Sonar-Based Rock and Mine Detection for Enhancing Efficiency in Underwater Exploration and Security

SONAR (Sound Navigation and Ranging) is one of the critical techniques used for detection and location of underwater objects. Distinguishing harmless objects such as rocks from hazardous threats like mines is vital for successful naval operations and maritime security. This article discusses how the introduction of machine learning techniques will bring about a difference in this process. The results have shown that 80-90% accuracy could be achieved in the process using an appropriately designed machine learning model. The improved classification of underwater objects is, no doubt the most significant benefit of incorporating machine learning into the SONAR technology, which ensures the process both safer and efficient. Machine learning does not directly involve humans; hence, unlike traditional methods, it reduces risks to operators.Four different machine learning algorithms were used, namely Logistic Regression, K- Nearest Neighbors (KNN), Random Forest, and Support Vector Machine (SVM). The model performance metrics analysis was done to ensure a detailed comparison of which algorithm is best. Besides proving the supremacy of machine learning, this comparative study provides an accurate performance evaluation to ensure that the best model is selected for practical applications. The integration of multiple algorithms underscores the adaptability and robustness of machine learning, highlighting its potential to enhance SONAR technology for safer and more efficient maritime operations.