融合多域特征的VAE模型在肌肉疲劳分析中的应用

董博,吕东澔,喻大华,杜晓炜

VAE model combined with multi-domain feature for muscle fatigue analysis

Bo DONG,Donghao LV,Dahua YU,Xiaowei DU

表 3 不同特征肌肉疲劳分类的结果

Tab.3 Result of muscle fatigue classification with different feature %

分类模型 特征 R Spe P F1 A 支持向量机 iEMG 55.303 77.652 56.578 55.103 55.299 MF 62.879 81.439 66.062 63.298 62.877 FDE 77.273 88.636 79.410 77.655 77.350 TSMFDE 78.030 89.015 82.107 78.429 78.063 VAE 90.909 95.455 91.335 90.832 90.912 SSR-VAE 96.212 98.106 96.409 96.207 96.211 决策树 iEMG 53.030 76.515 54.157 53.280 53.020 MF 58.333 79.167 61.575 58.322 58.433 FDE 77.273 88.636 77.083 77.098 77.350 TSMFDE 83.333 91.667 84.376 83.357 83.419 VAE 87.121 93.561 87.319 87.117 87.123 SSR-VAE 95.455 97.727 95.613 95.478 95.442 随机森林 iEMG 50.758 75.379 50.060 50.146 50.798 MF 52.273 76.136 52.268 52.210 52.365 FDE 73.485 86.742 73.089 73.231 73.618 TSMFDE 79.545 89.773 79.589 79.561 79.601 VAE 87.121 93.561 87.190 87.087 87.151 SSR-VAE 93.939 96.970 94.074 93.981 94.960 K近邻 iEMG 56.818 78.409 56.554 56.188 56.866 MF 52.273 76.136 53.324 52.514 52.336 FDE 79.545 89.773 80.647 79.851 79.573 TSMFDE 81.061 90.530 82.777 81.211 81.054 VAE 90.152 95.076 90.597 90.119 90.171 SSR-VAE 94.967 97.348 94.829 94.730 94.701 朴素贝叶斯 iEMG 57.579 78.788 58.311 57.777 57.578 MF 61.364 80.682 63.051 63.713 61.368 FDE 80.303 90.152 82.121 80.709 80.370 TSMFDE 81.061 90.530 81.764 81.175 81.083 VAE 90.152 95.076 90.219 90.080 90.171 SSR-VAE 95.455 97.727 95.453 95.446 95.470