Article
Engineering, Electrical & Electronic
Li Hu, Xiaodong Wang, Shilian Wang
Summary: This article investigates the detection and localization of underwater targets using a sensor network with acoustic sources, multiple sensors, and a fusion center. It proposes centralized and decentralized GLRT detection schemes, utilizing quantization of sensor observations and optimizing thresholds for accurate detection performance. The numerical results demonstrate that the decentralized GLRT detector can achieve similar performance to the centralized GLRT with only a small number of quantization bits.
IEEE SENSORS JOURNAL
(2021)
Article
Engineering, Marine
Xiao Chen, Mujiahui Yuan, Qi Yang, Haiyang Yao, Haiyan Wang
Summary: In this study, we propose the Underwater-YCC optimization algorithm based on YOLO v7 to enhance the accuracy of detecting small targets underwater. Our algorithm utilizes CBAM to obtain fine-grained semantic information and Conv2Former to handle underwater blurred images. By applying Wise-IoU, we improve detection accuracy by assigning multiple weights between high- and low-quality images. Experimental results on the URPC2020 dataset demonstrate that the proposed algorithm achieves a mean Average Precision (mAP) of up to 87.16% in complex underwater environments.
JOURNAL OF MARINE SCIENCE AND ENGINEERING
(2023)
Article
Environmental Sciences
Hui Huang, Zehao Sun, Shuchang Liu, Yanan Di, Jinzhong Xu, Caicai Liu, Ren Xu, Hong Song, Shuyue Zhan, Jiaping Wu
Summary: Underwater hyperspectral imaging was proposed for in situ detection of microplastics (MPs) on the seabed in turbid water, assisted by spectral image correction. Support vector machine (SVM) outperformed other classifiers, achieving high precision (PR) = 0.9839, recall (RE) = 0.9859, and F1-score (F1) = 0.9849 for identifying six types of MPs.
SCIENCE OF THE TOTAL ENVIRONMENT
(2021)
Article
Engineering, Multidisciplinary
Jing Yan, Zeqian Zhang, Xian Yang, Xiaoyuan Luo, Xinping Guan
Summary: This article focuses on the issue of non-cooperative target detection in underwater sensor networks by fusing active and passive measurements. The authors propose a Chi-square test based local decision rule and a Bayesian fusion algorithm to integrate detection decisions from different sensor nodes, improving accuracy and reducing communication load.
IEEE TRANSACTIONS ON NETWORK SCIENCE AND ENGINEERING
(2023)
Article
Engineering, Marine
Kaiyue Liu, Qi Sun, Daming Sun, Lin Peng, Mengduo Yang, Nizhuan Wang
Summary: This study proposes an improved YOLOv7 network (YOLOv7-AC) for underwater target detection, which utilizes ACmixBlock module to improve feature extraction and network reasoning speed, ResNet-ACmix module to avoid feature information loss and reduce computation, and Global Attention Mechanism (GAM) to enhance feature extraction. Experimental results show that the improved YOLOv7 network outperforms the original YOLOv7 model and achieves high mAP values on different datasets. The proposed network holds great potential for practical applications in various underwater tasks.
JOURNAL OF MARINE SCIENCE AND ENGINEERING
(2023)
Article
Automation & Control Systems
Lizhao Liu, Pinrui Li
Summary: With the rapid development in the field of marine exploration, underwater robots have become increasingly important in various areas such as military, fisheries, aquaculture, and resource protection. However, existing target detection algorithms often fail to perform well in underwater environments. This paper proposes the PILLO algorithm, based on plant intelligence and the YOLOv5 algorithm, to address the challenges of underwater target detection.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2023)
Article
Computer Science, Artificial Intelligence
Wen Wu, Xiao-Diao Chen, Wenya Yang, Jun-Hai Yong
Summary: This research proposes a robust shadow detection framework that can locate and correct noisy labels to improve detection performance. Experimental results show that the method outperforms state-of-the-art approaches on public benchmarks.
KNOWLEDGE-BASED SYSTEMS
(2023)
Article
Engineering, Marine
Fei Lei, Feifei Tang, Shuhan Li
Summary: This paper applies the advanced YOLOv5 algorithm to underwater target detection and improves it by combining with methods suitable for the underwater environment. The improved network model achieves impressive results in detecting underwater targets.
JOURNAL OF MARINE SCIENCE AND ENGINEERING
(2022)
Article
Optics
Zhan-Ming Li, Heng Zhou, Zhong-Yuan Li, Zeng-Quan Yan, Cheng-Qiu Hu, Jun Gao, Shi-Bao Wu, Si-Yuan Yin, Xian-Min Jin
Summary: This study demonstrates a thresholded single-photon imaging and detection scheme to extract photon signals from noisy underwater environments, achieving underwater detection down to 0.8 photons per pulse in high-loss underwater conditions. Results break the limits of classical underwater imaging and detection and open up possibilities for quantum-enhanced applications in deep-sea exploration.
Article
Environmental Sciences
Gaofei Xu, Daoxian Zhou, Libiao Yuan, Wei Guo, Zepeng Huang, Yinlong Zhang
Summary: This paper proposes a real-time object detection algorithm based on a lightweight convolutional neural network model for underwater targets. It addresses the challenges in underwater scenes and achieves high accuracy in target detection. The proposed method has been successfully deployed in practical scenarios and meets the requirements of real-time object detection.
FRONTIERS IN MARINE SCIENCE
(2023)
Article
Environmental Sciences
Natasha Y. Flores, Stephanie B. Oswald, Rob S. E. W. Leuven, Frank P. L. Collas
Summary: Plastic pollution in the ocean mainly comes from rivers. Monitoring floating plastics in rivers is time consuming, so exploring remote sensing options is necessary. Two imaging sonar technologies, ARIS and SSS, were tested in standing water and found to detect 100% of macroplastics. ARIS provided higher resolution images but had a lower detection rate in flowing water. SSS, a low-cost option, integrated with CHIRP sonar technology, provided 3D position of targets.
FRONTIERS IN ENVIRONMENTAL SCIENCE
(2022)
Article
Geochemistry & Geophysics
Gang Xiong, Fang Wang, Wenxian Yu, Trieu-Kien Truong
Summary: This article presents a novel image processing method SSMRI aimed at obtaining robust image features for SAR target detection at very low signal-to-noise ratio (SNR). By transforming SAR images based on singularity exponent and analyzing separability of SAR target in the space of singularity exponent, the stability and robustness of SSMRI-SAR feature extraction are demonstrated. Experiments show that the proposed SSMRI method outperforms traditional CFAR or 2-D-SPS methods, achieving a detection performance of over 99.4% even at SNR = -30 dB.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Physics, Multidisciplinary
Song Qiang, Sun Xiao-Bing, Liu Xiao, Ti Ru-Fang, Huang Hong-Lian, Wang Hao
Summary: Underwater optical imaging is important for seabed exploration and target recognition, but is often limited by bubbles' light scattering effects. This study investigates the polarization information of targets under different light incident angles in underwater environments and proposes using polarimetric image fusion methods to suppress bubble interference, improving target clarity. Future research may explore energy loss compensation or machine learning for target recognition in high-density bubble environments.
ACTA PHYSICA SINICA
(2021)
Article
Engineering, Marine
Ruoyu Chen, Ying Chen
Summary: This paper proposes a real-time sonar-based target detection system mounted on an autonomous underwater helicopter (AUH) using an improved convolutional neural network (CNN) to detect underwater targets quickly and precisely. YOLOv5 is used as the basic CNN network due to its strength, lightweight, and fast speed. Attention mechanisms are added and the structure of YOLOv5 is optimized to improve the performance of the detector, achieving a 1-3% increment in mAP up to 80.2% with an average speed of 0.025 s (40 FPS) on the embedded device. The system has been successfully verified in both a school tank and outdoor open water, meeting the requirements of real-time and lightweight detection using limited hardware.
JOURNAL OF MARINE SCIENCE AND ENGINEERING
(2023)
Article
Engineering, Electrical & Electronic
Yongqiang Cheng, Hao Wu, Zheng Yang, Hongqiang Wang
Summary: In this letter, a scheme for underwater target detection is proposed by measuring the water-surface vibrations caused by acoustic waves with a millimeter-wave radar. The power of the water-surface vibration due to the reflection of acoustic waves by the target is derived, giving a quantitative relationship between the perturbations on water-surface vibration and the position of the target. A signal model is established to accurately measure the water-surface vibration with millimeter-wave radar, and a method to estimate the power spectrum of the phase of water-surface vibrations is proposed. The effectiveness of the scheme is validated through water-tank experiments.
IEEE ANTENNAS AND WIRELESS PROPAGATION LETTERS
(2023)
Article
Optics
Aymeric Le Gratiet, Sylvain Rivet, Matthieu Dubreuil, Yann Le Grand
Article
Biochemical Research Methods
David Sevrain, Matthieu Dubreuil, Grace Elizabeth Dolman, Abed Zaitoun, William Irving, Indra Neil Guha, Christophe Odin, Yann Le Grand
BIOMEDICAL OPTICS EXPRESS
(2015)
Article
Cardiac & Cardiovascular Systems
Florine Tissier, Yassine Mallem, Christelle Goanvec, Romain Didier, Thierry Aubry, Nathalie Bourgeois, Jean-Claude Desfontis, Matthieu Dubreuil, Yann Le Grand, Jacques Mansourati, Karine Pichavant-Rafini, Emmanuelle Plee-Gautier, Philippe Roquefort, Michael Theron, Martine Gilard
Article
Optics
Aymeric Le Gratiet, Matthieu Dubreuil, Sylvain Rivet, Yann Le Grand
Article
Optics
Matthieu Dubreuil, Sylvain Rivet, Yann Le Grand
Article
Biochemical Research Methods
Matthieu Dubreuil, Florine Tissier, Lucas Le Roy, Jean-Pierre Pennec, Sylvain Rivet, Marie-Agnes Giroux-Metges, Yann Le Grand
BIOMEDICAL OPTICS EXPRESS
(2018)
Article
Optics
Sylvain Rivet, Matthieu Dubreuil, Yann Le Grand
JOURNAL OF THE OPTICAL SOCIETY OF AMERICA A-OPTICS IMAGE SCIENCE AND VISION
(2019)
Article
Biochemical Research Methods
Matthieu Dubreuil, Florine Tissier, Sylvain Rivet, Yann Le Grand
Summary: Linear diattenuation (LD) imaging in complex biological specimens has received little attention due to its challenges. Near infrared transmission Mueller scanning microscopy offers unprecedented resolution and sensitivity for imaging LD, highlighting anisotropic scattering by reducing the contribution of dichroism. Pixel-by-pixel comparison of LD images with LR and multiphoton images reveals under-resolved structures that are not visible by other means, particularly within the sarcomere of skeletal muscles.
BIOMEDICAL OPTICS EXPRESS
(2021)
Article
Biochemical Research Methods
George Alzeeb, Matthieu Dubreuil, Danielle Arzur, Sylvain Rivet, Laurent Corcos, Yann Le Grand, Catherine Le Jossic-Corcos
Summary: Gastric cancer is a highly deadly disease, and three-dimensional cancer cell cultures can better mimic the tumor microenvironment. Cancer-associated fibroblasts in the microenvironment play a role in promoting or restraining cancer cell proliferation, invasion, and drug resistance. The analysis of spheroid organization from different GC cell lines mixed with fibroblasts showed differences, suggesting the highly personalized nature of gastric cancer.
BIOMEDICAL OPTICS EXPRESS
(2022)
Article
Optics
Xavier Theillier, Sylvain Rivet, Matthieu Dubreuil, Yann Le Grand
Summary: This paper presents a novel configuration for null polarimetry, based on the concept of spectrally encoded light polarization. The method achieves high-speed measurements of linear or circular retardance with exceptional sensitivity. The compact and rapid nature of the method makes it potentially useful for laser scanning microscopes.
Article
Physics, Multidisciplinary
David Spenato, Matthieu Dubreuil, Denis Morineau, Philippe Giamarchi, David Dekadkevi, Jean-Philippe Jay, Alain Fessant, Sylvain Rivet, Yann Le Grand
Summary: In this study, Ta/NiO bilayers were shown to be high-efficiency, lithography free, reflective structural color filters for generating broad color gamut. Experimental results demonstrated adjustable strong structural colors, and the optical interference effect remained effective even on flexible substrates, enabling flexible color filtering applications.
JOURNAL OF PHYSICS COMMUNICATIONS
(2022)
Proceedings Paper
Optics
Sylvain Rivet, Matthieu Dubreuil, Adrian Bradu, Yann Le Grand
ADVANCES IN MICROSCOPIC IMAGING II
(2019)
Article
Chemistry, Multidisciplinary
Kevin Raimondeau, Sylvain Rivet, Matthieu Dubreuil, Yann Le Grand, Laurent Dupont, Jean-Louis de Bougrenet de la Tocnaye
MOLECULAR CRYSTALS AND LIQUID CRYSTALS
(2019)
Article
Multidisciplinary Sciences
Sylvain Rivet, Matthieu Dubreuil, Adrian Bradu, Yann Le Grand
SCIENTIFIC REPORTS
(2019)
Proceedings Paper
Biophysics
A. Le Gratiet, M. Dubreuil, S. Rivet, Y. Le Grand
BIOPHOTONICS: PHOTONIC SOLUTIONS FOR BETTER HEALTH CARE V
(2016)
