Tracking Hand Trajectory as a Preliminary Study for Hand Hygiene Stages

The process of hand washing involves complex hand movements. There are six principal sequential steps for washing hands as per the World Health Organisation (WHO) guidelines. In this work, a preliminary analysis was undertaken in order to develop an automated image processing system for tracking and classification of two-handed dynamic gestures involved in hand washing. To facilitate this study, videos of healthcare workers who were engaged in washing hands were sourced from the internet. The videos were analysed in order to extract the unique features of two-handed gestures associated with all hand hygiene (HH) stages. The combination of these unique features can be used to detect each HH stage. In the video recordings, the hand trajectory was found to be linear or circular for all six HH stages. In this paper, we attempt to track hand trajectory with the help of 2.5 Mega Pixel ELP-USB cameras and using an image processing approach for skin detection and the contour-centroid detection method. The YCbCr colour space is invariant to illumination intensity and therefore it was selected for the skin detection method. This work concludes that cameras are suitable for tracking one hand movement- linear and circular motion as a preliminary work and can be further expanded for detecting two hand movements in hand washing.