WiFi Based Vigilant Activity Recognition for Ship-Bridge Watchkeeping Officers

To guarantee the safety of marine traffic, ship watch-keeping officers must maintain a high level of vigilance during the watchkeeping period. In this paper, we utilized fine-grained channel state information (CSI) obtained using WiFi to track an officer-on-watch’s vigilant activities and determine in real time whether the ship officer was complying with safety guidelines without using cameras to violate the crew’s privacy. We adopted CSI velocity and activity models to determine bridge-officer activities by applying deep learning techniques. Finally, a watchkeeping vigilance evaluation module was developed and implemented on a commercial WiFi platform and extensively evaluated the system on an actual passenger ship. Our proposed system achieved accuracies of 95.4 and 93.8% for tracking walking movement and recognizing careless activities.