EEG mental workload recognition using deep learning techniques
EEG devices are becoming more commonly available on the market and have seen an increase in usage in many different applications, such as human factor studies and human performance assessment. Deep learning techniques are also being applied to EEG in order to streamline the EEG data collection and p...
Saved in:
| 主要作者: | |
|---|---|
| 其他作者: | |
| 格式: | Final Year Project |
| 語言: | English |
| 出版: |
2019
|
| 主題: | |
| 在線閱讀: | http://hdl.handle.net/10356/77528 |
| 標簽: |
添加標簽
沒有標簽, 成為第一個標記此記錄!
|
| 機構: | Nanyang Technological University |
| 語言: | English |
| 總結: | EEG devices are becoming more commonly available on the market and have seen an increase in usage in many different applications, such as human factor studies and human performance assessment. Deep learning techniques are also being applied to EEG in order to streamline the EEG data collection and processing. This paper aims to review the available state-of art mental workload recognition algorithms from EEG and compare the effectiveness of subject-dependent algorithms and subject-independent algorithms, such as transfer learning and CNN. In this project, the results show that transfer learning and convolutional neural networks can be used to classify EEG data, but it requires a significant amount of improvement before it can be used on a regular basis. |
|---|