Research Article
Classification of the Korean Sign Language Alphabet Using an Accelerometer with a Support Vector Machine
Table 2
Comparison studies on sign language alphabet recognition.
| | Author, year | Number of sensors | Sensor types (sampling frequency (Hz)) | Recognition targets | Classification model | Number of features | Classification accuracy (%) (random chance) |
| | Abualola et al., 2016 [15] | 6 | IMU (100) | American sign language alphabet (26) | LDAa | 5 | 85.00 (3.85) | | Yeo and Shin, 2018 [10] | 6 | sEMG (N/A), ACC (N/A), gyro (N/A) | Korean sign language alphabet (12) | Gaussian model | 6 | 92.40 (vowels) and 95.31 (consonants) (16.67) | | This study | 1 | ACC (20) | Korean sign language alphabet (31) | SVMb | 7 | 92.24 (3.23) |
|
|
aLinear discriminant analysis. bSupport vector machine.
|
