@article{ART003204632},
author={Won-Jun Han and Soo-Jin Lee and Gwang-Ho Kim and Ki-Pyong Park},
title={Lightweight Classification Model for Underwater Mines and Rocks Using Sonar Signals in Military Environment},
journal={Journal of The Korea Society of Computer and Information},
issn={1598-849X},
year={2025},
volume={30},
number={5},
pages={131-140}
TY - JOUR
AU - Won-Jun Han
AU - Soo-Jin Lee
AU - Gwang-Ho Kim
AU - Ki-Pyong Park
TI - Lightweight Classification Model for Underwater Mines and Rocks Using Sonar Signals in Military Environment
JO - Journal of The Korea Society of Computer and Information
PY - 2025
VL - 30
IS - 5
PB - The Korean Society Of Computer And Information
SP - 131
EP - 140
SN - 1598-849X
AB - The purpose of this study is to develop a lightweight machine learning model that can quickly and accurately classify underwater objects such as rocks and mines in a maritime military environment using SONAR data. To achieve this, we traind five machine learning models including KNN, SVM, Logistic Regression, LightGBM, and MLP, and analyzed and compared their classification performances. In order to improve the classification performance of the models, preprocessing techniques such as PCA and LDA were applied. Hyperparameters were also fine-tuned for each training model to derive optimized values. As a result of the experiment, the SVM model, which was trained by reducing the feature dimensions to 29 through PCA, achieved the highest accuracy of 96.15%, and all performances were superior to those of existing machine learning-based models. In particular, the proposed approach does not cause any misclassification of underwater mines at all, so it is expected to be useful for detecting submarines and underwater drones in a marine military environment in the future.
KW - Sonar;Underwater Mines;Machine Learning;SVM;Principal Component Analysis (PCA)
DO -
UR -
ER -
Won-Jun Han, Soo-Jin Lee, Gwang-Ho Kim and Ki-Pyong Park. (2025). Lightweight Classification Model for Underwater Mines and Rocks Using Sonar Signals in Military Environment. Journal of The Korea Society of Computer and Information, 30(5), 131-140.
Won-Jun Han, Soo-Jin Lee, Gwang-Ho Kim and Ki-Pyong Park. 2025, "Lightweight Classification Model for Underwater Mines and Rocks Using Sonar Signals in Military Environment", Journal of The Korea Society of Computer and Information, vol.30, no.5 pp.131-140.
Won-Jun Han, Soo-Jin Lee, Gwang-Ho Kim, Ki-Pyong Park "Lightweight Classification Model for Underwater Mines and Rocks Using Sonar Signals in Military Environment" Journal of The Korea Society of Computer and Information 30.5 pp.131-140 (2025) : 131.
Won-Jun Han, Soo-Jin Lee, Gwang-Ho Kim, Ki-Pyong Park. Lightweight Classification Model for Underwater Mines and Rocks Using Sonar Signals in Military Environment. 2025; 30(5), 131-140.
Won-Jun Han, Soo-Jin Lee, Gwang-Ho Kim and Ki-Pyong Park. "Lightweight Classification Model for Underwater Mines and Rocks Using Sonar Signals in Military Environment" Journal of The Korea Society of Computer and Information 30, no.5 (2025) : 131-140.
Won-Jun Han; Soo-Jin Lee; Gwang-Ho Kim; Ki-Pyong Park. Lightweight Classification Model for Underwater Mines and Rocks Using Sonar Signals in Military Environment. Journal of The Korea Society of Computer and Information, 30(5), 131-140.
Won-Jun Han; Soo-Jin Lee; Gwang-Ho Kim; Ki-Pyong Park. Lightweight Classification Model for Underwater Mines and Rocks Using Sonar Signals in Military Environment. Journal of The Korea Society of Computer and Information. 2025; 30(5) 131-140.
Won-Jun Han, Soo-Jin Lee, Gwang-Ho Kim, Ki-Pyong Park. Lightweight Classification Model for Underwater Mines and Rocks Using Sonar Signals in Military Environment. 2025; 30(5), 131-140.
Won-Jun Han, Soo-Jin Lee, Gwang-Ho Kim and Ki-Pyong Park. "Lightweight Classification Model for Underwater Mines and Rocks Using Sonar Signals in Military Environment" Journal of The Korea Society of Computer and Information 30, no.5 (2025) : 131-140.
