Self-Training Can Reduce Detection False Alarm Rate of High-Resolution Imaging Sonar

Imaging sonar is a primary means of underwater detection, but it faces challenges of high false alarm rates in sonar image target detection due to factors such as reverberation, noise, and resolution. This paper proposes a method to improve the false alarm rate by self-training a deep learning detec...

Full description

Saved in:
Bibliographic Details
Main Authors: Jingqi Han, Yue Fan, Zheng He, Zhenhang You, Peng Zhang, Zhengliang Hu
Format: Article
Language:English
Published: MDPI AG 2025-01-01
Series:Applied Sciences
Subjects:
Online Access:https://www.mdpi.com/2076-3417/15/3/1189
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850199450891321344
author Jingqi Han
Yue Fan
Zheng He
Zhenhang You
Peng Zhang
Zhengliang Hu
author_facet Jingqi Han
Yue Fan
Zheng He
Zhenhang You
Peng Zhang
Zhengliang Hu
author_sort Jingqi Han
collection DOAJ
description Imaging sonar is a primary means of underwater detection, but it faces challenges of high false alarm rates in sonar image target detection due to factors such as reverberation, noise, and resolution. This paper proposes a method to improve the false alarm rate by self-training a deep learning detector on sonar images. Self-training automatically generates proxy classification tasks based on the sonar image target detection dataset, and pre-trains the deep learning detector through these proxy classification tasks to enhance its learning effectiveness of target and background features. This, in turn, improves the detector’s ability to distinguish between targets and backgrounds, thereby reducing the false alarm rate. For the first time, this paper conducts target detection experiments based on deep learning using high-resolution synthetic aperture sonar images at two frequencies. The results show that, under the conditions of equal or higher recall rates, this method can reduce the false alarm rate by 3.91% and 18.50% on 240 kHz and 450 kHz sonar images, respectively, compared to traditional transfer learning methods.
format Article
id doaj-art-ea81daa853904d3495a9b86fd3866f8a
institution OA Journals
issn 2076-3417
language English
publishDate 2025-01-01
publisher MDPI AG
record_format Article
series Applied Sciences
spelling doaj-art-ea81daa853904d3495a9b86fd3866f8a2025-08-20T02:12:37ZengMDPI AGApplied Sciences2076-34172025-01-01153118910.3390/app15031189Self-Training Can Reduce Detection False Alarm Rate of High-Resolution Imaging SonarJingqi Han0Yue Fan1Zheng He2Zhenhang You3Peng Zhang4Zhengliang Hu5College of Meteorology and Oceanography, National University of Defense Technology, Changsha 410073, ChinaCollege of Electronic Engineering, Naval University of Engineering, Wuhan 430033, ChinaCollege of Meteorology and Oceanography, National University of Defense Technology, Changsha 410073, ChinaCollege of Meteorology and Oceanography, National University of Defense Technology, Changsha 410073, ChinaCollege of Meteorology and Oceanography, National University of Defense Technology, Changsha 410073, ChinaCollege of Meteorology and Oceanography, National University of Defense Technology, Changsha 410073, ChinaImaging sonar is a primary means of underwater detection, but it faces challenges of high false alarm rates in sonar image target detection due to factors such as reverberation, noise, and resolution. This paper proposes a method to improve the false alarm rate by self-training a deep learning detector on sonar images. Self-training automatically generates proxy classification tasks based on the sonar image target detection dataset, and pre-trains the deep learning detector through these proxy classification tasks to enhance its learning effectiveness of target and background features. This, in turn, improves the detector’s ability to distinguish between targets and backgrounds, thereby reducing the false alarm rate. For the first time, this paper conducts target detection experiments based on deep learning using high-resolution synthetic aperture sonar images at two frequencies. The results show that, under the conditions of equal or higher recall rates, this method can reduce the false alarm rate by 3.91% and 18.50% on 240 kHz and 450 kHz sonar images, respectively, compared to traditional transfer learning methods.https://www.mdpi.com/2076-3417/15/3/1189underwater target detectionsonar imagedeep learning
spellingShingle Jingqi Han
Yue Fan
Zheng He
Zhenhang You
Peng Zhang
Zhengliang Hu
Self-Training Can Reduce Detection False Alarm Rate of High-Resolution Imaging Sonar
Applied Sciences
underwater target detection
sonar image
deep learning
title Self-Training Can Reduce Detection False Alarm Rate of High-Resolution Imaging Sonar
title_full Self-Training Can Reduce Detection False Alarm Rate of High-Resolution Imaging Sonar
title_fullStr Self-Training Can Reduce Detection False Alarm Rate of High-Resolution Imaging Sonar
title_full_unstemmed Self-Training Can Reduce Detection False Alarm Rate of High-Resolution Imaging Sonar
title_short Self-Training Can Reduce Detection False Alarm Rate of High-Resolution Imaging Sonar
title_sort self training can reduce detection false alarm rate of high resolution imaging sonar
topic underwater target detection
sonar image
deep learning
url https://www.mdpi.com/2076-3417/15/3/1189
work_keys_str_mv AT jingqihan selftrainingcanreducedetectionfalsealarmrateofhighresolutionimagingsonar
AT yuefan selftrainingcanreducedetectionfalsealarmrateofhighresolutionimagingsonar
AT zhenghe selftrainingcanreducedetectionfalsealarmrateofhighresolutionimagingsonar
AT zhenhangyou selftrainingcanreducedetectionfalsealarmrateofhighresolutionimagingsonar
AT pengzhang selftrainingcanreducedetectionfalsealarmrateofhighresolutionimagingsonar
AT zhenglianghu selftrainingcanreducedetectionfalsealarmrateofhighresolutionimagingsonar