SSVEP EEG classification method based on convolutional neural model augmented with EMD data
A technology of convolutional neural and classification methods, applied in the fields of SSVEP EEG classification, artificial intelligence and pattern recognition, and brain-computer interface, to achieve the effect of optimizing models and inputs, high application prospects, and increasing ease of use
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Publication Date
- 2022-05-31
Smart Images

Figure 1 
Figure 2 
Figure 3
Abstract
Description
SSVEP EEG Classification Method Based on EMD Data Augmented Convolutional Neural Model technical field The present invention relates to the algorithm of brain-computer interface, artificial intelligence and pattern recognition, more particularly, relate to based on EMD data The SSVEP EEG classification method of the enhanced convolutional neural model can be applied to medical devices, human-computer interaction, robot control, etc. field. Background technique In recent years, with the development of computer science and artificial intelligence technology, brain-computer interface technology (BCI) as a new It has important application value in the field of rehabilitation science and control. BCI uses EEG signals to realize the human brain Communication or control with a computer or other electronic device, so it can help physically disabled patients to Improve the ability to communicate with the outside world. Steady-state visual evoked potentials are currently a...
Examples
Embodiment Construction
[0030] For the original EEG data, the preprocessing is mainly to filter out the DC component and band-pass filtering operations. first of all
[0036]
[0038] A large amount of artificial data is generated by randomly extracting and mixing different sequences of IMFs to train the network.
[0045] Referring to Figure 2, it is a structural diagram of a convolutional neural network model. The structure of the convolutional network is as follows, the first layer of convolution
[0048]
[0051] Results: As shown in Figure 5, the average correct rate of 5 subjects exceeded 95%, and the original training set was expanded to 2 times the original maximum.
[0053] Although the above-mentioned embodiments have been described, once those skilled in the art know the basic innovation