Drowsy driver detection using Keras and convolution neural networks.
Datasets:
Eye dataset: http://parnec.nuaa.edu.cn/xtan/data/datasets/dataset_B_Eye_Images.rar
Yawn dataset: http://www.discover.uottawa.ca/images/files/external/YawDD_Dataset/YawDD.rar
Eye dataset credits,Yawn dataset credits
Files included:
EyePreprocess.py and YawnPreprocess.py : Preprocesses the data by converting the images to grayscale and dividing them into training and testing sets
EyesCNN.py and YawnCNN.py : Trains a CNN based on the training data.
EyeDetect.py and FaceDetect.py : Simple eyes and face detection code. Uses a 16-layer cascade instead of the traditional one since the original one was not able to detect faces properly.
Pickle files contain the preprocessed datasets for closed eyes, open eyes and yawns.