Research Article
Abnormal Gait Behavior Detection for Elderly Based on Enhanced Wigner-Ville Analysis and Cloud Incremental SVM Learning
Table 3
Comparison results of different methods for activities classification.
| | Experiment | BPNN [12] | Method [34] | Method [35] | Method [36] | Method [37] | Our method |
| | Walking | 95.0% | 92.2% | 90.0% | 99.9% | 94.2% | 100% | | Sitting | 91.6% | 90.2% | 96.7% | 90.6% | 91.7% | 98.6% | | Lying | 90.0% | 98.4% | 74.1% | 83.3% | — | 96.7% | | Falling | 86.6% | — | 95.5% | 87.5% | — | 95.9% | | Squatting | — | — | — | — | — | 98.3% | | Stumbling | — | — | — | — | — | 96.2% | | Syncope | — | — | — | — | — | 96.6% | | Average | 90.8% | 93.6% | 89.1% | 90.3% | 92.9% | 97.4% |
|
|