underwater mapping_2022-2023

Underwater mapping 2022 - 2023

Authors: Chensheng Cheng, Can Wang, Dianyu Yang, Weidong Liu and Feihu Zhang. SLAM (Simultaneous Localization And Mapping) plays a vital role in navigation tasks of AUV (Autonomous Underwater Vehicle). However, due to a vast amount of image sonar data and some acoustic equipment’s inherent high latency, it is challenging to implement real- time underwater SLAM on a small AUV. This paper presents a filter based methodology for SLAM algorithms in underwater environments.

Authors: Emir Cem Gezer, Lin Zhao, Jordan Beason, Mingxi Zhou. Date of publication: 2022 Seafloor mapping and monitoring are essential for creating a sustainable economy and widening our understanding of the ocean environments. However, the cost of conducting underwater operations has always been high. This paper presents the progress on developing an affordable Autonomous Under-water Vehicle (AUV), equipped with an interferometric sonar for seabed mapping at low altitudes or in coastal water.

Authors: Jinkun Wang, Fanfei Chen, Yewei Huang, John McConnell, Tixiao Shan & Brendan Englot Date of publication: 2022 This paper presents a novel exploration framework for underwater robots operating in cluttered environments, built upon simultaneous localization and mapping (SLAM) with imaging sonar. The proposed system comprises path generation, place recognition forecasting, belief propagation, and utility evaluation using a virtual map, which estimates the uncertainty associated with map cells throughout a robot’s workspace.

Authors: Balint Z. Teglasy Emil Wengle, John R. Potter, and Sokratis Katsikas Secure digital wireless communication underwater is a crucial issue as the security of digital systems becomes challenged across all domains. The authors of this document address the requirements for security, namely, a confirmation of asset identities, also known as "authentication". They propose and validate an authentication protocol based on the first digital underwater communications standard primarily applicable to AUVs operating around offshore oil and gas facilities but also to other underwater devices.

Authors: Changho Yun To efficiently utilize nonexclusive underwater acoustic frequencies, The authors propose an Underwater Cooperative Spectrum Sharing (UCSS) protocol for a centralized underwater cognitive acoustic network consisting mainly of two parts. In the first part, to check the random occurrence of interferers periodically, the time domain is divided into frames that consist of a sensing and a non-sensing sub-frame. Then, they set the ratio of the two sub-frames to enhance the sensing rate via simulations.

Authors: Zheng Liu, Yaoming Zhuang, Pengrun Jia, Chengdong Wu, Hongli Xu, and Zhanlin Liu. The automatic detection and identification of marine organisms are critical for aquaculture resources. However, due to the low quality of underwater images and the characteristics of underwater biological detection, the lack of abundant features can impede traditional hand- designed feature extraction approaches or CNN-based object detection algorithms, particularly in complex aquatic environments. Therefore, this study aimed to perform object detection in underwater environments.

Author: Serkan Aksoy This document explains notions such as sonar equation, acoustic waves, underwater acoustic equations, reverberation, Helmholtz equation, noise sources, cavitation, and target strength.

Author: Aslanbek Naziev The authors study the (in)dependence of additivity and homogeneity conditions in the definition of a linear mapping between vector spaces over the same scalar field. Unlike other works on this theme, which deal with particular fields like real or complex numbers, we consider the general case. That enables them to obtain an almost complete picture.

Authors: Tuochao Chen, Justin Chan, Shyamnath Gollakota The authors present a novel underwater system that can perform acoustic ranging between commodity smartphones. To achieve this, they design a real-time underwater ranging protocol that computes the time-of-flight between smartphones. To address the severe underwater multipath, they present a dual-microphone optimization algorithm that can more reliably identify the direct path. Their evaluations show that our system has median errors of 0.48–0.86 m at distances up to 35 m.

02 - A Comprehensive Comparison and Analysis of Several Intensity Modulations Based on the Underwater Photon- Counting Communication System

Authors: Xiaotian Han, Peng Li, Chang Chang, Duorui Gao, Dongquan Zhang, Peixuan Liao, Wei Wang, and Xiaoping Xie This document proposes solutions to the problems of underwater wireless optical communication, which is facing absorption and scattering problems.

Authors: Zhimin Li, Zibin Lin, Longsheng Zeng, Hao Wu, and Xue- Feng Zhu Manipulating underwater acoustic waves along the prescribed trajectory has excellent potential for various applications. Traditional metasurfaces for underwater acoustic modulation usually have complex structural designs and are complicated to manufacture. Here, we propose a simple strategy of embedding air bubbles of different sizes inside the polymer to manipulate the transmitted underwater acoustic wave fields freely.

13 - Marine priority habitat mapping in a Mediterranean conservation area (Gyaros, South Aegean) through multi-platform marine remote sensing techniques.

Authors: Xenophon Dimas, Elias Fakiris, Dimitris Christodoulou, Nikos Georgiou, Maria Geraga, Vasillis Papathanasiou, Sotiris Orfanidis, Spyros Kotomatas, and George Papatheodorou The aim of this study was to present the results of the first complete marine habitat mapping through marine remote sensing techniques in Gyaros Island, a remote island in the Cyclades archipelago with a great historical and ecological value

Author: Usman Ali, and Muhammad Tariq Mahmood The authors of this paper propose a method for underwater image restoration utilizing a technique that incorporates the potential structural differences between transmission maps and the guidance map.

Authors: Xu Liu, Sen Lin, Zhiyong Tao Evidence suggests that vision is among the most critical factors in marine information exploration. Instead, underwater images are generally of poor quality due to color casts, lack of texture details, and blurred edges. The authors propose the Multiscale Gated Fusion conditional GAN (MGF-cGAN) for underwater image enhancement. The generator of MGF-cGAN consists of Multiscale Feature Extract Module (Ms-FEM) and Gated Fusion Module (GFM). In Ms-FEM, they use three parallel subnets to extract feature information, which can extract richer features than a single branch.

07 - Underwater Hyperspectral Imaging of Arctic Macroalgal Habitats during the Polar Night Using a Novel Mini-ROV-UHI Portable System.

Authors: Natalie Summers, Geir Johnsen, Aksel Mogstad, Håvard Løvås, Glaucia Fragoso, Jørgen Berge This document describes an Underwater Hyperspectral Imager (UHI) deployed on an instrument-carrying platform consisting of two interconnected mini-ROVs (Remotely Operated Vehicle) for the mapping and monitoring of Arctic macroalgal habitats in Kongsfjorden (Svalbard) during the Polar Night.

Author: Hong-Gi Kim, Jungmin Seo, and Soo Mee Kim Unmanned underwater operations using remotely operated vehicles or unmanned surface vehicles are increasing recently, guaranteeing human safety and work efficiency. Optical cameras and multi-beam sonars are generally utilized as imaging sensors in underwater environments. However, the obtained underwater images are difficult to understand intuitively, owing to noise and distortion. This study explains the development of an optical and sonar image fusion system that integrates the color and distance information from two images.

Authors: Zhenjing Zhu, Ning Hu, Junyi Wu, Wenxin Li, Jiabao Zhao, Maofa Wang, Fanzong Zeng, Huajie Dai, and Yongju Zheng. This document presents a new method for estimating the angular pulse spectrum. The process is based on the Husimi transform of a wave ﬁeld and can be realized with a short vertical array of nondirectional hydrophones. As a result, one obtains a diagram of the arrival pattern in the time–angle plane. Special attention is paid to sound scattering on a cold synoptic eddy along the waveguide.

28 - An End-to-End Underwater-Image-Enhancement Framework Based on Fractional Integral Retinex and Unsupervised Autoencoder.

Authors: Yang Yu, and Chenfeng Qin Underwater image enhancement faces challenges of light scattering, absorption, and distortion. Instead of using a specific underwater imaging model to mitigate the degradation of underwater images, the authors propose an end-to-end underwater-image-enhancement framework that combines fractional integral-based Retinex and an encoder–decoder network. The proposed variant of Retinex aims to alleviate haze and color distortion in the input image while preserving edges to a large extent by utilizing a modified fractional integral filter.

Authors: Denis V. Makarov, and Leonid E. Konkov This document presents a new method for estimating the angular pulse spectrum. The process is based on the Husimi transform of a wave field and can be realized with a short vertical array of nondirectional hydrophones. As a result, one obtains a diagram of the arrival pattern in the time–angle plane. Special attention is paid to sound scattering on a cold synoptic eddy along the waveguide.

Authors: Zefeng Zhao, Zhuang Zhou, Yunting Lai, Tenghui Wang, Shujie Zou, Haohao Cai, and Haijun Xie Clear underwater images are necessary for many underwater applications, while absorption, scattering, and different water conditions will lead to blurring and different color deviations. This paper proposed a fusion- based image enhancement method for various water areas to overcome the limitations of the available color correction and deblurring algorithms. The authors proposed two novel image processing methods, namely, an adaptive channel deblurring method and a color correction method, by limiting the histogram mapping interval.

Authors: Shlomi Amitai, Itzik Klein, and Tali Treibitz Depth estimation is critical for any robotic system. In the past years, the depth estimation from monocular images has shown significant improvement; however, in the underwater environment, results are still lagging due to appearance changes caused by the medium. The authors suggest training using subsequent frames self-supervised by a reprojection loss, as demonstrated above water. The authors suggest several additions to the self-supervised framework to cope with the underwater environment and achieve state-of-the-art results on a challenging forward-looking underwater dataset.

Authors: Mochou Yang, Yi Wu, and Guoying Feng In an underwater environment, conventional imaging is limited by low sensitivity, resulting in fuzzy images, while ghost imaging can solve this problem. This study proposes underwater laser ghost imaging based on Walsh speckle patterns. According to the simulated and experimental results, noise resistance and a low sampling rate of ghost imaging based on Walsh speckle patterns are proved. As the underwater environment's turbidity increases, the image quality of ghost imaging based on Walsh speckle patterns decreases.

Authors: Zheyong Li, Jinghua Li, Pei Zhang, Lihui Zheng, Yilong Shen, Qi Li, Xin Li, and Tong Li The detection of underwater targets through hyperspectral imagery is a relatively novel topic as the assumption of target background independence is no longer valid, making it difficult to detect underwater targets using land target information directly. Meanwhile, deep-learning-based methods have faced challenges regarding the availability of training datasets, especially in underwater conditions. To solve these problems, a transfer-based framework is proposed in this paper, which exploits synthetic data to train deep-learning models and transfers them to real-world applications.

Authors: Yang Guan, Xiaoyan Liu, Zhibin Yu, Yubo Wang, Xingyu Zheng, Shaoda Zhang, and Bing Zheng Underwater image enhancement is a fundamental requirement in the field of underwater vision. Along with the development of deep learning, underwater image enhancement has made remarkable progress. However, most deep learning-based enhancement methods are computationally expensive, restricting their application in real-time large-size underwater image processing. The authors propose a model based on a generative adversarial framework for large-size underwater image enhancement to solve these problems.

Authors: Yelena Randall, and Tali Treibitz Yan The authors collected forward-looking stereo-vision and visual-inertial image sets with two underwater imaging platforms, a stereo camera rig, and an ROV in the Mediterranean and Red Sea. To their knowledge, there is only one other public dataset in the underwater environment with this camera-sensor orientation. These datasets are critical for developing several underwater applications, including autonomous obstacle avoidance, visual odometry, 3D tracking, Simultaneous Localization and Mapping (SLAM), and depth estimation through deep learning.

Authors: Yuanheng Li, Shengxiong Yang, Yuehua Gong, Jingya Cao, Guang Hu, Yutian Deng, Dongmei Tian, and Junming Zhou Unpaired Image-to-Image Translation using Cycle- Consistent Adversarial Networks, also known as CycleGAN, has been broadly studied regarding underwater image enhancement, but it is difficult to apply the model because it can be difficult to train if the dataset used is not appropriate. In this article, the authors devise a new method of building a dataset and choosing the best images based on the widely used Underwater Image Quality Measure (UIQM) scheme to create the dataset.

Authors: Shijie Xu, Rendong Feng, Pan Xu, Zhengliang Hu, Haocai Huang, and Guangming Li Underwater environment observation with underwater acoustic tomography has been considerably developed in recent years. Moving sound transmission can obtain the observation of entire spatial area with sound station moving. Various internal structures, unique surface and submarine boundaries and changing environment constitutes a complex acoustic propagation channel. This paper focus on the inversion method and signal resampling for sound moving transmission. Also, the current field in three-dimensional (3D) scale is also studied.

Authors: Marios Xanthidis, Bharat Joshi, Monika Roznere, Weihan Wang, Nathaniel Burgdorfer, Alberto Quattrini Li, Philippos Mordohai, Srihari Nelakuditi, and Ioannis Rekleitis The authors of this paper discuss how to effectively map an underwater structure with a team of robots considering the specific challenges posed by the underwater environment.

Authors: Shiying Feng, Xiaofeng Li, Lu Ren, Shuiqing Xu Autonomous navigation of unmanned aerial vehicles (UAVs) is widely used in building rescue systems. As the complexity of the task increases, traditional methods based on environment models are hard to apply. this paper proposes a reinforcement learning (RL) algorithm to solve the UAV navigation problem.

Authors: Daniel Short, Tingjun Lei, Chaomin Luo, Daniel W. Carruth, Zhuming Bi This study aims to address a gap in existing research by integrating a biologically inspired Bat Algorithm (BA) for path planning. It outlines the motivation behind the study, the methodology involving the use of convolutional neural networks (CNNs) for feature extraction, and the implementation of the BA for global path planning. It also highlights the structure of the paper and the validation of the proposed approach through simulations and comparisons.

Author: Yian Wang An underwater target ranging and size measurement system based on binocular vision is proposed to measure the distance and size information of underwater targets accurately. The underwater binocular calibration method is used to obtain the underwater parameters of the binocular camera, and the underwater images are calibrated to make them coplanar and aligned.

Authors: Wei Song, Yaling Liu, Dongmei Huang, Bing Zhang,Zhihao Shen, & Huifang Xu This paper aims to systematically review underwater image restoration technology, bridging the gap between shallow-sea and deep-sea image restoration fields through experimental analysis. The authors first categorize shallow- sea image restoration methods into three types: physical model-based methods, prior-based methods, and deep learning-based methods that integrate physical models. The core concepts and characteristics of representative methods are analyzed. The research status and primary challenges in deep-sea image restoration are then summarized.

Authors: Xiaoyang Bai, Zuodong Liang, Zhongmin Zhu, Alexander Schwing, David Forsyth, & Viktor Gruev Monitoring water properties using autonomous underwater sampling robots remains a significant challenge due to a lack of underwater geolocalization capabilities. The authors present a new method for underwater geolocalization using deep neural networks trained on approximately 10 million polarization-sensitive images acquired globally, along with camera position sensor data.

Authors: Tuochao Chen, Justin Chan, & Shyamnath Gollakota The emergence of waterproof mobile and wearable devices designed for underwater activities opens up opportunities for underwater networking and localization capabilities on these devices. The authors present the first underwater acoustic positioning system for smart devices. Unlike conventional systems that use floating buoys as anchors at known locations, they design a system where a dive leader can compute the relative positions of all other divers, without any external infrastructure.

Authors: Boris Gasparovic, Jonatan Lerga, Goran Mausa, and Marina Ivasic-Kos This paper addresses the challenges posed by underwater conditions for image recognition and demonstrates the use of various Convolutional Neural Network (CNN) architectures, particularly focusing on the YOLO and Faster RCNN models. It evaluates these models based on detection accuracy, mean average precision, and processing speed, highlighting the effectiveness of the YOLOv4 model in achieving high accuracy and real-time detection capabilities. It also positions this work as a pioneering effort in the field of underwater pipeline detection using deep learning.

Authors: Xuanyao Bai, Kailun Wen, Donghong Peng, Shuangqiang Liu,and Le Luo The objective of this study is to explore the measurement principles and detection methods of atomic magnetometers, and it also examines the technological means and research progress of atomic magnetometers in various industrial fields, including magnetic imaging, material examination, underwater magnetic target detection, and magnetic communication. Additionally, this study discusses the potential applications and future development trends of atomic magnetometers.

Click on the octopus to return to the top of the page

Author: Yaofeng Xie, Zhibin Yu, Xiao Yu, & Bing Zheng This article proposes a new underwater image enhancement network designed to address the challenges of enhancing images captured in low-light underwater environments. It explains the limitations of existing algorithms, the need for a specialized approach, and the creation of a new dataset to train the proposed model. It also highlights the effectiveness and robustness of the new method through experimental results, emphasizing its superior performance compared to previous methods.

Authors: Xuebo Zhang , Haixin Sun, & Arata Kaneko This paper highlights the significance of marine environment and hydroacoustic data collection for military, economic, and ecological purposes. It discusses the challenges environmental interference poses to underwater communication and target detection and emphasizes the need for advanced technical solutions to improve detection and identification capabilities. The paper also underscores the importance of sound wave technology and signal processing in ocean exploration.

49 - Mapping Underwater Aquatic Vegetation Using Foundation Models With Air- and Space-Borne Images: The Case of Polyphytos Lake

Authors: Leonidas Alagialoglou, Ioannis Manakos, Sofia Papadopoulou, Rizos-Theodoros Chadoulis, & Afroditi Kita This study explains the effectiveness of different artificial intelligence (AI) techniques for mapping underwater aquatic vegetation (UVeg) using remote sensing data. It evaluates the performance of classical AI tools and a modern AI tool (Segment Anything Model, SAM) in the context of few-shot learning with limited annotation. The results highlight the superior performance of the SAM model for airborne imagery but also note its challenges for spaceborne imagery, providing researchers and practitioners with insights into the strengths and limitations of different AI models for UVeg mapping.

Authors: Shaowei Zhang, & Zhiqun Daniel Deng This document overviews the advancements and research in deep-sea exploration and observation technology. It highlights the development of sophisticated sensors and equipment for studying deep-sea environments, such as seabed samplers and MEMS sensors. The text also discusses specific technologies and methods, such as atmospheric evaporation waveguide communication and dual-axis rotary modulation inertial navigation systems, that enhance ocean exploration capabilities. Additionally, it mentions a study on evaporation duct properties in the East China Sea, emphasizing the importance of understanding this phenomenon for electromagnetic wave propagation.

Authors: Hualong Du, Huijie Fan, Qifeng Zhang, and Shuo Li This text describes a method for accurately measuring the volume of thick oil slicks in real-time, which is crucial for oil spill response strategies and evaluating disposal efficiency. It explains the use of ultrasonic inspection and image processing techniques, facilitated by a remotely operated vehicle and an airborne drone, to measure the thickness and area of oil slicks. It also highlights the successful verification of this method through laboratory experiments, suggesting its potential application in real- world oil spill scenarios.

Authors: Ziyang Wang, Liquan Zhao, Tie Zhong, Yanfei Jia, & Ying Cui This paper describes a proposed solution for improving the quality of underwater images using a generative adversarial network (GAN) with multi-scale and attention mechanisms. It explains the problem of degraded underwater images and introduces various modules and techniques integrated into the GAN to enhance image quality. It also mentions the evaluation of the proposed method using specific datasets and highlights the superior performance of the enhanced images compared to other methods.

Authors: Nicholas F. L. Vale, Juan C. Braga, Alex C. Bastos, Fernando C. Moraes, Claudia S. Karez, Ricardo G. Bahia, Luis A. Leão, Renato C. Pereira, Gilberto M. Amado-Filho, and Leonardo T. Salgado This document presents a study on rhodolith beds located on the Brazilian continental shelf, specifically off the Sergipe-Alagoas Coast. It describes the characteristics, distribution, and composition of these rhodolith beds, as well as the methods used for their study, such as ROV operations and bottom trawl sampling. It also provides information on the age and growth phases of the rhodoliths, highlighting their ecological significance and the diversity of red algal morpho-taxa identified in the area.

Authors: Kai Yan, Lanyue Liang, Ziqiang Zheng, Guoqing Wang, Yang Yang This text highlights the challenges of underwater visual perception due to factors like low illumination and suspended particles and critiques existing enhancement methods for their poor performance. It proposes a new approach that achieves better results with a simple and lightweight network configuration, demonstrating its effectiveness through experimental results. It also provides a link to access the code for this method, indicating an intent to share the research and facilitate further exploration or use by others.

Authors: Larissa Macedo Cruz de Oliveira, Aaron Lim, Luis A. Conti, and Andrew J. Wheeler This document describes a study that explores the use of machine learning methods combined with Structure-from- Motion (SfM) photogrammetry for the classification of cold-water coral (CWC) reefs in deep-water environments. It outlines the challenges faced in this domain, such as the lack of training data and benchmark datasets, and details the methodology used in the study, including the selection of the Piddington Mound area for data collection, the machine learning algorithms employed, and the evaluation of their performance.

Authors: Yiyang Chen, Pengqiang Ge, Guina Wang, Guirong Weng, Hongtian Chen This document is an analysis of the active contour model (ACM) approach in image segmentation within the field of computer vision. It reviews different types of ACMs, discuss their theoretical foundations, and evaluate their performance through experiments. Additionally, it seeks to compare ACMs with deep learning-based algorithms and identify future research directions in this area.

27 - Industry remotely operated vehicle imagery for assessing marine communities associated with subsea oil and gas infrastructure on the continental shelf of South-East Australia

Authors: Daniel Ierodiaconou, Dianne McLean, Matthew Jon Birt, Todd Bond, Sam Wines, Ollie Glade-Wright, Joe Morris, Doug Higgs, and Sasha K. Whitmarsh This study examines the ecological value of offshore oil and gas infrastructure in marine environments, particularly as these structures approach the end of their operational life. It informs decommissioning decisions by analyzing the diversity and distribution of fish, invertebrate, and benthic communities using remotely operated vehicle (ROV) imagery, and also provides an overview of the methods used, the results obtained, and highlights the ecological significance of certain species observed in the study.

Authors: Jiancheng Yin, Xiangyu You, Yu Yao This text informs the reader about the challenges and methods involved in detecting subsea oil pipelines, specifically focusing on the use of sonar and acoustic scattering characteristics. It also aims to describe a study that establishes an acoustic scattering model using the finite element method to analyze various factors affecting the detection of buried pipelines.

Authors: Zefeng Zhao, Zhuang Zhou, Yunting Lai, Tenghui Wang, Shujie Zou, Haohao Cai, and Haijun Xie This text presents findings on the effectiveness of different technologies, specifically the multibeam echo sounder (MBES) WASSP S3 and the remotely operated underwater vehicle (ROV) Chasing M2, in detecting and mapping marine litter on the sea floor. It evaluates the capabilities and limitations of these technologies in identifying marine debris, particularly in the St. Ante Channel area in Croatia, and provides guidelines for their integrated use in marine litter detection.

Authors: Shuangshuang Fan, Xinyu Zhang, Guangxian Zeng, and Xiao Cheng This study aims to improve the efficiency and accuracy of mapping the underside of sea ice by proposing adaptive sampling and map reconstruction methods. It highlights the challenges of mapping ice topography with limited resources and introduces a graphics-based adaptive mapping method and a sparse approximation method for reconstructing ice draft maps. The intent is to contribute to the scientific understanding of sea-ice interactions and provide a cost-effective solution for under-ice observation missions.

Authors: Ahila Priyadharshini R, Ramajeyam K This study introduces a new methodology for enhancing the visual quality of underwater images. It outlines the challenges faced in underwater imaging, such as speckle noise, backscatter noise, and blur, and explains how the proposed approach, which includes techniques like color correction, contrast enhancement, homomorphic filtering, and fusion, addresses these issues. It also highlights the effectiveness of the methodology through qualitative and quantitative assessments, claiming it surpasses existing state-of-the-art techniques. The ultimate goal is to improve the quality of underwater images for fields like marine biology, oceanography, and underwater archaeology.