Motor Imagery EEG Feature Extraction Based on Fuzzy Entropy with Wavelet Transform
Due to the nonlinear characteristics of EEG signals and the rhythm characteristics of motor imagery, the low recognition rate of using single feature extraction algorithm, a feature extraction method based on wavelet transform and fuzzy entropy is present
- PDF / 488,822 Bytes
- 11 Pages / 439.37 x 666.142 pts Page_size
- 12 Downloads / 188 Views
Abstract. Due to the nonlinear characteristics of EEG signals and the rhythm characteristics of motor imagery, the low recognition rate of using single feature extraction algorithm, a feature extraction method based on wavelet transform and fuzzy entropy is presented in this paper. The EEG signals are decomposed to three levels by the wavelet transform, according to the ERS/ERD phenomena during motor imagery, the alpha rhythm and beta rhythm signal can be extracted by the algorithm of fuzzy entropy. Finally, the motor imagery EEG signals are classified by a support vector machine classifier. BCI Competition IV Datasets1 has been used to conduct the experiment, the experimental results show that the feature extraction method combining wavelet transform and fuzzy entropy is much better than the ways of using single fuzzy entropy, sample entropy, or others, and its highest recognition rate is 90.25%. Keywords: Electroencephalograph signal Fuzzy entropy transform Feature extraction Support vector machine
Wavelet
1 Introduction Brain-computer Interface (BCI) is a new technology that integrates multiple disciplines such as artificial intelligence, computer science and information, biomedical engineering and neuroscience [1]. This technology has broad applications prospects in the fields of medical treatment, rehabilitation of the disabled and cognitive science of the brain. BCI is a non-muscle communication system, and it allows the intention of brain connect with environment directly [2]. Feature extraction is a key problem in BCI system, and the result is directly related to the design of classifier and recognition rate. Feature extraction is done for making pattern recognition easier by extracting the most representative features from the EEG signals. The EEG signals are non-linear, so there are many nonlinear analysis methods applied in analyzing the EEG signals. With the development of the relevant theory, Approximate Entropy Characteristic (ApEN) [3], Sample Entropy (SampEn) [4, 5] and etc. have appeared one after another, and they are used as the measure of complexity. Acharya et al. [6] used ApEN, SampEn to extract the features of epilepsy electro-encephalogram and the normal EEG signals, characterizing the features of epileptic seizure in EEG signals efficiently. Chen et al. [7] improved SampEn and defined fuzzy entropy, and used fuzzy entropy to the feature © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 Q. Liu et al. (Eds.): CENet 2020, AISC 1274, pp. 1668–1678, 2021. https://doi.org/10.1007/978-981-15-8462-6_190
Motor Imagery EEG Feature Extraction Based on Fuzzy
1669
extraction of the EMG signals with success. Meanwhile, according to the phenomenon that the event related desynchronization and the event related synchronization (ERD/ERS) [8] occur with the motor imaginary of unilateral limb, the feature of alpha rhythm (8–15 Hz) and beta rhythm (15–30 Hz) is most obvious. Thus, it is necessary to resolve the EEG signals into different
Data Loading...
