http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Classification of directional cognition EEG using deep learning
심희동,박현준,안장원,현예나,양석조 대한기계학회 2021 대한기계학회 춘추학술대회 Vol.2021 No.4
Brain-Computer Interface(BCI) decodes human`s thoughts into machine commands using biological signals such as Electroencephalography(EEG). However, most high-performance BCI applications are based on reactive methods like SSVEP(Steady-State Visual Evoked Potential). These methods are not only counter-intuitive, but also requiring external stimulus. Since these problems are due to low-level of understanding on human cognitive processes, to solve these, biosignal-based understanding is essential. One of the most non-intuitively used in BCI system is directional information. Even though directional information is frequently used in BCI systems, it is still used in an indirect way. In this study, we designed cognitive tasks to elicit directional cognition from subjects while measuring EEG. To elicit directional cognition, subjects were stimulated in visual and auditory sounds such as directional arrows, words and sounds. These datasets were labeled with the stimulated directions. Deep learning based neural networks were trained to classify and compared the classification results with conventional methods. We believe that the results of this study will improve the understanding in directional cognition processes through electroencephalography.