Deep Learning-Based Cascaded Light Source Detection for Link Alignment in Underwater Wireless Optical Communication

Obtaining the light source position from the image is an important solution for achieving link alignment in laser-based underwater wireless optical communication (UWOC) systems. However, in practical scenarios, the misalignment degree between the light source and camera is variable, and factors such...

Full description

Saved in:
Bibliographic Details
Main Authors: Bowen Jia, Wenmin Ge, Jingxuan Cheng, Zihao Du, Renming Wang, Guangbin Song, Yufan Zhang, Chengye Cai, Sitong Qin, Jing Xu
Format: Article
Language:English
Published: IEEE 2024-01-01
Series:IEEE Photonics Journal
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10663260/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850107327586238464
author Bowen Jia
Wenmin Ge
Jingxuan Cheng
Zihao Du
Renming Wang
Guangbin Song
Yufan Zhang
Chengye Cai
Sitong Qin
Jing Xu
author_facet Bowen Jia
Wenmin Ge
Jingxuan Cheng
Zihao Du
Renming Wang
Guangbin Song
Yufan Zhang
Chengye Cai
Sitong Qin
Jing Xu
author_sort Bowen Jia
collection DOAJ
description Obtaining the light source position from the image is an important solution for achieving link alignment in laser-based underwater wireless optical communication (UWOC) systems. However, in practical scenarios, the misalignment degree between the light source and camera is variable, and factors such as ambient light may introduce disturbances, leading to significant variations in the appearance of light spots in images. Existing research primarily relies on simple features like brightness, color, or shape, which makes it difficult to accurately obtain position information from these non-ideal images. In this paper, deep neural networks (DNNs) with strong feature extraction capabilities are introduced to automatically learn the patterns of the light source from diverse images. A detection architecture cascading an object detector and a keypoint detector is adopted, achieving better comprehensive performance in terms of accuracy and speed. To train and evaluate the deep learning model, we construct the UWOC Light Source Detection Benchmark (ULDB) dataset. This dataset comprises 2200 images captured in a standard swimming pool, covering a misalignment range far beyond existing studies. On the ULDB test set, the proposed detection method achieves an average precision (AP) of 99.1% and an average positioning error of 4.66 pixels, while the traditional method may frequently extract false light spots. To the best of our knowledge, the ULDB dataset is the first image dataset specifically designed for the task of link alignment between UWOC terminals.
format Article
id doaj-art-b3dd0820b062418d8486291dec281140
institution OA Journals
issn 1943-0655
language English
publishDate 2024-01-01
publisher IEEE
record_format Article
series IEEE Photonics Journal
spelling doaj-art-b3dd0820b062418d8486291dec2811402025-08-20T02:38:36ZengIEEEIEEE Photonics Journal1943-06552024-01-0116511210.1109/JPHOT.2024.345311610663260Deep Learning-Based Cascaded Light Source Detection for Link Alignment in Underwater Wireless Optical CommunicationBowen Jia0https://orcid.org/0009-0003-0881-3727Wenmin Ge1Jingxuan Cheng2https://orcid.org/0009-0008-7177-064XZihao Du3Renming Wang4Guangbin Song5Yufan Zhang6https://orcid.org/0009-0004-4347-1272Chengye Cai7https://orcid.org/0000-0003-3672-728XSitong Qin8Jing Xu9https://orcid.org/0000-0001-6430-5063Hainan Institute of Zhejiang University, Sanya, ChinaOptical Communications Laboratory, Ocean College, Zhejiang University, Zhoushan, ChinaOptical Communications Laboratory, Ocean College, Zhejiang University, Zhoushan, ChinaOptical Communications Laboratory, Ocean College, Zhejiang University, Zhoushan, ChinaHainan Institute of Zhejiang University, Sanya, ChinaHainan Institute of Zhejiang University, Sanya, ChinaOptical Communications Laboratory, Ocean College, Zhejiang University, Zhoushan, ChinaOptical Communications Laboratory, Ocean College, Zhejiang University, Zhoushan, ChinaOptical Communications Laboratory, Ocean College, Zhejiang University, Zhoushan, ChinaHainan Institute of Zhejiang University, Sanya, ChinaObtaining the light source position from the image is an important solution for achieving link alignment in laser-based underwater wireless optical communication (UWOC) systems. However, in practical scenarios, the misalignment degree between the light source and camera is variable, and factors such as ambient light may introduce disturbances, leading to significant variations in the appearance of light spots in images. Existing research primarily relies on simple features like brightness, color, or shape, which makes it difficult to accurately obtain position information from these non-ideal images. In this paper, deep neural networks (DNNs) with strong feature extraction capabilities are introduced to automatically learn the patterns of the light source from diverse images. A detection architecture cascading an object detector and a keypoint detector is adopted, achieving better comprehensive performance in terms of accuracy and speed. To train and evaluate the deep learning model, we construct the UWOC Light Source Detection Benchmark (ULDB) dataset. This dataset comprises 2200 images captured in a standard swimming pool, covering a misalignment range far beyond existing studies. On the ULDB test set, the proposed detection method achieves an average precision (AP) of 99.1% and an average positioning error of 4.66 pixels, while the traditional method may frequently extract false light spots. To the best of our knowledge, the ULDB dataset is the first image dataset specifically designed for the task of link alignment between UWOC terminals.https://ieeexplore.ieee.org/document/10663260/Communication link alignmentdeep learningobject detectionkeypoint detectionunderwater wireless optical communication (UWOC)
spellingShingle Bowen Jia
Wenmin Ge
Jingxuan Cheng
Zihao Du
Renming Wang
Guangbin Song
Yufan Zhang
Chengye Cai
Sitong Qin
Jing Xu
Deep Learning-Based Cascaded Light Source Detection for Link Alignment in Underwater Wireless Optical Communication
IEEE Photonics Journal
Communication link alignment
deep learning
object detection
keypoint detection
underwater wireless optical communication (UWOC)
title Deep Learning-Based Cascaded Light Source Detection for Link Alignment in Underwater Wireless Optical Communication
title_full Deep Learning-Based Cascaded Light Source Detection for Link Alignment in Underwater Wireless Optical Communication
title_fullStr Deep Learning-Based Cascaded Light Source Detection for Link Alignment in Underwater Wireless Optical Communication
title_full_unstemmed Deep Learning-Based Cascaded Light Source Detection for Link Alignment in Underwater Wireless Optical Communication
title_short Deep Learning-Based Cascaded Light Source Detection for Link Alignment in Underwater Wireless Optical Communication
title_sort deep learning based cascaded light source detection for link alignment in underwater wireless optical communication
topic Communication link alignment
deep learning
object detection
keypoint detection
underwater wireless optical communication (UWOC)
url https://ieeexplore.ieee.org/document/10663260/
work_keys_str_mv AT bowenjia deeplearningbasedcascadedlightsourcedetectionforlinkalignmentinunderwaterwirelessopticalcommunication
AT wenminge deeplearningbasedcascadedlightsourcedetectionforlinkalignmentinunderwaterwirelessopticalcommunication
AT jingxuancheng deeplearningbasedcascadedlightsourcedetectionforlinkalignmentinunderwaterwirelessopticalcommunication
AT zihaodu deeplearningbasedcascadedlightsourcedetectionforlinkalignmentinunderwaterwirelessopticalcommunication
AT renmingwang deeplearningbasedcascadedlightsourcedetectionforlinkalignmentinunderwaterwirelessopticalcommunication
AT guangbinsong deeplearningbasedcascadedlightsourcedetectionforlinkalignmentinunderwaterwirelessopticalcommunication
AT yufanzhang deeplearningbasedcascadedlightsourcedetectionforlinkalignmentinunderwaterwirelessopticalcommunication
AT chengyecai deeplearningbasedcascadedlightsourcedetectionforlinkalignmentinunderwaterwirelessopticalcommunication
AT sitongqin deeplearningbasedcascadedlightsourcedetectionforlinkalignmentinunderwaterwirelessopticalcommunication
AT jingxu deeplearningbasedcascadedlightsourcedetectionforlinkalignmentinunderwaterwirelessopticalcommunication