Table 2.
Summary of Machine Learning Approaches to Traffic Sign Recognition.
| Author | Algorithm | Dataset | Accuracy (%) |
|---|---|---|---|
| Kerim and Efe (2021) [1] | ANN | GTSRB | 95 |
| Soni et al., (2019) [2] | LBP, HOG, PCA, SVM | TSRD (Chinese) | 84.44 |
| Namyang and Phimoltares (2020) [3] | HOG, CLD, SVM, Random Forest | Self-collected (Thai) | 93.98 |
| Li et al., (2022) [4] | Color Histogram, HOG, PCA | GTSRB | 99.99 |
| Madani and Yusof (2018) [5] | Border Color, Shape, Pictogram, SVM | GTSRB | 98.23 |
| Sapijaszko et al., (2019) [6] | DWT, DCT, MLP | BTSD | 96.0 |
| GTSRB | 95.7 | ||
| TSRD | 94.9 | ||
| Aziz and Youssef (2018) [7] | HOG, CLBP, Gabor, ELM | GTSRB | 99.10 |
| BTSC | 98.30 | ||
| Weng and Chiu (2018) [8] | CCL, HOG, SVM | GTSRB | 90.85 |
| Wang (2022) [9] | LR | GTSRB | 97.75 |
| MLP | 98.88 | ||
| SVM | 95.51 |