This work demonstrates a sonar prototype generating synthetic datasets for mine detection, suggesting improved data accessibility under varying environmental conditions.
Key Points
The sonar prototype successfully generated high fidelity acoustic datasets for underwater mine detection.
Under varying salinity from 0-35 ppt and temperature from 10-30°C, the system operated effectively.
Using signal processing techniques like low-pass filtering and echo detection, robust data was obtained.
The developed prototype aims to enhance machine learning model training through improved dataset quality.