一般情況下,汽車駕駛人應使用雙手控制方向盤,否則將被視為不安全的駕駛行為。基於Aditya的論文,基於Aditya的論文,我們可以通過使用empirical mode decomposition (EMD)以及 Hilbert-Huang transform (HHT)從手錶的加速度計信號中萃取振動信息來檢測這些不安全的駕駛行為。 我們提出了一種新的方法來提高單手與雙手轉向操縱位置檢測,即為透過改用RBF-SVM和individual model和grouping model的建模方法。 我們的實驗結果顯示,通用模型中雙手檢測的準確率從75%提高到89%, individual model準確率為97%, grouping model準確率為93%。;Normally when driving the car, the driver should hold the steering wheel with both hands. Otherwise, the action is considered as an unsafe behavior. Based on Aditya’s thesis, potentially we are able to detect those unsafe behaviors by extracting vibration information from watch’s accelerometer signal using empirical mode decomposition (EMD) and Hilbert-Huang transform (HHT). We propose new approach to improve the accuracy value of one hand versus two hand steering handling position detection by changing the classifier using SVM with RBF kernel and modeling approach using individual model and grouping model. Our experiment result shows that the accuracy is increasing from 75% to 89% for two hand detection in universal model, 97% average in individual model, and 93% average in grouping model.