Sonar-based underwater sensing system

Abstract

A sonar-based underwater sensing method and systems are provided. The method includes performing intermittent scanning and image reconstruction; performing dynamic noise removal and object detection; and performing multi-dimensional activity recognition. To overcome the low frame rate due to the sonar's physical limitation, a scanning strategy is developed and an image reconstruction method is applied to accelerate the scanning speed without compromising the performance of motion detection. To overcome the dynamic interferences in the underwater scenario, a signal processing pipeline based on a physical model is developed to remove noises and localize human subjects. Features such as motion, time, and spatial information are further extracted from sonar images and a state-transfer-based activity recognition system is provided. The method was implemented across three public swimming pools for a total period of 94 hours. The evaluation results show that the method successfully recognizes around 90.5% human activities in the water.

Dependencies are inferred from the claim text.