Detection of Pharmaceutical Personal Protective Equipment in Video Stream.

ABSTRACT

Pharmaceutical manufacturing is a complex process, where each stage requires a high level of safety and sterility. Personal Protective Equipment (PPE) is used for this purpose. Despite all the measures of control, human factor (improper PPE wearing) causes numerous losses for human health and material property. This research proposes solid computer vision system for ensuring safety in pharmaceutical laboratories. For this we have tested wide range of state-of-the-art object detection methods. Composing previously obtained results in this sphere with our own approach to this problem, we have reached the high accuracy (mAP@0.5) ranging from 0.77 up to 0.98 in detecting all the elements of common set of PPE used in pharmaceutical laboratories. Our system is the step towards safe medicine producing.