Brain-computer interfaces (BCIs) based on Steady-State Visual Evoked Potentials (SSVEPs) has been attracting much attention because of its high information transfer rate and little user training. However, most methods...
详细信息
ISBN:
(数字)9781538683446
ISBN:
(纸本)9781538683453
Brain-computer interfaces (BCIs) based on Steady-State Visual Evoked Potentials (SSVEPs) has been attracting much attention because of its high information transfer rate and little user training. However, most methods applied to decode SSVEPs are limited to CCA and some extended CCA-based methods. This study proposed a comparing network based on Convolutional Neural Network (CNN), which was used to learn the relationship between EEG signals and the templates corresponding to each stimulus frequency of SSVEPs. This novel method incorporated prior knowledge and a spatial filter (task related component analysis, TRCA) to enhance detection of SSVEPs. The effectiveness of the proposed method was validated by comparing it with the standard CCA and other state-of-the art methods for decoding SSVEPs (i.e., CNN and TRCA) on the actual SSVEP datasets collected from 17 subjects. The comparison results indicated that the CNN-based comparing network significantly could significantly improve the classification accuracy compared with the standard CCA, TRCA and CNN. Furthermore, the comparing network with TRCA achieved the best performance among three methods based on comparing network with the averaged accuracy of 84.57% (data length: 2s) and 70.21% (data length: 1s). The study validated the efficiency of the proposed CNN-based comparing methods in decoding SSVEPs. It suggests that the comparing network with TRCA is a promising methodology for target identification of SSVEPs and could further improve the performance of SSVEP-based BCI system.
Numerical simulation of semiconductor devices enable the quick obtain of the device physical characteristics. However, the trial productions of the semiconductor devices have the disadvantages of long cycle and high c...
详细信息
ISBN:
(纸本)9781665409889
Numerical simulation of semiconductor devices enable the quick obtain of the device physical characteristics. However, the trial productions of the semiconductor devices have the disadvantages of long cycle and high cost. Therefore, it is necessary to carry out numerical simulation investigating semiconductor devices. We propose the coupling approach and the decoupling-Gummel method to solve the semiconductor equations. The results of the two methods are compared with the commercial software. Numerical experiment indicates that the results obtained by our method are accurate. When the transistor is at a low voltage bias, the decoupling method converges faster than the coupling method. In the case of high voltage bias, the decoupling method has a slower convergence speed than the coupling method, even difficult to converge for higher voltage bias. However, the coupling method has higher memory requirements than the Gummel method. Therefore, we need to balance the advantages of the coupling method and the Gummel method when conduct the numerical simulation.
A characteristic mode (CM) method that relies on a global multi-trace formulation (MTF) of surface integral equations is proposed to compute the modes and the resonance frequencies of microstrip patch antennas with fi...
详细信息
On 6 September 2017, the global scholars and professionals were mourning the death of professor L. A. Zadeh. This paper briefly recalls the historical significance of fuzzy mathematics, Zadeh’s friendship with China ...
On 6 September 2017, the global scholars and professionals were mourning the death of professor L. A. Zadeh. This paper briefly recalls the historical significance of fuzzy mathematics, Zadeh’s friendship with China and what his academic thoughts have been carried out in China.
In this paper, we proposed a frame-based approach to synthetic aperture radar (SAR) images fusion. Specifically, the method consists two parts: image fusion preprocessing and fusion reconstruction. Firstly, we fuse a ...
ISBN:
(数字)9781728129129
ISBN:
(纸本)9781728129136
In this paper, we proposed a frame-based approach to synthetic aperture radar (SAR) images fusion. Specifically, the method consists two parts: image fusion preprocessing and fusion reconstruction. Firstly, we fuse a set of low-resolution (LR) SAR images to form an aligned coarse SAR image which still has image blur caused by the point spread function (PSF) by fusion preprocessing. Then Frame fundamental iterative regularization is used to reconstruct high-resolution (HR) SAR image. Both simulated and real SAR images are used to verify the validity of the method. The results show that details and definition of the LR image processed with the method are effectively improved.
暂无评论