Vision-based underwater target real-time detection for autonomous underwater vehicle subsea exploration

被引:3
|
作者
Xu, Gaofei [1 ]
Zhou, Daoxian [2 ,3 ,4 ]
Yuan, Libiao [2 ,3 ,4 ]
Guo, Wei [1 ]
Huang, Zepeng [5 ]
Zhang, Yinlong [2 ,3 ,4 ]
机构
[1] Chinese Acad Sci, Inst Deep Sea Sci & Engn, Sanya, Peoples R China
[2] Chinese Acad Sci, Shenyang Inst Automat, State Key Lab Robot, Shenyang, Peoples R China
[3] Chinese Acad Sci, Key Lab Networked Control Syst, Shenyang, Peoples R China
[4] Chinese Acad Sci, Inst Robot & Intelligent Mfg, Shenyang, Peoples R China
[5] Natl Ctr Archaeol, Underwater Archaeol Dept, Beijing, Peoples R China
基金
中国国家自然科学基金;
关键词
autonomous underwater vehicle; subsea exploration; real-time target detection; lightweight convolutional neural network; underwater image enhancement;
D O I
10.3389/fmars.2023.1112310
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
Autonomous underwater vehicles (AUVs) equipped with online visual inspection systems can detect underwater targets during underwater operations, which is of great significance to subsea exploration. However, the undersea scene has some instinctive challenging problems, such as poor lighting conditions, sediment burial, and marine biofouling mimicry, which makes it difficult for traditional target detection algorithms to achieve online, reliable, and accurate detection of underwater targets. To solve the above issues, this paper proposes a real-time object detection algorithm for underwater targets based on a lightweight convolutional neural network model. To improve the imaging quality of underwater images, contrast limited adaptive histogram equalization with the fused multicolor space (FCLAHE) model is designed to enhance the image quality of underwater targets. Afterwards, a spindle-shaped backbone network is designed. The inverted residual block and group convolutions are used to extract depth features to ensure the target detection accuracy on one hand and to reduce the model parameter volume on the other hand under complex scenarios. Through extensive experiments, the precision, recall, and mAP of the proposed algorithm reached 91.2%, 90.1%, and 88.3%, respectively. It is also noticeable that the proposed method has been integrated into the embedded GPU platform and deployed in the AUV system in the practical scenarios. The average computational time is 0.053s, which satisfies the requirements of real-time object detection.
引用
收藏
页数:12
相关论文
共 50 条
  • [21] Monocular Vision-Based Real-Time Vehicle Detection at Container Terminals
    Liu, Zijian
    Zhang, Tianlei
    He, Bei
    Liu, Yu
    Sun, Li
    Tang, Wenyang
    PROCEEDINGS OF CHINA SAE CONGRESS 2018: SELECTED PAPERS, 2020, 574 : 821 - 830
  • [22] Real-time path planning and obstacle avoidance for an autonomous underwater vehicle
    Universita degli Studi di Napoli, 'Federico II', Napoli, Italy
    Proc IEEE Int Conf Rob Autom, (78-83):
  • [23] Vision-based Navigation Solution for Autonomous Underwater Vehicles
    Alves, Tiago
    Hormigo, Tiago
    Ventura, Rodrigo
    2022 IEEE INTERNATIONAL CONFERENCE ON AUTONOMOUS ROBOT SYSTEMS AND COMPETITIONS (ICARSC), 2022, : 226 - 231
  • [24] A real-time control architecture for a semi-autonomous underwater vehicle
    Li, JH
    Jun, BH
    Lee, PM
    Honga, SW
    OCEANS '04 MTS/IEEE TECHNO-OCEAN '04, VOLS 1- 2, CONFERENCE PROCEEDINGS, VOLS. 1-4, 2004, : 271 - 275
  • [25] Real-time Path Planning for Autonomous Underwater Vehicle Mobile Docking
    Zeng, Daheng
    Chen, Shumin
    Li, Zengni
    Ma, Xinqi
    Xu, Yuanxin
    GLOBAL OCEANS 2020: SINGAPORE - U.S. GULF COAST, 2020,
  • [26] Real-time path planning and obstacle avoidance for an autonomous underwater vehicle
    Antonelli, G
    Chiaverini, S
    Finotello, R
    Morgavi, E
    ICRA '99: IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION, VOLS 1-4, PROCEEDINGS, 1999, : 78 - 83
  • [27] Monocular Vision-Based Underwater Object Detection
    Chen, Zhe
    Zhang, Zhen
    Dai, Fengzhao
    Bu, Yang
    Wang, Huibin
    SENSORS, 2017, 17 (08)
  • [28] Real-time underwater target detection based on improved YOLOv7
    Wu, Qingqi
    Cen, Lihui
    Kan, Shichao
    Zhai, Yongping
    Chen, Xiaofang
    Zhang, Hong
    JOURNAL OF REAL-TIME IMAGE PROCESSING, 2025, 22 (01)
  • [29] Autonomous Vision-based Target Detection Using Unmanned Aerial Vehicle
    Deeds, Jeff
    Engstrom, Zach
    Gill, Caleb
    Wood, Zack
    Wang, Jing
    Ahn, In Soo
    Lu, Yufeng
    2018 IEEE 61ST INTERNATIONAL MIDWEST SYMPOSIUM ON CIRCUITS AND SYSTEMS (MWSCAS), 2018, : 1078 - 1081
  • [30] Integrated real-time vision-based preceding vehicle detection in urban roads
    Chong, Yanwen
    Chen, Wu
    Li, Zhilin
    Lam, William H. K.
    Zheng, Chunhou
    Li, Qingquan
    NEUROCOMPUTING, 2013, 116 : 144 - 149