Research Article
Motor Imagery EEG Decoding Based on New Spatial-Frequency Feature and Hybrid Feature Selection Method
Table 7
The classification accuracy achieved by various methods (data set 2).
| Subject | DBN [32] (2018) | CapsNet [33] (2019) | SGRM [34] (2019) | NCFS [35] (2020) | CSP-FBLBP |
| B01 | 70.38 | 78.75 | 77.3 | 79.25 | 76.25 | B02 | 70.34 | 55.71 | 59.1 | 63.48 | 61.25 | B03 | 71.2 | 55 | 51.5 | 56.65 | 50 | B04 | 71.24 | 95.93 | 97 | 99.28 | 97.5 | B05 | 71.21 | 83.12 | 87.4 | 88.67 | 92.5 | B06 | 70.52 | 83.43 | 72.5 | 79.96 | 78.75 | B07 | 70.79 | 75.62 | 86.7 | 88.76 | 87.5 | B08 | 70.49 | 91.25 | 84.7 | 92.66 | 95 | B09 | 70.32 | 87.18 | 85.6 | 84.95 | 78.75 | Mean ± Std | 70.72 ± 0.4 | 78.44 ± 14.44 | 78.0 ± 14.65 | 81.52 ± 13.72 | 79.72 ± 15.86 |
|
|