Driver Drowsiness Detection

An Artificial Intelligence that doing Driver Drowsiness Detection

Our Research

Driver Drowsiness Detection

Driver Drowsiness Detection powered by Artificial Intelligence: Our state-of-the-art technology comprehends and interprets the subtle signs of driver drowsiness, facilitating uninterrupted vigilance and safety on the road. Pioneering new pathways for enhanced road safety through the capabilities of AI.

Image

DRIVER DROWSINESS DETECTION WITH DEPTHWISE SEPARABLE CONVOLUTIONAL LSTM

Depthiwse Separable Convolutional LSTM Diagram


Depthiwse Separable Convolutional LSTM Pictures

Model Architecture With Convolutional LSTM


Model Architecture Convolutional LSTM Pictures

Model Architecture With Depthwise Separable Convolutional LSTM


Model Architecture Depthiwse Separable Convolutional LSTM Pictures

README


Training is done using 2 architectures, the first using architecture with Convolutional LSTM, Second using architecture with Depthwise Separable Convolutional LSTM. Before training the dataset is extracted as many as 20 frames for each video, each frame is resized to 100 * 100 with 3 RGB Channels. Training uses epoch 10, learning rate 0.001, batch size 32, and Adam optimizer
Research thesis link : Open PDF

Prediction with 1 video

Sample data can be downloaded at : Data

Upload a Video to Predict

Select Method

Prediction Result :

Copyright © . All Rights Reserved. — Created by Trunojoyo AI Team 2023