hz;
3) to through step 2) down-sampled after EEG signals carry out the bandpass filtering of 8-30Hz;
4) extract the fractal intercept feature of box peacekeeping of each passage in EEG signals, to the step method of the described fractal intercept of extraction EEG signals each passage box peacekeeping be wherein:
A) by length after down-sampled and filtering be the EEG signals S of L point, average mark cuts G the subsegment of growing into H point, then calculates the fractal intercept of box peacekeeping of each subsegment;
B) to each subsegment of EEG signals S, continued to split T the segment of growing into r, r=2
h, 0 < h < log
2h, T equal the integral part of H/r, and the length of side of getting the box covering signal equals r, to i-th segment, wherein i=1 ... T, if the minimum value of its amplitude and maximal value drop in kth and l box respectively, then box number n (i) of covering needed for i-th segment is:
n(i)=l-k+1;
C) box sum Num (r) covered needed for this subsegment is:
D) the box counting dimension D of this subsegment EEG signals is:
E) when box the length of side r change time, steps d) described in formula meet straight-line equation:
logNum(r)=D·log(1/r)+C
Wherein, the slope of straight line is D, and intercept is C; Get different r values, calculate some groups (r, Num (r)); Application least square curve fitting algorithm, tries to achieve slope D and the intercept C of this straight-line equation; Here, slope D is the box counting dimension of this subsegment EEG signals, and intercept C is then the fractal intercept of this subsegment EEG signals;
5) by step 4) the fractal intercept feature of box peacekeeping extracted is input to Boosting sorter and classifies, and obtains output probability value;
6) output probability value and predetermined threshold value are compared, wherein said predetermined threshold value is 0.5, obtains brain electricity condition testing result and is converted to control command:
When output probability value is greater than predetermined threshold value, then the EEG signals when brain electricity condition judging now is imagination tongue, and be converted to control command 1;
When output probability value is less than or equal to predetermined threshold value, then the EEG signals when brain electricity condition judging now is the imagination left little finger of toe, and be converted to control command 2.
2. a kind of brain-machine interface method based on the fractal intercept feature of box peacekeeping according to claim 1, is characterized in that, step 3) described in the method for EEG signals being carried out to filtering, comprise step as follows:
The Butterworth filter on J rank is utilized to carry out the bandpass filtering of 8-30Hz to EEG signals.
3. a kind of brain-machine interface method based on the fractal intercept feature of box peacekeeping according to claim 2, is characterized in that, described J=5.
4. according to claim 1, step 5) in the performing step of Boosting sorter be:
A) the training dataset X={X used trained by sorter
j∈ R
k, j=1 ..., N}, the mark Y={y of its correspondence
j∈ 0,1}, j=1 ... N}, wherein, K=Ch × s is the number of feature, wherein Ch is that port number equals 64, and s to be single experiment single lead on the number of proper vector that extracts equal 6, N and be the single experiment comprised in training data number equals 278; F
mrepresent the sorter after m step; Setting iterations is M=200; The feature vector, X of the EEG signals of setting jth time single thought experiment
jfor thinking languet probability P
0(y
j=1|X
j)=0.5, j=1 ..., N, the feature vector, X of the EEG signals of setting jth time single thought experiment
jpreliminary classification device be F
0(X
j)=0, j=1 ..., N;
B) m represents iterative steps, from m=1, carry out following loop iteration:
I. sorter F is asked
mthe gradient of likelihood function
Wherein,
be after m-1 walks iteration, feature vector, X
jbelong to the probable value of imagination tongue brain electricity;
Ii., in least square meaning, the Weak Classifier f matched most with gradient is selected
m:
Wherein, regression coefficient vector w is tried to achieve by least-squares algorithm;
Iii. F is obtained according to given training data
mbernoulli Jacob's log-likelihood function:
Iv. Weak Classifier f is calculated
mweights γ
mfor:
γ
m=argmaxL(F
m-1+γf
m;X,Y);
V. sorter is upgraded:
F
m=F
m-1+εγ
mf
m;
Wherein, ε is a minimal value, is set to 0.05;
Vi. by sorter F
mcalculate feature vector, X
jbelong to the probable value of imagination tongue brain electricity:
Wherein, F
m(X
j) represent the rear corresponding training data X of m step
jsorter;
Vii. make m=m+1, repeat above-mentioned circulation, if m=M, then loop iteration terminates, the sorter F=F obtained
m.
5. a kind of brain-machine interface method based on the fractal intercept feature of box peacekeeping according to claim 1, is characterized in that, step 5) described in by the method for classifier calculated output probability value be:
By step 4) in the fractal intercept feature vector, X of box peacekeeping send into sorter F, utilize formula:
Obtain EEG signals for thinking languet probability P.
6. one kind utilizes method as claimed in claim 1 to carry out the device of brain-computer interface, it is characterized in that, this device comprises the eeg amplifier, data collecting card and the computing machine that connect with circuit, the brain electro-detection module detecting brain electricity condition is provided with in described computing machine, be transferred in computing machine after utilizing eeg amplifier and data collecting card to gather EEG signals, utilize brain electro-detection module to EEG signals carry out filtering and box dimension, fractal intercept feature extraction, and extracted proper vector is sent in Boosting sorter, obtain output probability value; Output probability value is compared with predetermined threshold value, obtains brain electricity condition testing result and be converted into the control command controlling external unit.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310095778.0A CN103164026B (en) | 2013-03-22 | 2013-03-22 | Based on brain-machine interface method and the device of the fractal intercept feature of box peacekeeping |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310095778.0A CN103164026B (en) | 2013-03-22 | 2013-03-22 | Based on brain-machine interface method and the device of the fractal intercept feature of box peacekeeping |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103164026A CN103164026A (en) | 2013-06-19 |
CN103164026B true CN103164026B (en) | 2015-09-09 |
Family
ID=48587175
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201310095778.0A Expired - Fee Related CN103164026B (en) | 2013-03-22 | 2013-03-22 | Based on brain-machine interface method and the device of the fractal intercept feature of box peacekeeping |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103164026B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110174947A (en) * | 2019-05-27 | 2019-08-27 | 齐鲁工业大学 | The Mental imagery task recognition method to be cooperated based on fractals and probability |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101219048A (en) * | 2008-01-25 | 2008-07-16 | 北京工业大学 | Method for extracting brain electrical character of imagine movement of single side podosoma |
CN101923392A (en) * | 2010-09-02 | 2010-12-22 | 上海交通大学 | Asynchronous brain-computer interactive control method for EEG signal |
CN102708288A (en) * | 2012-04-28 | 2012-10-03 | 东北大学 | Brain-computer interface based doctor-patient interaction method |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100280403A1 (en) * | 2008-01-11 | 2010-11-04 | Oregon Health & Science University | Rapid serial presentation communication systems and methods |
EP2442714A1 (en) * | 2009-06-15 | 2012-04-25 | Brain Computer Interface LLC | A brain-computer interface test battery for the physiological assessment of nervous system health |
-
2013
- 2013-03-22 CN CN201310095778.0A patent/CN103164026B/en not_active Expired - Fee Related
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101219048A (en) * | 2008-01-25 | 2008-07-16 | 北京工业大学 | Method for extracting brain electrical character of imagine movement of single side podosoma |
CN101923392A (en) * | 2010-09-02 | 2010-12-22 | 上海交通大学 | Asynchronous brain-computer interactive control method for EEG signal |
CN102708288A (en) * | 2012-04-28 | 2012-10-03 | 东北大学 | Brain-computer interface based doctor-patient interaction method |
Non-Patent Citations (2)
Title |
---|
基于脑电的想象运动分类算法研究;高均波;《医学卫生科技辑》;20090115(第01期);全文 * |
脑电信号的分形截距特征分形及在癫痫检测中的应用;王玉、周卫东、李淑芳、袁琦、耿淑娟;《中国生物医学工程学报》;20110831;第30卷(第4期);全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN103164026A (en) | 2013-06-19 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102722727B (en) | Electroencephalogram feature extracting method based on brain function network adjacent matrix decomposition | |
CN101980106B (en) | Two-dimensional cursor control method and device for brain-computer interface | |
CN102521505B (en) | Brain electric and eye electric signal decision fusion method for identifying control intention | |
Gao et al. | A recurrence network-based convolutional neural network for fatigue driving detection from EEG | |
CN103793058A (en) | Method and device for classifying active brain-computer interaction system motor imagery tasks | |
CN101923392A (en) | Asynchronous brain-computer interactive control method for EEG signal | |
CN106527732B (en) | The selection of characteristic signal and optimization method in body-sensing electro photoluminescence brain-computer interface | |
CN104548347A (en) | Pure idea nerve muscle electrical stimulation control and nerve function evaluation system | |
CN103892829B (en) | A kind of eye based on common space pattern moves signal recognition system and recognition methods thereof | |
CN103092971B (en) | A kind of sorting technique for brain-computer interface | |
CN109171713A (en) | Upper extremity exercise based on multi-modal signal imagines mode identification method | |
CN107822629A (en) | The detection method of extremity surface myoelectricity axle | |
CN108268844A (en) | Movement recognition method and device based on surface electromyogram signal | |
CN110013248A (en) | Brain electricity tensor mode identification technology and brain-machine interaction rehabilitation system | |
CN107981997A (en) | A kind of method for controlling intelligent wheelchair and system based on human brain motion intention | |
CN105708587A (en) | Lower-limb exoskeleton training method and system triggered by brain-computer interface under motion imagination pattern | |
Li et al. | EEG signal classification method based on feature priority analysis and CNN | |
CN106109164A (en) | Rehabilitation system and the control method of rehabilitation system | |
CN104571504A (en) | Online brain-machine interface method based on imaginary movement | |
Wang et al. | Classification of EEG signal using convolutional neural networks | |
Donovan et al. | Simple space-domain features for low-resolution sEMG pattern recognition | |
CN108509869A (en) | Feature set based on OpenBCI optimizes on-line training method | |
Attenberger et al. | Modeling and visualization of classification-based control schemes for upper limb prostheses | |
Alansari et al. | Study of wavelet-based performance enhancement for motor imagery brain-computer interface | |
Geng et al. | A novel design of 4-class BCI using two binary classifiers and parallel mental tasks |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20150909 |