Hand wash Hygiene Monitoring during COVID-19 using Deep Learning Approach

Hand washing has been considered one of the most important procedures to combat COVID-19 situation. In this research, the Visual Geometry Group16 (VGG16) machine learning model architecture has been applied in order to monitor hand hygiene procedure adapted by individual. The model developed has been tested on a real time dataset. The software used is the Python. For categorizing the inputs (image classification) the deep learning in python is used. To optimize the accuracy, the Adam optimizer algorithm has been applied. Similarly, to minimize the loss of the model, categorical cross-entropy loss estimation method was used. The model was trained, tested and validated where the research found that the developed model is an excellent-fit for predicting and classifying the hand-washing gestures-based images. The model achieved 97% accuracy rate, through classification accuracy as the performance metric evaluation technique. Keywords: Hand-washing gestures, hand washing detection model, hand washing detector, VGG16, convolutional neural network, Adam optimizer, classification accuracy.