Search

Aiming at solving the problems of low recognition accuracy and poor practicability caused by the small amount of data and insufficient training of the network weights when the deep neural network algorithm is directly used to classify the 4-class motor-imagery electroencephalogram signals(MI-EEG), We combine the one-versus-the-rest common spatial pattern (OVR-CSP) algorithm and a novel convolution neural networks (CNN) algorithm to extract features and classify 4-class MI-EEG signals. Firstly, the original EEG signal data is truncated and expanded by using a fixed-size overlapping window, the features of intercepted sub-signals are extracted by CSP algorithm and the obtained feature vectors are merged as the input sample matrix of CNN. Secondly, CNN algorithm performs second feature extraction and final classification on the input sample matrix. Finally, the validity of the proposed algorithm was verified by the datasets IIIa of the BCI2005 competition. The average recognition accuracy of the three subjects has reached 91.9%, this is an improvement over other algorithms in recent years.

Electrical Capacitance Tomography (ECT) is a branch of Process Tomography based on capacitance-sensitive mechanism, which has the advantages of simple structure, low cost, high speed and relatively safe. To ensure the accuracy of image reconstruct, a large amount of research on sensitivity field is necessary. In this paper, we compared the differences of sensitivity fields between Electrical Capacitance Volume Tomography (ECVT) and Adaptive Electrical Capacitance Tomography (AECVT) system at first, studied the imaging characteristics of a regular symmetrical square sensor by a fast calculation method of three-dimensional AECVT sensitivity based on electric intensity, to simulate and analyze square tubes applied in some industries. Excited by different voltage envelopes synthetic plates adopted in AECVT system provides more variations and possibilities for imaging. The experiment explores the effects of different voltage packaging parameters on a series of characteristics such as the size and uniformity of the sensitivity field, in the hope that it will stimulate further research.

Adaptive electrical capacitance volume tomography (AECVT) is a powerful and flexible imaging technique that has received widespread attention in recent years. However, the accuracy of reconstructed image in AECVT system is still not ideal enough to meet the needs of industrial problems. Although differential evolution (DE) algorithm has played a great role in many other areas, very few studies have noted the role it plays in electrical capacitance volume tomography (ECVT) system, let alone AECVT system. Therefore, this paper contrasts the quality of the reconstruction image among the Particle Swarm Optimization algorithm and the DE algorithm with different strategies respectively applied in AECVT system when the adaptive plates are only excited by single voltage. Then we present a modified DE algorithm with a novel mutation strategy, which has a satisfied capability of avoiding convergence to the local optimum. By preliminary discussion, the modified DE algorithm is demonstrated to be slightly superior and stable to DE algorithm for single voltage excitation. In addition, in order to make full use of the flexibility of AECVT system, we conduct a series of experiments to find the best voltage envelope for the performance of proposed algorithm.

A key issue in brain-computer interface systems (BCI) based on motor-imagery electroencephalogram signals (MI-EEG) is the classification accuracy of EEG signals. Although deep learning (DL) methods have achieved great success in many research fields, only a limited number of works investigate its potential in BCI application research. In order to optimize the classification performance of MI-EEG signals, we propose a deep learning end-to-end classification model which is combined with convolutional neural network (CNN) and stacked autoencoders (SAE). A new type of CNN is introduced into the model for learning generalized features from time and spatial domains and for dimension reduction. Finally, the features extracted in the CNN are classified by a deep network SAE. The effectiveness of the proposed approach has been evaluated by using datasets of BCI competition data III and BCI competition data Ⅳ. Our results show that DL should be considered as an alternative to other state of art approaches, if the amount of data is large enough.

In order to realize the fast and accurate automated detection and classification of EEG signals during the normal, inter-ictal and ictal periods of patients, we propose an automated classification method for feature extraction of epileptic EEG signals based on the sample entropy and fast-slow-wave energy ratio (FSR)of each frequency sub-band in this paper. EEG signals are decomposed into frequency sub-bands using wavelet packet decomposition (WPD)in this method. The SampEn and FSR of different sub-bands are calculated, which are used to form feature vectors and these vectors are used as inputs to three different classifiers: support vector machine (SVM), k-nearest neighbor (KNN), and probabilistic neural network (PNN), to evaluate four famous classification problems. Our results show that the SVM classifier using radial basis function (RBF)is able to distinguish the above four problems with high accuracy more than 98.67%.

As an important research direction in the field of artificial intelligence, emotion recognition has become a hot topic in current research. Because of the use of multi-channel EEG acquisition equipment nowadays, which brings many problems in practical use and post-calculation, we have also studied channel selection. In this paper, the DEAP database is used as EEG data. The multi-feature fusion in a time domain and the composite features based on wavelet feature and information entropy are used as EEG features for emotion recognition. The average recognition accuracy reached 72.03% and 71.7% respectively. We also use the ReliefF algorithm to select EEG channels. Under the premise of a slight loss of emotional recognition preparation rate, we selected the optimal combination of 6 and 13 channels, and the brain data was reduced from 32 channels to 13 channels. It lays a foundation for the development of portable, wearable devices.

Electrical capacitance volume tomography(ECVT) is a simple-structure and low-cost imaging technique. ECVT is used to reconstruct an image of the permittivity distribution in the three-dimensional(3-D) domain with the limited capacitance measurements and sensitivity map by a suitable reconstruction algorithm. We can get the predicted image of higher resolution by improving the sensitivity map and the capacitance value and image reconstruction methods. In this paper, we contrast the quality of the reconstruction image between ECVT system and adaptive electrical capacitance volume tomography(AECVT) system. The experiment sensors used for comparison are the same size but different plates: conventional and synthetic which are both excited by single voltage. Then, we investigate the effect of different voltage envelopes on the quality of reconstruction image in AECVT. The examples are presented to demonstrate that AECVT system is slightly superior to ECVT system for single voltage excitation. Further more, for AECVT system, it can improve the 3-D reconstruction performance by using voltage excitations of multiple amplitudes and changing the voltage position.

Large-scale sensor-based Electrical Capacitance Tomography systems have a bright industrial application prospect. However, in terms of the hardware design of capacitance measuring circuits of the systems, a particular difficulty is to eliminate strong stray capacitance measuring interference produced by overlong single-shield coaxial cables and single analog switches. Therefore, in this paper different drive cable technologies are analyzed and complete drive cable technology is applied to the measuring circuits in order to eliminate coaxial cable stray capacitance interference. Furthermore, a double-T-switch connection is utilized for eliminating stray capacitance interference caused by using single switch arrangements. Also, a double electrode excitation strategy is implemented to suppress the noise generated in large-scale sensors and to enhance the Signal-to-Noise Ratio (SNR) of measuring circuits. Experimental results show that capacitance measuring circuits using the proposed anti-interference hardware have good stray capacitance suppression properties, a high SNR (57 dB) and good stability.