CN114469134A - System and method of design thinking EEG using residual network to identify concept derivation stage - Google Patents
System and method of design thinking EEG using residual network to identify concept derivation stage Download PDFInfo
- Publication number
- CN114469134A CN114469134A CN202111547266.4A CN202111547266A CN114469134A CN 114469134 A CN114469134 A CN 114469134A CN 202111547266 A CN202111547266 A CN 202111547266A CN 114469134 A CN114469134 A CN 114469134A
- Authority
- CN
- China
- Prior art keywords
- data
- electroencephalogram
- residual error
- thinking
- brain
- 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
- 238000013461 design Methods 0.000 title claims abstract description 64
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000009795 derivation Methods 0.000 title claims abstract description 11
- 238000012549 training Methods 0.000 claims abstract description 22
- 238000007781 pre-processing Methods 0.000 claims abstract description 21
- 238000000605 extraction Methods 0.000 claims abstract description 14
- 238000012545 processing Methods 0.000 claims abstract description 14
- 238000004140 cleaning Methods 0.000 claims abstract description 3
- 230000009466 transformation Effects 0.000 claims abstract description 3
- 210000004556 brain Anatomy 0.000 claims description 24
- 239000011159 matrix material Substances 0.000 claims description 10
- 238000012880 independent component analysis Methods 0.000 claims description 8
- 238000003062 neural network model Methods 0.000 claims description 8
- 108010076504 Protein Sorting Signals Proteins 0.000 claims description 6
- 230000000694 effects Effects 0.000 claims description 6
- 238000011960 computer-aided design Methods 0.000 claims description 5
- 238000001914 filtration Methods 0.000 claims description 5
- 230000004913 activation Effects 0.000 claims description 4
- 238000000354 decomposition reaction Methods 0.000 claims description 3
- 238000011276 addition treatment Methods 0.000 claims description 2
- 238000004891 communication Methods 0.000 claims description 2
- 238000011176 pooling Methods 0.000 claims description 2
- 239000000284 extract Substances 0.000 abstract 1
- 238000004458 analytical method Methods 0.000 description 8
- 230000008569 process Effects 0.000 description 7
- 238000011160 research Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 230000005611 electricity Effects 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000002372 labelling Methods 0.000 description 2
- 238000003786 synthesis reaction Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000003340 mental effect Effects 0.000 description 1
- 230000035790 physiological processes and functions Effects 0.000 description 1
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/24—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
- A61B5/369—Electroencephalography [EEG]
- A61B5/372—Analysis of electroencephalograms
- A61B5/374—Detecting the frequency distribution of signals, e.g. detecting delta, theta, alpha, beta or gamma waves
-
- 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]
- A61B5/377—Electroencephalography [EEG] using evoked responses
-
- 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
- A61B5/7203—Signal processing specially adapted for physiological signals or for diagnostic purposes for noise prevention, reduction or removal
-
- 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
-
- 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
- A61B5/725—Details of waveform analysis using specific filters therefor, e.g. Kalman or adaptive filters
-
- 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
- A61B5/7267—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems involving training the classification device
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Surgery (AREA)
- Public Health (AREA)
- Pathology (AREA)
- Veterinary Medicine (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Molecular Biology (AREA)
- Psychiatry (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Artificial Intelligence (AREA)
- Signal Processing (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physiology (AREA)
- Psychology (AREA)
- Evolutionary Computation (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
The invention discloses a system and a method for recognizing thought electroencephalogram data designed in a concept derivation stage by using a residual error network. The head-wearing electroencephalogram equipment collects design thinking electroencephalogram data in real time; the data preprocessing module is used for cleaning and denoising original design thinking electroencephalogram data; the characteristic extraction module extracts characteristic data in the design thinking electroencephalogram data by wavelet packet transformation; the training module acquires electroencephalogram data under different design thinking, obtains characteristic data after processing, obtains labeled electroencephalogram data to form training data, and inputs the training data to obtain a trained model; and the classification display output module inputs the electroencephalogram data to be classified collected in real time into the trained model and outputs a classification result. The invention constructs the electroencephalogram characteristic data comprising three aspects of time, frequency and pole position, can quickly read, efficiently and accurately identify the classification of the design thinking electroencephalogram data, and has higher reading precision and speed.
Description
Technical Field
The invention relates to a system and a method for classifying and identifying design thinking electroencephalogram data, in particular to a system and a method for identifying design thinking electroencephalogram data at a concept derivation stage by applying a convolutional neural network.
Background
Design thinking is an innovative reasoning process, and the design problem faced by it is an undefined ambiguity.
When a person is faced with such a problem, the solution is made by the process of naming-framing-action-evaluation. Such complex mental activities have attracted the research of numerous scholars because of their potential to assist in the development of fields such as brain-like computing. Spoken language protocol analysis is one of the methods commonly used in the research field, and the method analyzes thinking changes in the thinking process by analyzing the thinking against the thinking process after the experiment is tested and encoding.
Although the spoken language protocol analysis opens the door of thinking analysis, due to thinking, especially complexity and convergence of design thinking, precise coding cannot be achieved by means of behavior labeling. Meanwhile, the spoken language protocol analysis needs long-time observation and labeling, and has huge cost in the aspects of manpower and material resources and limited accuracy.
Thus, physiological features such as brain electrical activity represent advantages. The research of the brain electricity in the design field has quite abundant achievements, and the automatic design thinking recognition based on the brain electricity has feasibility and is significant.
Therefore, in the prior art, only the processing mode is used for analyzing and identifying the physiological process of the design thinking, and other more effective means are lacked for solving the problem of analyzing and identifying the design thinking.
Disclosure of Invention
In order to solve the problems in the background art, the invention provides a system and a method for recognizing concept-derived stage design thinking electroencephalogram data by using a convolutional neural network.
The invention utilizes the physiological electroencephalogram characteristics to analyze and process to obtain the identification result of which stage the design thinking electroencephalogram data belongs to, can fully utilize the advantages of the physiological electroencephalogram characteristics to automatically design thinking identification, and also verifies that the method is an implementable scheme.
As shown in fig. 1, the technical scheme adopted by the invention is as follows:
the brain-computer-aided design method comprises the steps that brain-computer-aided design equipment is worn, brain-computer data of a human body under a design thinking is collected in real time to obtain the brain-computer-aided design thinking data, and the brain-computer-aided design data are sent to a data preprocessing module to be analyzed in real time;
the device comprises a data preprocessing module, a data processing module and a data processing module, wherein the data preprocessing module is used for receiving design thinking electroencephalogram data from head-wearing electroencephalogram equipment, and cleaning and denoising the original design thinking electroencephalogram data, and comprises operations of re-referencing, filtering, independent component analysis, eye charge removal and the like;
the device comprises a characteristic extraction module, a data preprocessing module and a data processing module, wherein the characteristic extraction module is used for receiving the processed design thinking electroencephalogram data from the data preprocessing module and extracting the characteristic data in the design thinking electroencephalogram data by using wavelet packet transformation so as to facilitate the subsequent input into a neural network for identification;
the brain-computer training system comprises a training module, a data preprocessing module and a feature extraction module, wherein the training module is connected to head-mounted brain-computer equipment, induces different design thoughts through presetting corresponding design task experimental settings, controls the head-mounted brain-computer equipment to acquire brain-computer data under the different design thoughts, obtains feature data after being processed by the data preprocessing module and the feature extraction module, and sets classification labels of the different design thoughts so as to collect the corresponding brain-computer data and print the labels, and obtains labeled brain-computer data to form training data; inputting training data into the neural network model for training to obtain a trained model and sending the trained model to the classification display output module;
the classification display output module receives the trained model from the training module, inputs the electroencephalogram data to be classified under the design thinking collected in real time into the trained model, and outputs the classification result to be displayed on a screen.
The design thinking comprises classification aspects such as thinking resistance, synthesis, analogy, situation building, problem definition and the like.
The details are as follows
The method comprises the steps of collecting design thinking electroencephalogram data of a human body, specifically, arranging electrode poles on the brain of the human body by adopting an electroencephalogram instrument when the human body carries out design thinking activities, and collecting signals of the electrode poles to form the design thinking electroencephalogram data of the human body.
12 electrode poles are arranged in the human brain, and the 12 electrode poles are respectively positioned at F3, T7, CP5, Pz, P7, O1, O2, P8, CP6, T8, F4 and Cz.
The invention is particularly provided with 12 electrodes, and can efficiently and accurately collect electroencephalogram data in the design thinking process.
As shown in fig. 3, in the data preprocessing module, for the signal of each brain electrode point in the design thinking electroencephalogram data, the signal of the pole is an analog signal, and the following processing is installed: the method comprises the steps of firstly re-referencing an original signal by taking a Tp9 pole and a Tp10 pole as references, then limiting the frequency band of the signal within 0-45Hz through band-pass filtering, namely reserving the signal within 0-45Hz, discarding the rest, and finally performing Independent Component Analysis (ICA) to remove the interference eye electrical signal by taking an Fp1 pole and an Fp2 pole as references.
The classification display system also comprises a display screen which is in communication connection with the classification display output module and displays the classification result sent from the classification display output module on the screen.
As shown in fig. 3, in the feature extraction module, for a signal of each brain electrode point in design thinking electroencephalogram data, feature data with the same length under five feature frequencies of δ (0.5-3.5Hz), θ (3.5-8Hz), α (8-13Hz), β (13-30Hz) and γ (31-45Hz) are extracted through wavelet packet variation decomposition, a continuous feature signal sequence is obtained by reconstructing according to a wavelet packet function for the feature data under each feature frequency, then the feature signal sequence is sliced once every 200ms, the overlapping rate between sliced segments is kept at 50%, all the sliced segments under the five feature frequencies form a two-dimensional matrix of 5 × m, wherein m represents the segment length after the feature data are sliced under a single feature frequency; the total n brain electrodes form a three-dimensional matrix with a single data format of 5 x m x n, and the three-dimensional matrix is standardized to be used as data input into the neural network model.
Thus, the data are processed to construct electroencephalogram characteristic data comprising three aspects of time, frequency and pole position.
The neural network model is mainly formed by sequentially connecting three continuous first residual error layers, four continuous second residual error layers, six continuous third residual error layers, three continuous fourth residual error layers, a discarding layer, an average pooling layer and a full-link layer, the topological structures of the first residual error layers, the second residual error layers, the third residual error layers and the fourth residual error layers are the same and respectively comprise two continuous convolution layers, an addition layer and an activation function, the input of the residual error layers is subjected to addition treatment by the two continuous convolution layers and the addition layer, and finally the input is output as the output of the residual error layers through the activation function; the output characteristic sizes of the first residual error layer, the second residual error layer, the third residual error layer and the fourth residual error layer are different and gradually increased.
The implementation also sets a number of different numbers of brain pole positions appropriate for the design thinking reading to test the uniqueness of the selection of 12 poles embodying the present invention.
The invention trains a network model with accurate recognition, and has higher reading precision and speed compared with the traditional spoken language protocol analysis method.
The invention has the beneficial effects that:
the method carries out real-time acquisition and processing on the data of the design thinking, constructs electroencephalogram characteristic data comprising three aspects of time, frequency and pole position, and can quickly read, efficiently and accurately identify the classification of the electroencephalogram data of the design thinking; compared with the traditional spoken language protocol analysis method, the method has higher reading precision and speed.
The invention solves the problems of high subjectivity and unobvious external thinking in the conventional common method in the design thinking research, and utilizes the design thinking with complex electroencephalogram physiological data analysis and research value to ensure that the thinking research is more accurate and objective. In addition, the high reading precision and speed of the invention solves the problems of high requirement of experts and high time cost of the existing common method, and further reduces the cost of thinking reading and analysis. Meanwhile, the invention provides a plurality of electroencephalogram electrode combinations capable of achieving similar recognition effects, and is beneficial to further industrial design and practice aiming at different analysis requirements and product requirements subsequently.
Drawings
FIG. 1 is a general logic block diagram of the method of the present invention.
Fig. 2 is an electrode arrangement position diagram of brain electricity.
FIG. 3 is a diagram of the subdivision logic relationship on the training side of the present invention.
FIG. 4 is a diagram of the subdivision logic relationship on the recognition side of the present invention.
Fig. 5 is a detailed topology diagram of the residual layer.
Detailed Description
The invention is described in further detail below with reference to the figures and the embodiments.
As shown in fig. 1, the embodiment of the present invention and its implementation are as follows:
1) collecting electroencephalogram data of human under design thinking through head-wearing electroencephalogram equipment
In the specific implementation, different pole numbers (6, 12, 18, 24 and 30 poles) are selected, and electroencephalogram data in the design thinking process are collected.
6. The selection of 12, 18, 24, 30 poles is as follows.
The positions of the respective electroencephalogram points are set according to medical placement standards as shown in fig. 2.
2) And data preprocessing, including operations of re-referencing, filtering, independent component analysis, eye removing and the like which are sequentially performed.
Aiming at the signal of each brain electrode point in the brain electrical data of design thinking, the signal of the pole is an analog signal, and the following modes are installed for processing: the method comprises the steps of firstly carrying out re-reference processing on an original signal by taking a Tp9 pole and a Tp10 pole as references, then limiting the frequency band of the signal within 0-45Hz through band-pass filtering, namely reserving the signal within 0-45Hz, discarding the rest, and finally carrying out independent component analysis ICA by taking an Fp1 pole and an Fp2 pole as references to remove an ocular signal interfered by the signal.
3) Feature extraction
Aiming at the signal of each brain electrode point in the design thinking electroencephalogram data,
extracting feature data with the same length under five feature frequencies of delta (0.5-3.5Hz), theta (3.5-8Hz), alpha (8-13Hz), beta (13-30Hz) and gamma (31-45Hz) through wavelet packet variation decomposition, reconstructing the feature data under each feature frequency according to a wavelet packet function to obtain a continuous feature signal sequence, then slicing the feature signal sequence every 200ms, keeping the overlapping rate between sliced segments at 50%, and combining all the sliced segments under the five feature frequencies into a two-dimensional matrix of 5 x m, wherein m represents the segment length of the feature data under a single feature frequency after slicing; the total n brain electrodes form a three-dimensional matrix with a single data format of 5 × m × n, and m is 200ms in specific implementation and is used as data input into the neural network model.
4) Training module
Through design task experiment setting, the corresponding design thinking (situation building, problem definition, analogy reasoning, synthesis and thinking resistance) is induced.
As shown in fig. 4, electroencephalogram data is collected and labeled, and preprocessing and feature extraction are sequentially performed, so that a 5 × m × 12 three-dimensional matrix composed of all 12 electroencephalogram polar points is obtained.
And then inputting the three-dimensional matrix into a neural network for training, performing model verification by using a 5-fold cross verification method, dividing the data into 5 parts, selecting one part of the 5 parts of data as test data in turn, using the other 4 parts of data as training data, repeating the training for 5 times, and obtaining the average value of the classification accuracy obtained by the training for 5 times by 30 rounds each time to obtain the final classification accuracy.
The average accuracy of the neural network model is shown in the table below.
TABLE 1 mean accuracy of classification for two classification modes proposed by the present invention
It can be seen from the above table that the method provided by the invention is applied to design the accuracy of thinking identification in the concept derivation stage, can achieve better identification effect, and has more than 95% identification accuracy for each pole combination.
Therefore, the implementation of the method can be seen that 12 poles are adopted, the classification result of the design thinking can be accurately identified and obtained, the identification can be completed more quickly and efficiently due to the fact that the number of the poles is small and the data is simplified, and the method has a better cost performance operation effect.
Under the training model 1, compared with the situation of 18 poles, 24 poles and 30 poles, the method of the invention adopts 12 poles, and has the advantages of high accuracy, easier data identification and high speed.
5) And finally, a classification display output module inputs the label-free electroencephalogram data collected in real time into the preprocessing module and the feature extraction module, then inputs the data into the trained model, and outputs a classification result to a screen.
Claims (7)
1. A design thinking electroencephalogram data identification system for concept derivation stages by using a residual error network is characterized in that:
the brain-computer-aided design system comprises a head-wearing brain-computer device, a data preprocessing module and a data processing module, wherein the head-wearing brain-computer device is used for acquiring brain-computer data of a human body under a design thinking in real time to obtain the brain-computer data of the design thinking and sending the brain-computer data to the data preprocessing module;
the device comprises a data preprocessing module, a data processing module and a data processing module, wherein the data preprocessing module is used for receiving design thinking electroencephalogram data from head-wearing electroencephalogram equipment and cleaning and denoising the original design thinking electroencephalogram data;
the device comprises a characteristic extraction module, a data preprocessing module and a data processing module, wherein the characteristic extraction module is used for receiving the processed design thinking electroencephalogram data from the data preprocessing module and extracting the characteristic data in the design thinking electroencephalogram data by wavelet packet transformation;
the device comprises a training module, a data preprocessing module and a feature extraction module, wherein the training module induces different design thoughts through design task experimental setting and controls head-wearing electroencephalogram equipment to acquire electroencephalogram data under the different design thoughts, the characteristic data is acquired after the characteristic data is processed by the data preprocessing module and the feature extraction module, and then classification labels of the different design thoughts are set to acquire labeled electroencephalogram data to form training data; inputting training data into the neural network model for training to obtain a trained model and sending the trained model to the classification display output module;
the classification display output module is used for inputting the electroencephalogram data to be classified collected in real time into a trained model and outputting a classification result.
2. The system for designing thinking brain electrical data recognition system of concept derivation stage by residual error network as claimed in claim 1, wherein: the method comprises the steps of collecting design thinking electroencephalogram data of a human body, specifically, arranging electrode poles on the brain of the human body by adopting an electroencephalogram instrument when the human body carries out design thinking activities, and collecting signals of the electrode poles to form the design thinking electroencephalogram data of the human body.
3. The system for designing thinking brain electrical data recognition system of concept derivation stage by residual error network as claimed in claim 2, wherein: 12 electrode poles are arranged in the human brain, and the 12 electrode poles are respectively positioned at F3, T7, CP5, Pz, P7, O1, O2, P8, CP6, T8, F4 and Cz.
4. The system for recognizing thinking electroencephalogram data designed for concept derivation stage by using residual error network as claimed in claim 1 or 2, wherein: in the data preprocessing module, aiming at the signal of each brain electrode point in the design thinking electroencephalogram data, the following modes are installed for processing: the method comprises the steps of firstly re-referencing an original signal by taking a Tp9 pole and a Tp10 pole as references, then limiting the frequency band of the signal within 0-45Hz through band-pass filtering, and finally carrying out Independent Component Analysis (ICA) to remove the interference eye electrical signal in the signal by taking an Fp1 pole and an Fp2 pole as references.
5. The system for designing thinking brain electrical data recognition system of concept derivation stage by residual error network as claimed in claim 1, wherein: the classification display system also comprises a display screen which is in communication connection with the classification display output module and displays the classification result sent from the classification display output module on the screen.
6. The system for designing thinking brain electrical data recognition system of concept derivation stage by residual error network as claimed in claim 1, wherein: in the characteristic extraction module, aiming at a signal of each brain electrode point in designed thinking electroencephalogram data, characteristic data with the same length under five characteristic frequencies of delta (0.5-3.5Hz), theta (3.5-8Hz), alpha (8-13Hz), beta (13-30Hz) and gamma (31-45Hz) are extracted through wavelet packet variation decomposition, a continuous characteristic signal sequence is obtained by adopting reconstruction according to a wavelet packet function for the characteristic data under each characteristic frequency, then the characteristic signal sequence is sliced once every 200ms, the overlapping rate of the sliced segments is kept at 50%, all the sliced segments under the five characteristic frequencies form a two-dimensional matrix of 5 x m, wherein m represents the segment length after the slicing of the characteristic data under a single characteristic frequency; the total of n brain electrode points form a three-dimensional matrix with a single data format of 5 x m x n, and the three-dimensional matrix is standardized to be used as data input into the neural network model.
7. The system for designing thinking brain electrical data recognition system of concept derivation stage by residual error network as claimed in claim 1, wherein: the neural network model is mainly formed by sequentially connecting three continuous first residual error layers, four continuous second residual error layers, six continuous third residual error layers, three continuous fourth residual error layers, a discarding layer, an average pooling layer and a full-link layer, the topological structures of the first residual error layers, the second residual error layers, the third residual error layers and the fourth residual error layers are the same and respectively comprise two continuous convolution layers, an addition layer and an activation function, the input of the residual error layers is subjected to addition treatment by the two continuous convolution layers and the addition layer, and finally the input is output as the output of the residual error layers through the activation function; the output characteristic sizes of the first residual error layer, the second residual error layer, the third residual error layer and the fourth residual error layer are different and gradually increased.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111547266.4A CN114469134B (en) | 2021-12-16 | 2021-12-16 | System and method for identifying design thinking EEG in concept derivation stage using residual network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111547266.4A CN114469134B (en) | 2021-12-16 | 2021-12-16 | System and method for identifying design thinking EEG in concept derivation stage using residual network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114469134A true CN114469134A (en) | 2022-05-13 |
| CN114469134B CN114469134B (en) | 2024-09-10 |
Family
ID=81493430
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111547266.4A Active CN114469134B (en) | 2021-12-16 | 2021-12-16 | System and method for identifying design thinking EEG in concept derivation stage using residual network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114469134B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120330869A1 (en) * | 2011-06-25 | 2012-12-27 | Jayson Theordore Durham | Mental Model Elicitation Device (MMED) Methods and Apparatus |
| CN104635934A (en) * | 2015-02-28 | 2015-05-20 | 东南大学 | Brain-machine interface method based on logic thinking and imaginal thinking |
| CN108563323A (en) * | 2018-02-05 | 2018-09-21 | 北京理工大学 | A kind of product design process based on EEG signals method by stages |
| WO2020186651A1 (en) * | 2019-03-15 | 2020-09-24 | 南京邮电大学 | Smart sports earphones based on eeg thoughts and implementation method and system thereof |
-
2021
- 2021-12-16 CN CN202111547266.4A patent/CN114469134B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120330869A1 (en) * | 2011-06-25 | 2012-12-27 | Jayson Theordore Durham | Mental Model Elicitation Device (MMED) Methods and Apparatus |
| CN104635934A (en) * | 2015-02-28 | 2015-05-20 | 东南大学 | Brain-machine interface method based on logic thinking and imaginal thinking |
| CN108563323A (en) * | 2018-02-05 | 2018-09-21 | 北京理工大学 | A kind of product design process based on EEG signals method by stages |
| WO2020186651A1 (en) * | 2019-03-15 | 2020-09-24 | 南京邮电大学 | Smart sports earphones based on eeg thoughts and implementation method and system thereof |
Non-Patent Citations (3)
| Title |
|---|
| JIA, WJ等: "EEG signals respond differently to idea generation, idea evolution and evaluation in a loosely controlled creativity experiment", SCIENTIFIC REPORTS, 8 March 2021 (2021-03-08), pages 1 - 20 * |
| 杨程;彭怡腾;唐智川;: "泛设计领域中的脑电研究现状与发展趋势", 包装工程, no. 16, 20 August 2020 (2020-08-20), pages 77 - 88 * |
| 胡人君;李坤;吴小培;: "脑机接口应用中的思维任务分类", 计算机工程与应用, no. 03, 21 January 2007 (2007-01-21), pages 205 - 207 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114469134B (en) | 2024-09-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11564612B2 (en) | Automatic recognition and classification method for electrocardiogram heartbeat based on artificial intelligence | |
| CN109657642A (en) | A kind of Mental imagery Method of EEG signals classification and system based on Riemann's distance | |
| CN110353702A (en) | A kind of emotion identification method and system based on shallow-layer convolutional neural networks | |
| CN112487945B (en) | Pulse condition identification method based on double-path convolution neural network fusion | |
| CN110367980A (en) | EEG signals Emotion identification method based on polynary empirical mode decomposition | |
| CN108937968B (en) | Lead selection method of emotion electroencephalogram signal based on independent component analysis | |
| CN104035563B (en) | W-PCA (wavelet transform-principal component analysis) and non-supervision GHSOM (growing hierarchical self-organizing map) based electrocardiographic signal identification method | |
| CN113180659B (en) | Electroencephalogram emotion recognition method based on three-dimensional feature and cavity full convolution network | |
| CN114469137B (en) | Cross-domain electroencephalogram emotion recognition method and system based on space-time feature fusion model | |
| CN106419911A (en) | Emotional detection method based on brain electric wave analysis | |
| CN109871831A (en) | Emotion recognition method and system | |
| CN108852348A (en) | The collection point sort method and system of scalp brain electricity | |
| CN117407748A (en) | EEG emotion recognition method based on graph convolution and fused attention | |
| CN106643722A (en) | Method for pet movement identification based on triaxial accelerometer | |
| CN106127191A (en) | Brain electricity sorting technique based on WAVELET PACKET DECOMPOSITION and logistic regression | |
| CN118986349A (en) | Electroencephalogram depression recognition system and method based on knowledge distillation and meta-shift learning | |
| CN113331845A (en) | Electroencephalogram signal feature extraction and accuracy discrimination method based on continuous coherence | |
| CN105718953A (en) | Single-time P300 detection method based on matrix grey modeling | |
| CN114469134A (en) | System and method of design thinking EEG using residual network to identify concept derivation stage | |
| Zhang et al. | A pruned deep learning approach for classification of motor imagery electroencephalography signals | |
| CN119184694A (en) | Four-dimensional attention recognition method based on electroencephalogram feature fusion selection | |
| CN114209340A (en) | A system and method for identifying concept derivation stage design thinking using convolutional neural networks | |
| Samarpita et al. | Differentiating mental stress levels: Analysing machine learning algorithms comparatively for EEG-based mental stress classification using MNE-Python | |
| Bhagwat et al. | Human disposition detection using EEG signals | |
| CN114587376A (en) | Multi-lead and multi-scale ECG detection method and system based on deep learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |