Dangerous Posture Monitoring for Undersea Diver Based on Frame Difference Method

Traditionally, the data continuity of monitoring method for undersea diver dangerous posture is poor, leading to the lack of edge information. Therefore, a frame difference method to monitor the dangerous posture of undersea diver was put forward. This method improved the multi-target tracking algorithm based on the combination of multi-feature by resetting the color feature and SIFT feature extraction technology, and thus to capture the underwater pose of diver with stronger continuity. Meanwhile, the frame difference method was used to set the morphological filter and identify the dangerous target of pose. Experimental results show that the proposed method has higher continuity of monitoring data and better edge image feedback effect of dangerous posture than the traditional monitoring method.