Hybrid CNN-LSTM Machine Learning Algorithm for Driver Distraction Detection

Driver distraction is a critical safety concern on roadways, with potentially life-threatening consequences. The proposed system combines the spatial characteristic extraction capabilities of CNNs with the temporal modeling capabilities of LSTMs to effectively capture and analyze distractions in real-time. In the hybrid model, the CNN component processes visual input data, such as images or video frames, to extract spatial features that characterize driver behavior. These features are then fed into the LSTM component, which accounts for the temporal context and dependencies within the data. This approach promises to significantly enhance road safety by providing an intelligent system that can alert drivers and authorities to potential distractions, thereby reducing the risk of accidents. This study signifies a pivotal advancement in the progression of advanced driver assistance systems (ADAS) and driverless vehicles. The hybrid CNN-LSTM technique offers a promising route for safer and more reliable transportation systems with 93 % accuracy in detecting drivers' distractions.