rishi-naath/Driver_Drowsiness_Detection_Using_CNN

A standalone project that can detect a driver's drowsiness based on their retinal actions and posture # Driver_Drowsiness_Detection_Using_CNN **Driver drowsiness** is a critical factor contributing to road accidents, as it severely impairs attention, reaction time, and decision-making. Factors like long driving hours, lack of sleep, and monotonous road conditions make drivers more susceptible to fatigue, often leading to serious or fatal incidents. Instructions: 1. clone this repository via cmd: cd your/directory/to/clone git clone github.com 3. after cloning, run: pip install -r requirements.txt 5. then open the dataset links and download the dataset via kaggle Huge thanks to : **Ismail Nasri**, Mohammed First University, Morocco. **Dhir Devansh**, TIET Patiala. 4. once done, specify the directory paths inside aiko (cnn_model(aiko)) and Driver_Drowsiness_Detection (DDD). 5. run the CNN model first so that it creates a fresh memory, and repeatedly train with both datasets (only one dataset can be trained at once, but memory stays) for better efficiency. 6. train the model with larger epochs, it'll stop in between from time to time, which is fine. 7. for better detection in DDD, focus over a moderately-lit area. for any doubts, contact: rishinaath33@gmail.com