Deep Learning-Based Sonar Image Object Detection System

Md Shahazul Islam (Information and Communication Engineering, Nanjing University of Information Science and Technology, China)

International Journal of Informatics, Information System and Computer Engineering (INJIISCOM)
Vol. 6 No. 2 (2025): INJIISCOM: VOLUME 6, ISSUE 2, DECEMBER 2025

Sonar image object detection is crucial for underwater tasks, yet practical applications of advanced deep learning in this field remain underexplored due to environmental challenges. To bridge this gap, this study proposes a modular, deep learning-based sonar image object detection system. The system comprises three interdependent subsystems: dataset generation, algorithm model training and testing, and model deployment. Designed for high accuracy, speed, portability, and deployment adaptability, it effectively processes challenging sonar data. Experimental results from underwater suspicious object detection tasks confirm that the system achieves reliable, accurate performance and excellent real-world application outcomes. This work significantly advances sonar-based target localization and exploration.