Research Article
Controlling Embedded Systems Remotely via Internet-of-Things Based on Emotional Recognition
Table 3
HOG and SVM optimized parameters for training and testing with the SMO.
| Exp. No. | HOG-CS | HOG-bs | Image size | Feature length/image | Training time | Testing time | Accuracy | ConMat |
| 1 | 7 | 2 | 100 | 2808 | 97.74 | 4.0 | 87.80 | 800 | 119 | 125 | 956 | 2 | 9 | 2 | 100 | 1440 | 92.01 | 2.5 | 88.45 | 808 | 111 | 120 | 961 | 3 | 11 | 2 | 100 | 864 | 3.32 | 0.5 | 88.60 | 809 | 110 | 118 | 963 | 4 | 13 | 2 | 100 | 432 | 3.2 | 0.5 | 87.70 | 801 | 118 | 128 | 953 | 5 | 15 | 2 | 100 | 360 | 2.95 | 0.3 | 88.10 | 806 | 113 | 125 | 956 |
|
|